Full text of "The prevention of tuberculosis"

See other formats

THE NEW LIBRARY OF MEDICINE

EDITED BY C. W. SALEEBY, M.D.. F.R.S.E.

THE
PREVENTION OF TUBERCULOSIS

THE PREVENTION
OF TUBERCULOSIS

] ARTHUR NEWSHOLME,

M.D.',"F.R.C.P.

PRESIDENT OF THE EPIDEMIOLOGICAL SECTION OF THE ROYAL SOCIETY OF MEDICINE
LATE MEDICAL OFFICER OF HEALTH OF BRIGHTON

WITH THIRTY-NINE DIAGRAMS

SECOND EDITION

METHUEN & CO.

36 ESSEX STREET W.C.

LONDON

First Published . . August 2Oth, fqo8
Second Edition . . rgro

PREFACE

THE promise to write this book as one of a series dealing
with the public aspects of Medicine was made in 1906.
The greater part of it was written over a year ago,
Part I. almost entirely so, the quotations from the Second
Interim Report of the Royal Commission appointed to inquire .
into the Relations of Human and Animal Tuberculosis, with
such slight modifications of inference as were necessitated by
it, being added subsequently.

Part II. is a restatement of an investigation of which the
results were last set forth in the Journal of Hygiene for July 1906.
Although necessarily lengthy and full of detail, the argument
and conclusion that institutional segregation is the predominant 1V'
cause of the decline of phthisis in this country has great import-
ance in its bearing on the administrative measures considered
in Part III. It is perhaps unfortunate that the argument is
continuous from end to end, and that its effect is misconceived
when only parts of it are considered disjoined from the remainder.
In the absence of this continuity the investigation could have
yielded no more than ground for surmise or conjecture. The
history of the public health service gives familiar proof of the
important place taken by scientific hypothesis among the tools
at our disposal. When, however, conclusions can be tested
by actual experience, such experience obviously affords a surer
basis for administrative action ; and in a matter of such
immense importance to public health as the control of tuber-
culosis, the intricacies of a statistical inquiry embodying historical
and international experience are worth undertaking and master-
ing if, as happens often and is certainly so in the present case,
the question cannot be discussed conclusively without it.

The chapters on statistics are indispensable to the main
arguments of the book, especially to those in Part II. If, for
instance, statistics of phthisis are largely vitiated by trans-

vi THE PREVENTION OF TUBERCULOSIS

ference between this disease and bronchitis, important reasoning
as to the course of phthisis can scarcely be based on them.
Where not otherwise stated, the English statistics are derived
from the Reports of the Registrar-General of Births and Deaths
and from Dr. Tatham's letters therein ; some of these tables
have been calculated separately, or readjusted for my special
purposes.

The bibliography on p. 415 does not pretend to be complete.
It comprises only the papers and books actually quoted in this
volume. It is hoped that the index of names of places and
persons will form a useful supplement to the subject-index.

This volume is written almost solely from the standpoint
of the public health administrator, and is intended primarily
for medical officers of health. It is believed, however, that it
will also be interesting and useful to all medical practitioners,
to many members of Sanitary Authorities and Hospital Com-
mittees, to patients themselves, and to that increasing proportion
of the public who desire to know more of preventive medicine.
As therapeutics in the more limited sense of the word has been
entirely excluded from its scope, there appears to be no impedi-
ment, except, perhaps, lack of interest, to this wider utility of
the discussion of tiiberculosis here attempted.

I have to thank my friend H. C. Lecky, M.A., M.B., and H. P.
Newsholme B.A., B.Sc., for reading portions of the manuscript,
and for valuable suggestions, and the latter for seeing the
volume through the press.

A. N.

February 4t/i, 1908.

LIST OF FIGURES

FIG.

1 . Comparative Magnitude of some of the Chief Preventable Causes

of Death in England and Wales . ... 5

2. England and Wales, 1904. — Male and Female Death-rates from

Phthisis at different Age-periods . . . • 9

3. Deaths from Phthisis at each Age-period per 100 Deaths from the

same Disease at all Ages . . . . .10

4. Deaths from Phthisis at each Age-period per 100 Deaths from all

Causes at the same Age-period . . . .11

5. Relative Death-rates from (a) Phthisis, (ft) Bronchitis and Pneu-

monia in England and Wales, the rates for 1901-04=100 . 25

6. Comparison between 1861-70 and 1901 of relative Death-rates at

different Age-periods from Bronchitis plus Pneumonia and
from Phthisis . . . . . . .28

7. Relative Death-rates from different Tuberculous Diseases from

1850-54 to 1901-04, the Death-rate in the most recent period in
each instance being stated as 100 . . • -34

8. Section of a Lung chiefly in the first stage of Phthisis . . 46

9. Section of a Lung chiefly in the second stage of Phthisis . . 46

10. Section of a Lung chiefly in the third stage of Phthisis . . 46

1 1 . Section of a Lung exhibiting Cretaceous Masses . . .46

12. Acinus of the Lung, enlarged ten times . . . in

13. Death-rates from Phthisis for Males and Females at different Age-

periods in England and Wales, Sheffield, and Birmingham . 166

14. Female Death-rate from Phthisis at each Age-period in 1861-70,

and in 1891-1900, that of Males at the same Age-period = 100 . 170

15. Male Death-rate from Phthisis at each Age-period in Urban and

Rural Counties . . . . • • • J73

1 6. Female Death-rate from Phthisis at each Age-period in Urban

and Rural Counties . . • • J74

17. Death-rate from Phthisis in Males and Females at each Age-period

in Urban Counties .

1 8. Death-rate from Phthisis in Males and Females at each Age-

period in Rural Counties . . • • .170

19. Showing steady improvement in Housing Conditions in Ireland . 227

20. Proportional Phthisis Death-rates and Wheat Prices in the U.K. 233

21. Proportional Phthisis Death-rates and Wheat Prices in Paris . 234

22. Proportional Tuberculosis Death-rates and Wheat Prices in

Prussia. ....

vii

viii THE PREVENTION OF TUBERCULOSIS

23. Proportional Phthisis Death-rates and Wheat Prices in Mass., -^

U.S.A. ........ 235

24. Proportional Phthisis Death-rates and Cost of Food in the U. K. . 237

25. Proportional Tuberculosis Death-rates and Cost of Food in Prussia 237

26. Proportional Phthisis Death-rates and Cost of Living in England 239

27. Relative Changes in Pauperism and Phthisis Death-rate in

England ... . . . . . 245

28. Relative Changes in Pauperism and Phthisis Death-rate in

London ....... 246

29. Relative Changes in Pauperism and Phthisis Death-rate in

Scotland ....... 247

30. Relative Changes in Pauperism and Phthisis Death-rate in

Ireland . ...... 248

3 1 . Comparison of the Changes in the Death-rates from Typhus and

from Phthisis in Ireland and in England and Wales . 260, 261

32. Number of Total Lepers and of Lepers in Asylums in Norway . 264

33. Rates of Changes in Phthisis Death-rates and in the ratio of

Institutional to Total Deaths in England . . .271

34. Rates of Changes in Phthisis Death-rates and in the ratio of

Institutional to Total Deaths in London . . . 272

35. Rates of Changes in Phthisis Death-rates and in the ratio of

Total to Indoor Pauperism in England .... 278

36. Rates of Changes in Phthisis Death-rates and in the ratio of

Total to Indoor Pauperism in Scotland . . . 279

37. Brighton. Annual Notifications, Sputa examined, and Admissions

to the Sanatorium ...... 341

38. Comparison of Death-rate from Phthisis and other forms of

Tuberculosis at different Age-periods . . . . 361

39. Block Plan of Isolation Hospital . . . . -397

CONTENTS

PART I
CAUSATION OF TUBERCULOSIS

CHAP. PAGE

I. MAGNITUDE OF THE EVIL : A. MORTALITY . . 3

II. Do. B. SICKNESS AND ECONOMICS 13

III. ARE THE STATISTICS RELATING TO TUBERCULOSIS TRUST-

WORTHY ? ...... 22

IV. THE HISTORY OF PHTHISIS . . . « 35
V. THE MORBID ANATOMY AND SYMPTOMS OF PHTHISIS . 43

VI. THE TUBERCLE BACILLUS . . . . -51

VII. INFECTIVITY OF TUBERCULOSIS : A HISTORY OF VIEWS

HELD . . . . . . «55

VIII. INFECTIVITY OF TUBERCULOSIS: B. EXPERIMENTAL

EVIDENCE . . . . . -59

IX. INFECTIVITY OF TUBERCULOSIS: C. STATISTICAL AND

CLINICAL EVIDENCE . . . . .62

X. LATENCY IN TUBERCULOSIS . . . -74

XI. SOURCES OF INFECTION: MINOR SOURCES . . 86

XII. Do. : DUST AND SPRAY . . 89

XIII. CIRCUMSTANCES LIMITING THE AMOUNT OF INFECTION BY

DUST AND SPRAY . . . . . 101

XIV. THE PORTALS OF INFECTION : A. INFECTION BY INHALA-

TION . . . . . . . 106

XV. THE PORTALS OF INFECTION : B. INFECTION BY INGESTION 115

XVI. RELATION OF BOVINE AND HUMAN TUBERCULOSIS . 121
XVII. EVIDENCE OF THE OCCURRENCE OF BOVINE TUBERCULOSIS

IN MAN ...... I31

XVIII. TUBERCULOSIS FROM MEAT AND FROM MILK AND OTHER

DAIRY PRODUCTS . . • • • *39
XIX. DOMESTIC INFECTION . . . • .146
XX. INFECTION IN ATTENDANCE ON THE SICK . 152
XXI. INDUSTRIAL INFECTION . . . • 157
XXII. PERSONAL INFLUENCES OTHER THAN INFECTION FAVOUR-
ING TUBERCULOSIS : SUSCEPTIBILITY TO INFECTION . 161

XXIII. AGE AND SEX ... - 164

XXIV. PERSONAL CONDITIONS LOWERING RESISTANCE TO IN-

FECTION
XXV. HEREDITARY DISPOSITION TO PHTHISIS . . .182

x THE PREVENTION OF TUBERCULOSIS

CHAP. PAGE

XXVI. CONDITIONS OF ENVIRONMENT LOWERING RESISTANCE
TO INFECTION, SOCIAL MISERY, AND INSANITARY
CIRCUMSTANCES . . . . .191

XXVII. CLIMATE AND SOIL . . . . 194

PART II

THE MEANS BY WHICH THE REDUCTION OF
MORTALITY FROM TUBERCULOSIS ALREADY
OBTAINED HAS BEEN SECURED

XXVIII. INTRODUCTORY . . * . . . 205

XXIX. TUBERCULOSIS AND GENERAL HEALTH IN VARIOUS
COMMUNITIES : VIRULENCE, NATURAL SELECTION, AND
DECADENCE . . . . .210

XXX. TUBERCULOSIS IN URBAN AND IN RURAL COMMUNITIES . 220
XXXI. TUBERCULOSIS IN OVERCROWDED COMMUNITIES . 224

XXXII. TUBERCULOSIS IN COMMUNITIES OF VARYING WELL-
BEING ... . . . 230

XXXIII. TUBERCULOSIS IN COMMUNITIES OF VARYING SANITARY

EDUCATION AND SANATORIUM PROVISION . .252

XXXIV. THE GENERAL RELATIONS OF TUBERCULOSIS AND OTHER

CHRONIC INFECTIOUS DISEASES TO INSTITUTIONAL
SEGREGATION . . . . . . 256

XXXV. TUBERCULOSIS IN COMMUNITIES WITH VARYING

AMOUNTS OF INSTITUTIONAL SEGREGATION . . 266

XXXVI. THE RELATIVE INFLUENCE OF INSTITUTIONAL SEGREGA-
TION AND OF OTHER MEASURES FOR THE CONTROL

OF TUBERCULOSIS ..... 292

PART III

MEASURES FOR THE REDUCTION AND ANNIHILA-
TION OF TUBERCULOSIS

XXXVII. GENERAL SCHEME OF PREVENTIVE MEASURES . . 301

XXXVIII. THE EARLY RECOGNITION OF PHTHISIS IN RELATION

TO PREVENTION ..... 306
XXXIX. THE MEDICAL PRACTITIONER IN RELATION TO PREVEN-
TIVE MEASURES . . . . .316
XL. THE CONSUMPTIVE PATIENT IN RELATION TO PREVEN-
TIVE MEASURES . . . . -324
XLI. THE PREVENTION OF INDISCRIMINATE EXPECTORATION 331
XLII. THE NOTIFICATION OF PHTHISIS . . -338
XLIII. THE SANITARY AUTHORITY IN RELATION TO PREVENTIVE

MEASURES AGAINST TUBERCULOSIS . . . 351

XLIV. EDUCATION AUTHORITIES AND THE PREVENTION OF

TUBERCULOSIS . . . . • 359

CONTENTS xi

CHAP. PAC.K

XLV. THE BOARD OF GUARDIANS AND THE PREVENTION

OF TUBERCULOSIS ..... 366

XLVI. INSURANCE AND FRIENDLY SOCIETIES IN RELATION

TO THE PREVENTION OF TUBERCULOSIS . -372

XLVII. DISPENSARIES AND THE PREVENTION OF TUBER-
CULOSIS ...... 377

XLVIII. THE R&LE OF SANATORIA IN THE TREATMENT AND

PREVENTION OF PHTHISIS . . 382

XLIX. THE INSTITUTIONAL TREATMENT OF PHTHISIS FROM THE

PUBLIC HEALTH STANDPOINT . . . 394

L. THE PREVENTION OF TUBERCULOSIS ARISING FROM

FOOD ....... 403

LI. THE CO-ORDINATION OF ADMINISTRATIVE MEASURES

AGAINST TUBERCULOSIS . .411

BIBLIOGRAPHY . . .415

INDEX OF NAMES OF PLACES . . 423

INDEX OF NAMES OF PERSONS . . 424

INDEX OF SUBJECTS . 427

PART I
CAUSATION OF TUBERCULOSIS

TERMS EMPLOYED

TUBERCULOSIS: the general name given to the disease result-
ing from the invasion of any part of the body by the tubercle
bacillus.

General Tuberculosis .
Acute Miliary Tuberculosis
Acute Tuberculosis
Phthisis

Pulmonary Phthisis
Pulmonary Tuberculosis
Consumption

Tabes Mesenterica
Tuberculous Peritonitis

Tuberculous Meningitis
Acute Hydrocephalus .
Brain Fever (in part) .
Lupus ....
Caries ....
Scrofula

\ Names given to tuberculosis where
many parts of the body are
J attacked simultaneously.

Tuberculosis of the lungs.

^Tuberculosis ot the peritoneum
j- and of the abdominal lym-
J phatic glands.

[Tuberculosis of the membranes
f surrounding the brain.

Tuberculosis of the skin,
bone.

lymphatic
glands.

Consumption, Tabes (both \ of Latin origin), and Phthisis (of
Greek origin) are all words the literal meaning of which is
"wasting."

The term Phthisis has been used sometimes in a sense wider
than that of Tuberculous Phthisis or Pulmonary Tuberculosis,
e.g. miners' phthisis, knife-grinders' phthisis, etc. In most, if
not in all such diseases, tuberculosis forms an important, though
possibly super added, cause of death. Possibility of error from
this cause will only affect the statistics of special localities.

CHAPTER I

MAGNITUDE OF THE EVIL: A. MORTALITY

TUBERCULOSIS is a disease caused by the destructive
lesions set up in the lungs or in other parts of the body
by a special bacillus or microbe. The disease is infectious,
i.e. is communicable from man to man and from animals to
man ; and it never originates in the body apart from the invasion
of the special bacillus.

Being an infective disease, tuberculosis comes into the same
category as the infectious diseases enumerated in Tables I. and
III. Large sums of money very properly are spent each year
in the prevention of these diseases ; hitherto but little has of
set purpose been spent on measures for the prevention of tuber-
culosis. We may, therefore, with advantage consider, in the
first place, the relative magnitude of these different causes of
death. In Table I. are set out the deaths from the acute infectious
diseases and from tuberculosis.

TABLE I.1 — ENGLAND AND WALES, 1904
Number of Deaths from —

Measles and German Measles
Whooping - Cough
Diarrhoea and Dysentery
Enteric Fever .
Diphtheria
Scarlet Fever .
Typhus Fever .
Small-pox

Pulmonary Phthisis .

All other Tuberculous Diseases

12,341
11,909
29,674
3,i53
5,763
3,770
37
5Q7

67,154
. 41,851

60,205

Thus tuberculous diseases in 1904 caused 60 deaths for every

1 All the statistical material relating to England and Wales contained in
this volume is derived from Dr. Tatham's valuable annual letters to the Registrar-
General of Births and Deaths, unless otherwise stated.

THE PREVENTION OF TUBERCULOSIS

67 caused by the aggregate of the chief acute infectious diseases.
These figures do not bring out fully the relative importance
and seriousness of deaths from tuberculosis. Although infantile
deaths are regrettable, they do not cause so great a loss to the
community and so much distress and suffering to the survivors
in a bereaved family as do deaths in early and middle life. The
following table is important in this connection : —

TABLE II. — ENGLAND AND WALES, 1904

Out of every 100 Deaths at all Ages the number occurring at different
Ages from each Cause of Death was —

Under 10.

10-20.

20-45.

45-65-

65 and over.

Measles ....

99-1

0'5

0'3

O'l

Whooping-Cough .

99 '9

O'l

Diarrhoea

93 '5

O'2

1-6

4-0

Phthisis ....

4-8

IO'I

56-5

25-3

3'3

Thus 99 out of every 100 total deaths from measles and whooping-
cough, and 94 out of every 100 deaths from diarrhoea, occurred
under 10 years of age, while only 5 out of 100 deaths from
pulmonary tuberculosis occurred under this age ; and during
the working years of life (20-65) 82 occurred out of every 100
total deaths from phthisis, as against no deaths from whooping-
cough, less than a half per cent, of the total deaths from measles,
and less than 2j per cent, of the total deaths from diarrhoea.

If we compare the mortality from tuberculosis with that
from infective diseases, other than those enumerated in Table I.,
we have the following result : —

TABLE III. — ENGLAND AND WALES, 1904

Number of Deaths from —
Influenza ..... 5,694 (the highest number in any one year

was 13,756, in 1890).
1,654
1,206
1,871 (doubtless understated).

Puerperal Fever

Erysipelas .

Syphilis and other Venereal Diseases

Tetanus (Lock-jaw)

Malaria

Anthrax

Glanders .

Hydrophobia

All forms of Tuberculosis .

1 06

o (in 1885 the number was 60 ; it has
not been so high since).

10,812
. 60,205

MAGNITUDE OF THE EVIL 5

Evidently none of these diseases occupies so important a place
as tuberculosis, though in the public administration of the

FIG. i —Comparative Magnitude of some of the Chief Preventable Causes of
Death in England and Wales

THE PREVENTION OF TUBERCULOSIS

country much larger sums are spent in the control of hydro-
phobia, glanders, anthrax, and puerperal fever than have hitherto
been spent in direct measures against tuberculosis.

In the Registrar-General's returns for England and Wales other
diseases than those enumerated above are classified as infective,
i.e. produced by infection received from without. Omitting pneu-
monia for separate consideration, the number of deaths returned
as due to infective diseases in 1904 was 140,431, the total number
of deaths from all causes in the same year being 547,784. Of
the total (140,431), 60,205 were caused by tuberculosis, 77,966
by the other infective diseases named in Tables I. and III.
Rheumatic fever, which is undoubtedly infective, though not
classified as such in the official returns, caused 1788 deaths in
1904. Probably most, if not all the diseases of the respiratory
organs have an infective origin, and many not recognised as
such are tuberculous. Pneumonia in 1904 caused 43,372 deaths,
bronchitis 42,188, all other diseases of the respiratory organs
excepting pulmonary tuberculosis 8059 deaths. The relative
magnitude of the most important preventable causes of death
is shown in Fig. i. The list is not complete, but the most
important items are included. Pneumonia and bronchitis have
been added, although only partially preventable under present
conditions. Cancer has also been added, because, although
not directly preventable, many of the deaths from it are pre-

TABLE IV. — ENGLAND AND WALES, 1904
Death-rate from Phthisis per 100,000 living at each Age-group

Ages.

Males.

Females.

Persons of
Both Sexes.

IO-

15- .

1 02

20-

161

12$

142

25- .

213

I58

184

35- •

270

170

218

45- •

310

I48

226

55- .

255

117

182

65- •

126

All Ages .

146

103

124

MAGNITUDE OF THE EVIL 7

ventable by early recognition and removal of the diseased
parts.

MORTALITY IN TERMS OF THE POPULAI ON— DEATH-RATES.—
In 1904 the death-rate in England and Wales from phthisis
was 1-46 per 1000 of population among males and 1-03 per 1000
among females. The death-rate varies greatly at different
ages, as will be seen from the table on preceding page, derived
from Dr. Tatham's Report to the Registrar-General. In this
table the death-rates are stated per 100,000 living at each
age-period separately for the two sexes.

The facts in this table can be more clearly seen when set
out graphically as in Fig. 2.

The significance of the different age distribution of the
phthisis death-rate in the two sexes will be subsequently con-
sidered (p. 168). At present we need only record the fact,
as bearing on the value of the lives sacrificed to this disease.
The age distribution of deaths from phthisis may be stated in
three different ways : —

(1) The death-rate from this disease may be given per 1000
or per 100,000 living at each period of life, as in Fig. 2.

(2) The deaths from this disease may be stated in proportion
to the total deaths from the same disease at all ages.

(3) Or these deaths may be stated in proportion to the total
deaths from all causes at the same age-period.

The first is the only method which can be employed in com-
paring the age incidence of the disease in different populations.
The second and third methods are useful for special purposes.
By means of the second method we can ascertain the proportional
incidence of deaths from phthisis at different ages, and by the
third we can state its importance in proportion to other causes
of death at each age-period. From these standpoints the second
and third methods tell us more than the first ; for a high death-
rate may occur among a relatively small population. Thus
the male death-rate from phthisis of 126 per 100,000 at ages
over 65 is higher than that of 39 per 100,000 in male children
under 5, but the two rates represent an equal percentage
(3*1) of the total male mortality from this disease at all ages.
In the following table the second and third ratios mentioned
above are given for each sex : —

THE PREVENTION OF TUBERCULOSIS

TABLE V. — ENGLAND AND WALES, 1904
Proportional Mortality from Phthisis

Males.

Females.

Age.

(l) In propor-
tion to 100
Deaths from
Phthisis at all
Ages.

(2) In propor-
tion to 100
Deaths from all
Causes in the
same Age-
period.

(i) In propor-
tion to loo
Deaths from
Phthisis at all
Ages.

(2) In propor-
tion to loo
Deaths from all
Causes in the
same Age-
period.

3'i

3 '4

4-2

2'0

10-

i '4

9*5

4'3

20'6

15-

5'6

26-1

35'4

20-

10-8

38-5

1 1 '9

367

25-

23-1

37-o

25'3

32-3

35-

22-8

28-1

20-3

21 '2

45-

18-9

18-2

13-0

1I'2

55-

IO'I

6-9

4*4

65 and upwards

3'i

i '4

3-2

0-8

All Ages .

lOO'O

8-5

lOO'O

6-0

The same facts are set forth graphically in Figs. 3 and 4.
Comparing the three sets of facts depicted in Figs. 2-4, it
will be noted that the highest male death-rate from phthisis
occurs at the age-period 45-55, the age-periods 35-45 and 55-65
coming next. The highest proportion of the total male deaths
from phthisis occurs at the ages 25-35 an(l 35~45 \ and phthisis
bears the highest proportion to deaths from all causes at the
ages 20-25 and 25-35.

In the female sex the highest death-rate from phthisis occurs
at the ages 35-45, 25-35 coming next, the highest proportion
to deaths from phthisis at all ages occurs at the ages 25-35,
and to deaths from all causes at the corresponding age-period
at ages 20-25.

Of the total deaths from phthisis 91-3 per cent, in males
and 87-1 per cent, in females occur at the ages 15-65, the working
years of life.

TUBERCULOUS DISEASES OTHER THAN PHTHISIS. — Phthisis is
not the only fatal disease due to tuberculous infection. In 1904

\JHDL

YEAR

10-

-10

15-
-20

20-
25

25-35

35-45

45-55

55-65

65,

JPWARDS

FIG. 2.— England and Wales, 1904.— Male and Female Death-rates

from Phthisis at different Age-periods
(Males — continuous line; females — interrupted line)

THE PREVENTION OF TUBERCULOSIS

FIG. 3. — Deaths from Phthisis at each Age-period per 100 Total

Deaths from the same Disease at all Ages
(Males — continuous line; females — interrupted line)

MAGNITUDE OF THE- EVIL

32-

Ill

ill1

UHDER

5
YEARS

10-
-15

IS-
-20

20-
-25

25-35

35-45

45-55

UPWARDS

FIG. 4. — Deaths from Phthisis at each Age-period per 100 Deaths from all
Causes at the same Age-period

THE PREVENTION OF TUBERCULOSIS

the number of deaths caused by each form of tuberculosis was
returned as follows :—

TABLE VI. — ENGLAND AND WALES, 1904
Number of Deaths caused by various forms of Tuberculosis

Males.

Females.

Total.

Pulmonary Phthisis

23,850

18,001

41,851

Tuberculous Meningitis

3,359

3,030

6,389

Tuberculous Peritonitis \

1,994

1,921

3,915

Tabes Mesenterica J

1,064

834

1,898

Lupus ....

Tubercle of other Organs

957

705

1,662

General Tuberculosis

2,253

2,062

4,315

Scrofula .....

IOQ

33,552

26,653

60,205

The death-rate in 1904 from all the tuberculous diseases ex-
cluding phthisis was 54 per 100,000 persons, 59 per 100,000 males,
and 50 per 100,000 females ; the corresponding figures for all
tuberculous diseases being 178, 205, and 153. Thus phthisis
accounts for 69-5 per cent, of the total deaths ascribed to tuber-
culosis.

The age distribution of the deaths from tuberculous diseases
other than pulmonary tuberculosis enumerated in Table VI.
will be more conveniently discussed in the chapter on Accuracy
of Certification.

CHAPTER II

MAGNITUDE OF THE EVIL: B. SICKNESS AND
ECONOMICS

IT has been shown in Chapter I. that n per cent, of the
total deaths in England and Wales are registered as due
to tuberculosis, and that seven-tenths of these are caused
by phthisis. Table V. also shows that among males 91 per
cent, and among females 87 per cent, of these deaths occur
between the ages 15 and 65, and 86 and 77 per cent, in the two
sexes respectively at ages 20-65.

ECONOMIC VALUE OF LIVES LOST. — Each child during his
years of helplessness and until he is able to support himself by
his own exertions is having expended upon him time, money,
and effort, which may be regarded as so much capital invested
with a prospect of future returns. If he dies in infancy, the
measurable loss is much less than if death is postponed until
the age of 15. Although it is scarcely necessary to make
elaborate calculations as to the expenditure on maintenance,
etc., which is lost by death occurring before or during school-
life, it obviously represents a considerable capital sum. Between
the ages of 15 and 20 it is probably exceptional for the earnings
to more than balance personal expenditure, and, if this be so,
all deaths up to the age of 20 may be regarded as involving a
serious loss of capital expenditure. After this age the problem
becomes more complicated. During the next five years a large
proportion of the population marry, and thus incur new obliga-
tions before the balance against them can possibly have been
paid off. It is during the ages from 25-65, and especially during
the ages 25-55, that the worker can hope to pay back the value
of his own earlier maintenance (a) by personal savings, (b) by
investing capital in the formation of a home and the upbringing
of a family in his turn. Each family represents in this respect
an investment on the instalment system, and the only hope of

THE PREVENTION OF TUBERCULOSIS

completing the investment, and leaving no debt for survivors
to redeem or owe to the community, is for the worker to live
and to remain able to work, until all his children are able to
earn their livelihood, and until his wife and himself can maintain
themselves in their old age. That is the ideal. It can only be
realised when the worker is not cut down by illness or killed
by disease or accident. Hence the immense economic significance
of the fact that among men nine out of every ten deaths from
phthisis occur between the ages of 15 and 65. Some data for
the determination of this loss have been calculated by Dr. T. E.
Hay ward (igo^.1

EFFECT ON THE DURATION OF LIFE OF THE ELIMINATION OF
PHTHISIS. — Dr. Hay ward calculated by the life-table method
what would be the effect of totally abolishing phthisis from
the death-returns of England and Wales for the decade 1891-
1900. The main results thus obtained are summarised in the
following table : —

TABLE VII. — ENGLAND AND WALES, 1891-1900
Survivors and Future Expectation of Life at Different Ages in Males

Number of Survivors at

Future Expectation of

each Age out of
100,000 born.

Life (Mean After
Lifetime).

Percentage
Increase

in the

Expectation

Age.

Based on the

of Life

Based on the

Mortality

Based on the

Mortality

produced

Mortality

from all

Mortality

from all

by the

from all

Causes

from all

Causes

Elimination

Causes.

excluding

Causes.

except

of Phthisis.

Phthisis.

100,000

44-1

46-3

5'0

75>093

75,256

53 '4

56-2

5'3

15-

72,592

72,897

45'i

47'9

6'3

25-

69,446

70,654

36-9

39'2

6'3

35-

64,716

67,676

29-2.

30-8

5'3

45-

57,655

62,138

22'2

23-0

3 '9

55-

47,424

52,742

I5-8

16-2

2-3

65-

33,163

37,830

I0'3

10-4

I'O

75-

15,813

18,303

6-1

6'2

85-

3,121

3,629

...

1 It is convenient to note here that when a date is given in brackets after a
name, the full title of the paper or book quoted will be found in the bibliography
at the end of this volume. The same remark applies when a name and a page
reference are given in brackets.

MAGNITUDE OF THE EVIL 15

It will be observed that the number of survivors from infancy
to the age of 15 out of a given number born is not materially
increased by the elimination of phthisis. From this point
onwards the elimination of this disease would steadily increase
the number of survivors. At the age of 55, for instance, the
number of survivors would be n per cent, greater than under
the actual conditions holding good in 1891-1900, while the mean
expectation of life would be increased by 2*3 per cent.

FINANCIAL GAIN BY THE ABOLITION OF PHTHISIS IN MEN. —
Some conception of the financial gain that would be secured
were pulmonary tuberculosis abolished is given by Table VII.,
which shows that, judging by the experience of 1891-1900 in
England and Wales, the abolition of phthisis would increase the
expectation of life of every male aged 15 years by 2 '8 years, and
of every male aged 25 years by 2*3 years. Taking the average
increase of expectation for the 3,080,166 males aged 15-25 at the
last census (1901) to be 2*5 years, it follows that these males
who, in 1901, were at or near the beginning of their working
life would, but for phthisis, live in the aggregate 7,700,315 years
more than under present conditions they can expect to do. A
reference to Table VII. shows that the greatest part of this
increase of life would be in the working years of life before 65 ;
and if we assume that the average wage of each is 2os. a week,
a possible gain of over £400,000,000 might be obtained on the
above lives, or not far from ten millions sterling annually. And
this makes no allowance for the loss sustained by protracted
sickness ; nor for the further loss from premature death of
women from the same cause.

ILLUSTRATIONS OF FINANCIAL Loss BY PHTHISIS. — (i) The
experience of Friendly Societies throws light on this point.
Mr. A. W. Watson (1902) has published an investigation of the
experience of 819,716 members of the Oddfellows Society during
the years 1893-97. These members represented persons exposed
in the aggregate for 2,995,724 years to risk of sickness, and for
3,180,378 years to risk of death. During these years the average
annual death-rate per 1000 members was 12*3. This Society
has not published any results as to causes of mortality, but
the Ancient Order of Foresters has published (1903) a report
summarising for the five years 1897-1901 the number 'of total
deaths and deaths from consumption which occurred among its

THE PREVENTION OF TUBERCULOSIS

580,405 benefit members, equivalent to 2,721,822 years of life.
The following table summarises the results for them and for
224,374 wives and widows of members during the same period : —

TABLE VIII. — FORESTERS
Death-rates from all Causes and from Consumption, 1897-1901

Benefit Members.

Wives and Widows of Members.

Death-rate per 1000 from —

All Causes.

Consumption.

All Causes.

Consumption.

England
Ireland .
Scotland
Wales .

United Kingdom .

I3'2

I2'I

9'6
127

r8
27

2'6

1-8

I2'I
12-0
IO'I
I2'8

i;s

1-8

'7.;J

12-9

1-9

11-9

i -5

It is evident that the experience of the Foresters and the
Oddfellows as regards general death-rates is very similar, and
it may be assumed that this is so also for consumption, and that
in both Societies this disease causes at least 15 per cent., or
about one-seventh of the deaths from all causes. Returning
for a moment to Table V. and Fig. 4, it will be noted that the
proportion of deaths from phthisis to total deaths from all
causes is greatest from 20 to 45 years of age, at which ages
it varies from a third to a fourth of the total number. At ages
55-65 it has declined to one-twelfth of the total deaths from all
causes. In the total experience of the Foresters the proportion
is, as we have seen, one-seventh, and the proportion must be
higher than this in the working years of life 15-65. Further
allowance has to be made for the fact that consumption only
causes death after prolonged disablement, and almost certainly
causes a higher proportion of the total sickness than of the total
mortality. Assuming that it causes one-fifth of the total dis-
ablement at ages 15-65, we can calculate what this meant for
the 819,716 members of the Manchester Unity of Oddfellows
during the years 1893-97. According to Mr. Watson's tables,
these men experienced in these years 4,707,680 weeks of sickness,
of which 941,575 must be attributed to consumption. The

MAGNITUDE OF THE EVIL 17

expense to the Oddfellows of this amount of sick relief, and of
the deaths associated with it, must have been at least half a
million sterling, and the loss of wages to the men themselves
at least double this amount.

At a time when Friendly Societies are finding that the claims
on their funds are necessitating higher contributions or smaller
benefits, their wisest policy evidently is to aid by every means
in their power in diminishing this serious drain on their resources.

(2) As bearing on the experience of English Friendly Societies,
facts given by Mr. Hoffman (1901) relating to the experience of
the Prudential Insurance Company of America may be given.
He shows that " at the ages of most importance for Industrial
insurance purposes almost one-half of the entire mortality is due
to consumption." His statistics, unfortunately, do not give
the number of lives at risk, but his facts are nevertheless most
suggestive. He says : —

The annual cost of deaths from tubercular diseases to the Prudential
Insurance Company of America is approximately, on the basis of three
years' experience, the sum of $800,000. Over 6000 deaths are annually
due to this cause in our experience at the present time. . . . While on the
average we have received $24.00 from those who died from consumption,
we returned to the beneficiary under Industrial policies over $134.00, a net
loss of about $110.00 on every case, or more than half a million dollars
during the course of a year. Of course, there is a great difference as to the
losses sustained at different age-periods, and naturally the income is least
at the younger ages. As age increases, the average duration of insurance
increases, and the amounts paid in premiums to the companies tend
more to approach the amounts paid out in claims, but the fact remains,
that taking the business as a whole we lose about $i 10.00 on every death
from consumption which occurs in our experience at the present time.
If you examine these facts closely you will realise the great interest of
the Industrial companies in the problem of diminishing the mortality
from tuberculosis, especially at the early ages when, as for instance at
25-29, we will have received $18.00 in premiums to every $150.00 paid out
for losses.

(3) Dr. Hermann Biggs (1903) after a careful estimate
places the expense of tuberculosis to the people of the United
States at $330,000,000 (£66,000,000). He first calculates the
loss to New York City by putting a value of $1500 (£300) upon
each life at the average age at which deaths from tuberculosis
occur. This gives a total value of £3,000,000 for the lives lost
annually. To this has to be added the loss due to the fact that

i8 THE PREVENTION OF TUBERCULOSIS

for at least nine months before death these patients cannot
work ; and the loss of service at $i a day, and the cost of food,
nursing, medicines, attendance, etc., at $1.50 a day results in
a further loss of $8,000,000 (£1,600,000), making a yearly loss
to the city from tuberculosis of $23,000,000 (£4,600,000). The
estimated annual total of 150,000 deaths from tuberculosis
in the United States represents in the same way a loss of
$330,000,000 (£66,000,000). He further points out that the
total expenditure in the City of New York in the care of tuber-
culous patients is not at present over $500,000 (£100,000) a
year — that is, it does not exceed 2 per cent, of the actual loss
by death, etc. " If this annual expenditure were doubled or
trebled, it would mean the saving of several thousand lives
annually, to say nothing of the enormous saving in suffering."

(4) The experience of the German Imperial Insurance Office
ascribes a much higher proportion of the total sickness to con-
sumption than the one-fifth which I have tentatively given on
the basis of the one-seventh proportion of deaths in the experi-
ence of the Foresters. Bielefeldt reports that of every 1000
German workmen aged 20-25 who are rendered unfit for work,
548 owe their sickness to tuberculosis, while at ages between 25
and 30 the proportion per 1000 is 521. At the higher ages, as
the amount of non-tuberculous sickness increases, the proportion
of tuberculosis becomes less.

(5) In publications of the National Association for the Pre-
vention of Tuberculosis, it is estimated that one-eleventh of the
total cost incurred in the relief of pauperism in England and
Wales is caused by consumption. The total expenditure in
poor-law administration in the year ending Lady day 1907 was
£14,035,888, so that on this basis considerably over a million
sterling is annually spent on paupers who were made such by
consumption.

(6) The experience of the workhouse infirmary of Brighton
gives some insight into the immense cost incurred in the support
of parochial consumptive patients. That part of the borough
of Brighton comprised within the parish of Brighton has a
population of about 102,000. During the eight years 1897-
I9°5> 372 consumptive patients were treated in its infirmary.
The average and total stay of these patients in the institution
is shown in the following table : —

MAGNITUDE OF THE EVIL

TABLE IX. — PHTHISIS

Brighton Workhouse Infirmary Statistics from July 15, 1897, to
May 23, 1905

Total Number of Days spent in Workhouse by
Patient before-

I. Patient only Once in Work-
house.

Leaving
Workhouse.

Death.

May 23, 1905.
(Still In).

Number of days under each head-

11,128

21,306

9133

ing-

Number of patients under each head-

148

ing.

Average number of days for each

114

144

507

patient.

2. Patient Twice in Workhouse.

Leaving
Workhouse
2nd Time.

Death during
2nd Stay.

May 23, 1905.
(Still In).

Number of days under each head-

2998

5883

5521

ing.

1 Number of patients under each head-

ing-

Average number of days for each

I76

490

1380

patient.

3. Patient Three Times in Work-
house.

Leaving
Workhouse
3rd Time.

Death during
3rd Stay.

May 23, 1905.
(Still In).

Number of days under each head-

2146

2874

261

ing-
Number of patients under each head-

ing.

4. Patient Four Times in Work-
house.

Leaving
Workhouse
4th Time.

Death during
4th Stay.

May 23, 1905.
(Still In).

Number of days under each head-

966

3217

924

ing.
Number of patients under each head-

ing.

5. Patient Five Times in Work-
house.

Leaving
Workhouse
5th Time.

Death during
5th Stay.

May 23, 1905.
(Still In).

Number of days under each head-

613

...

ing.

Number of patients under each head-

...

ing.

6. Patient Six Times in Workhouse.

Leaving
Workhouse
6th Time.

Death during
6th Stay.

May 23, 1905.
(Still In).

Number of days under each head-

337

3259

ing.

Number of patients under each head-

ing.

N,B. — The word "Workhouse" is used to include Infirmary.

The average stay of each patient was 221 days, including those still in.

20 THE PREVENTION OF TUBERCULOSIS

This on the basis of 145. a week l means a total cost for main-
tenance and treatment of £8221, or an annual cost of over £1000
a year. If we assume that the expenditure per 1000 of popula-
tion is the same in other parts of the country as in Brighton,
this implies that on the indoor relief, i.e. on the institutional
treatment of consumptives in workhouse infirmaries, an annual
sum of about £331,000 is spent in England and Wales. This
estimate makes no allowance for the large sums given in relief
of the relatives of consumptives both before and after their
death, and in relief of consumptives who are allowed to remain
at home instead of going into infirmaries. If these items be
added together, it is likely that they would exceed the annual
sum of a million sterling, and would confirm the estimate quoted
in paragraph (5).

(7) Farr (1885) stated that the number constantly sick
to one annual death was 2*8 in the police and in some friendly
societies. According to the experience of the Manchester Unity
of Oddfellows during 1893-97 there were 3-35 years of sickness
for every annual death at ages 20-65. Although consumption is
more chronic than most disabling forms of disease it is doubtful
if it causes on an average 3 years of disabling sickness. Doubt-
less in the above average (3-35 years for every death) is included
much sick-leave for minor complaints ; and it appears likely
that the amount of sick-leave given for comparatively slight
ailments has increased. If, however, we assume that only
one year's disablement is caused by every fatal case of con-
sumption, then the direct loss per annum in England and Wales
produced by the death of men aged 20-65 from consumption,
reckoning wages at £50 a year, judging by the experience of 1904,
amounts to £1,015,400. This is the loss in wages, reckoned at
the above rate. No allowance is made for the cost of the illness,
for the interference which every sickness involves with the work
of others, or for the infection of others and resultant further loss
of health and money.

The preceding calculations are merely given as illustra-
tions of the terrible national loss of money and efficiency caused
by tuberculosis. They fail to show the full extent of the mis-
chief wrought. Looking at the subject from the standpoint

1 This is about the average cost in an infirmary calculated separately from
the workhouse.

MAGNITUDE OF THE EVIL 21

of national economics, it is not open to dispute that the most
elaborate and complete measures of every description against
tuberculosis would only cost a fraction of the present total
loss inflicted by this disease, and that this expenditure would
as time goes on be paid for many times over in the prevention
of sickness and increase of efficiency of the community.

CHAPTER III

ARE THE STATISTICS RELATING TO TUBERCULOSIS
TRUSTWORTHY ?

HAVING obtained some idea of the amount of havoc at
present wrought by tuberculosis, we must — before con-
sidering the changes in this respect in this and other
countries — ascertain what degree of confidence can be placed
in the official statistics of this disease.

COMPLETENESS OF CERTIFICATION OF CAUSES OF DEATH.—
In drawing deductions from our national statistics, it must
be borne in mind that, although national registration of births
and deaths was inaugurated in 1837, it was not until January i,
1875, that it became compulsory for medical practitioners to
give certificates of the cause of death of each patient dying
under their care. Before this duty became compulsory, medical
practitioners certified the majority of deaths, but Farr (1885,
p. 523) notes that in 1871 about 8 per cent, of the total deaths
were not medically certified. The proportion in 1904 had
declined to 1*4 per cent.

There is little doubt that the incomplete medical certification of
deaths must affect the trustworthiness of the statistics for phthisis
for years before 1875, though to what extent cannot be stated.
It is not likely that it does so to such an extent as to make the
figures before and after 1875 incomparable. This appears to
follow from the regularity of the fall in the death-rate from
phthisis before and after this year ; but the gradually increasing
completeness in medical certification of causes of death needs
to be borne in mind.

Beyond this there is the further point as to the gradually
increasing accuracy of medical certificates. There can be
little doubt that deaths certified at the present time in this
country to be due to phthisis are, as a rule, correctly returned.
The following exceptions to this rule require to be noted : —

ARE THE STATISTICS TRUSTWORTHY ? 23

(a) Inaccurate Diagnosis in Children. — In children, the term
broncho-pneumonia not infrequently conceals acute tuber-
culosis, especially when the " broncho-pneumonia " occurs
after imperfect recovery from such diseases as whooping-cough
and measles. Coates (1891) has drawn attention to the fre-
quency of errors of diagnosis in children. He quotes the figures
of the Great Ormond Street Children's Hospital, London, for
1877, which showed that of 77 deaths from all causes 35*5 per
cent, were due to tuberculosis ; and he considers that we may
safely affirm that of the total deaths under 10 years of age
among the masses of the people, one-third are due to tuberculosis.
In Paris, according to Landouzy (Trans. Tuber c. Congr. Paris,
1888, p. 202), one-third of the deaths under 2 are due to tuber-
culosis. Compare these statements with the experience shown
by our national returns for 1904, as given in Table X.

TABLE X.

Percentage at each Age of the total Deaths from all Causes in England
and Wales in 1904, which were returned as caused by Tuberculosis
(all forms)

Aged

All Ages under
5-

Aged
5-10.

O-I.

1-2.

2-5-

4-2

11-9

9-2

6'2

19-1

Table V. gives similar facts for phthisis alone. In explaining
the discrepancy between the percentages in early life given
in Table X., and the statements made by Coates and Landouzy,
it has to be remembered that the latter are dealing only with
hospital statistics, and both probably have included deaths
in which tuberculosis was secondary to other diseases (e.g.
whooping-cough), whereas in the Registrar-General's returns
these would be entered under the heading of the primary disease.
When allowance is made for these facts, there remains, probably,
in the national returns considerable understatement of the
mortality from tuberculosis in early life, which is not com-
pletely counterbalanced by the return of many deaths as " tabes
mesenterica," in which there is no tuberculosis. Th'ere is no
evidence that recent statistics of tuberculosis in early life are

THE PREVENTION OF TUBERCULOSIS

not fairly comparable with those of past years, and there is
some evidence to the contrary.

(b) Inaccurate Diagnosis in Old Age. — Concerning the other
extreme of life Dr. Glover Lyon has expressed the belief that " if
all the deaths from senile phthisis were properly registered, the
registered mortality from phthisis would increase right up to the
end of life, as is the case in New York." The diminution of
mortality from phthisis after the age of 60 he believes is entirely
due to erroneous certification. Dr. Lister also has drawn atten-
tion to the fact that in cases in which there is senile emphysema
and bronchitis, great difficulty is often experienced in diagnosing
phthisis clinically. Error may therefore creep in at these old
ages. There is no internal evidence to show that in our national
statistics these possible sources of error have been acting at
different periods to a markedly varying extent.

(c) Inaccurate Diagnosis at all Ages. — I. Confusion between
Phthisis and other respiratory Diseases. — The most likely sources
of error in phthisis statistics are deaths returned under the
headings of bronchitis and pneumonia. In the following table
the comparative death-rates from these diseases are given for
a series of years :—

TABLE XL — ENGLAND AND WALES
Death-rates per 100,000 of Population in successive Periods from —

Period.

Bronchitis.

Pneumonia.

Bronchitis
and
Pneumonia.

Phthisis.

5 years, 1866-70

5 1871-75
5 1876-80
5 1881-85
5 1886-90
5 1891-95
5 1896-1900
4 1901-04

191

222
238

215

214
207
I56

168

107
103

100
IOO

113
125

120
121

298

315
327
332
276
289

245

222
204

183
164
146
132
123

The question arises whether the rates in Table XL for years
before 1875 are comparable with the later rates. Comparing
1871-75 with the two succeeding quinquennial periods, no
change in the pneumonia death-rate is visible, and little, if
any, change in the death-rate from bronchitis. The following

ARE THE STATISTICS TRUSTWORTHY ? 25

diagram shows the difference in the course of phthisis and of
bronchitis and pneumonia together (thus combined because

cfl

FIG. 5.— Relative Death-rates from (a) Phthisis, (b) Bronchitis and Pneumonia
in England and Wales, the rates for 1901-04= 100

26 THE PREVENTION OF TUBERCULOSIS

there may have been transference between these two, especially
between capillary bronchitis and broncho-pneumonia). The
death-rates from phthisis and from bronchitis and pneumonia
respectively in 1901-04 are stated as 100, and earlier rates given
in proportion to this figure. By this method, which is adopted
in several other instances throughout this work, the items com-
pared start from a point of the same magnitude, and the varia-
tions under each heading are comparable on the same scale.

There is no evidence in Fig. 5 that phthisis has declined in
consequence of transfer of deaths from that heading to bronchitis
and pneumonia. The possibility of confusion between pneu-
monia and bronchitis and phthisis can be further tested by
a comparison of the age distribution of the death-rates from
these diseases in 1861-70 with that of 1901. This is done in
the table on next page for males, and in Fig. 6, which sets out
the same facts graphically.

It will be noted that in Fig. 6, and in each of the columns of
comparative figures in Table XII., the death-rate at all ages in
the aggregate is stated as 100, and the rates for different age-
periods are stated in proportion to this. It has not been thought
necessary to reproduce the table and diagram for the female sex,
as the result is the same as for males. By means of Fig. 6 we
can compare for each age-period the relative incidence of fatal
phthisis and of fatal bronchitis plus pneumonia at each age-
period in 1861-70 with that in 1901. The comparison is in-
teresting, as it affords no evidence that there has been any
considerable transfer between bronchitis plus pneumonia and
phthisis. Some postponement of the maximum death-rate from
phthisis is seen in Fig. 6 to have occurred in males, and the
same change has occurred for females.

So far, then, as confusion with other diseases is concerned,
it does not appear likely that the phthisis statistics of recent
years are to any serious extent incomparable with those of
earlier years. Phthisis when a fatal disease is easily recognised,
and the official figures within a limited margin may be regarded
as approximately true.

2. Return of Phthisis as " Tuberculosis." — Nor does it appear
probable that the tendency on the part of doctors which has
shown itself in recent years, to return deaths as " tuberculosis "
without any statement of organ affected, has caused a serious

ARE THE STATISTICS TRUSTWORTHY ?

Q
53
<

<^>
•«
^

.> .

N 1-1 fO'd-rotnu->M o t^>-<
C^ t-i i-iu-)i-i rfC\i-Ht^O •*!•

c3 <u

Ou tJD

|£

vO *>•

O oo LO o\ r^\o ONVO ^J- N t-
•^ HH N oo LOVQ vo LO roao N

cx>c»

HH •— '

|
|

o voo ooovo HH inmoNHH
conn NOOVO i-i CT.HH mn mvo

HH N N CO N HH

ON
TJ-

"Si

00 00

ON eo HH o^oo OM^VO O N vO
^^JHC^O^COOVO

•* .

.a
|

M-j'Ji-ioo ^Nt^vo O O"^

r>.|_ HHHHNThONMNVO

en N -^-oo

I.&

JL d

vO *>•

Tt- ^- U-JOQ M O i>- LOOO ON r^
vr> HH HH N rl-00 OO 00 HH
CT> HH rot>.

t/j
IS

£
1

§,

Tf t^ mvO ON CNCQ mt^Tj-ON
«N COHH N roLONvo r^mr^

O HH C4 IO HH CO
HH HH N

xn
&

2
«

,C
1

« 6

vO *^-

oooo

rt m TJ- moo moo fo o N "*•
O> ^ HH d cnvo N vo 00 O HH
O HH N 10 N N

M HH N

.> .
11

HH HHOooot^ON^nO ooo O

OSN HH CJ C0t^ M t^m N

C<^ HH HH d CO

&.SP

jfe

w 6
vD *^»

oooo

MNOOfONHHONHHio rovo
VON HHNCO^t^O "^OO

C3
£

vr> ONOO -^-vo CO ro •<*• covo ON
COM NcouoOvOcoNco

IT! HH HH N fO-*

CO
M

J3
1

JL d

<o t^

00 00

MOO O t^t^ONHH ONN COrj-
OO W HH HH M COO 00 CO ON CO

VO HH HH N

vO
N

ra <U

lOVOHH HH N T^OO N mO O
U-> HH tv. COO HH

CO M VO ri-

z/5

1-5

'rS

§-.§>

JL d

O O N covO •^-t^vo m HH O
oo HH HH co ON rr >-noQ
N N mO

c/5

"cd

O>00 M M covO "^ ON Tf OO O
00 N ONN N ^T
•^J- COCX) ON

g£

1— 1 w

N VO ^vo HH int>,Tj-OO ON HH

HHM HHNVOt^^OOO

vr> HH rT O ON

tf>

O "^O mO uiinmu^inro

HH HH N N CO <5f "^VO t^

•!

FIG. 6. — Comparison between 1861-70 and 1901 of relative Death-rates at
different Age-periods from Bronchitis plus Pneumonia and from Phthisis

ARE THE STATISTICS TRUSTWORTHY ?

transfer from phthisis. When checking the mortality returns
of Brighton for three years, I found that 496 deaths were
returned as phthisis and 39 as tuberculosis, acute tuberculosis,
or miliary tuberculosis. Many of these, doubtless, had not had
recognisable pulmonary tuberculosis, and were properly re-
turned ; and the residuum would only slightly reduce the great
decline in the death-rate from pulmonary tuberculosis which
has occurred. Thus, if in the figures for the whole of England
and Wales, given in Table VI. on page 12, half of the 4315 deaths
from general tuberculosis were transferred to phthisis, the
phthisis death-rate would only be changed from 1*24 to 1*30
per 1000 of population.

TUBERCULOUS DISEASES OTHER THAN PHTHISIS. — Tuberculous
diseases other than pulmonary in 1904 caused 29 per cent, in
males and 33 per cent, in females of the total deaths from tuber-
culosis. We must next inquire into the validity of the death-
returns under these headings.

TUBERCULOUS MENINGITIS. — We may adopt the same method
of age comparison as for phthisis ; only in this instance 1871-80
must be compared with 1901, because in 1861-70 the decennial
supplement of the Registrar-General did not separately tabulate
this disease.

TABLE XIII. — ENGLAND AND WALES

Annual Death-rate from Tuberculous Meningitis per 100,000 Persons
of both Sexes at each Age-period

Age-period.

1871-80.

1901.

190

109

10-

15-

20 and upwards .....

All Ages

There has been a reduction in the death-rate from tubercu-
lous meningitis (acute hydrocephalus) , which corresponds roughly
with that from phthisis (Table XVI. ). Tuberculous meningitis
is nearly always secondary to other tuberculous diseases, as of

THE PREVENTION OF TUBERCULOSIS

the glands or joints. Apart from the presence of such other
diseases, and unless an autopsy is made, tuberculous cannot
with certainty be distinguished from other forms of menin-
gitis.

Most of the deaths from tuberculous meningitis occur at
ages under 5. In the following table the death-rates at each
individual year of the first five years of life at the earliest
period available in the Registrar-General's reports are compared
with those for 1901.

TABLE XIV. — ENGLAND AND WALES

Annual Death-rate from Tuberculous Meningitis per 100,000 Persons
of both Sexes at each Age

Ages

•

All Ages

All

erio .

under 5.

Ages.

1871 80 ......

7.68

IQO

•22

178

144

IOQ

Percentage Decline of Death-rate from

Tuberculous Meningitis

Corresponding Percentage Decline of
Death-rate from Phthisis between

1871-80 and 1901 ....

It will be noted that both in 1871-80 and [in 1901 the death-
rate from tuberculous meningitis at ages under 5 was about six
times that at all ages. This appears to indicate that the statistics
of the two periods are comparable. Of course it does not follow
that they are accurate, and the comparisons given in the two
lowest columns of Table XIV. of percentage declines in the
death-rate for each of the first five years of life with those of
phthisis, throw further doubt on this point. The statistics of
tuberculous meningitis in the first year of life are especially
open to doubt. H. Armstrong (1902) states that the post-
mortem records for eighteen years at the Liverpool Infirmary
for Children contain particulars of 85 necropsies in which tuber-
culous meningitis was found. Of these 10 were in the second year,

ARE THE STATISTICS TRUSTWORTHY ?

1 8 in the third year, and not one in the first year of life.
Fagge states that only three cases of tuberculous meningitis
in the first year of life were verified in Guy's Hospital in forty
years.

TABES MESENTERICA AND TUBERCULOUS PERITONITIS. —
Tabes mesenterica is a name which should correctly be applied
only when it is clear that the patient has tuberculous disease of
the abdominal lymphatic glands. Unfortunately it is often used
in death certificates when the patient has died from a slow
wasting disease accompanied or not by abdominal symptoms
such as diarrhoea. As Drs. Ashby and Wright state in their
work on Diseases of Children, " Mesenteric disease is much
more frequently diagnosed than discovered post-mortem."
Similarly, Dr. Donkin (Brit. Med. Journ., vol. ii. p. 1046, 1899),
says, " All kinds of intestinal and other disorders are constantly
styled tabes mesenterica by those who fail to cure them." The
usual condition mistaken for it is wasting or marasmus caused
by chronic gastro-intestinal catarrh. In the great majority of
fatal cases of tabes mesenterica, this disease is accompanied by
general tuberculosis. Tabes mesenterica is seldom and tuber-
culous peritonitis still less frequently a direct cause of death.
In the following tables these two diseases are included together.
The separate tabulation of tuberculous peritonitis in the Registrar-
General's returns was not begun till 1901.

TABLE XV. — ENGLAND AND WALES

Annual Death-rate from Tabes Mesenterica and Tuberculous Peritonitis
per 100,000 Persons of both Sexes at each Age-period

Age-period.

1871-80.

1901.

o- .......

203

10- .......

15-

20 and upwards .....

6
3

5
3

All Ages

Here again there has been a reduction in the death-rate

THE PREVENTION OF TUBERCULOSIS

similar to that in phthisis. The comparison for each of the first
five years of life is shown in the following table :—

TABLE XVI. — ENGLAND AND WALES

Annual Death-rate from Tabes Mesenterica and Tuberculous Peritonitis
tier 100,000 Persons of both Sexes at each Age

Ages.

T> •_ J

All Ages

All

rerioa.

under 5.

Ages.

1871-80 . . .

C77

?,06

20 T

1901

jjj

364

138

125

Percentage Decline of Death-rate from

Tabes Mesenterica ....

Corresponding Percentage Decline of
Death-rate from Phthisis between

1871-80 and 1901 ....

It will be noted that under i year of age the death-rate
from tabes mesent erica declined 32 per cent. This should be
especially noted because Thorne in his Harben Lectures for 1895,
comparing 1891-95 with 1851-60, showed an increase in the
infantile death-rate under this head of 27*7 per cent, and founded
on it an important inference as to the importance of bovine milk-
supply in the causation of tabes mesenterica. In view of the
opposing experience when 1871-80 is compared with 1901, the
only inference justifiable is that the statistics of infantile tabes
cannot be trusted. The general experience of pathologists is
that the number of deaths from tabes increase as the end of the
first year of life approaches. Compare with this the fact that of
the 2977 deaths registered in 1901 as caused by tabes mesenterica,
594 were at ages under 3 months, 1036 at ages 3-6 months, and
1347 at ages 6-12 months. Evidently many of the deaths
returned as tabes mesenterica would be found to be due to causes
other than tuberculosis were all death certificates verified by
autopsies.

Although there is a close correspondence in the aggregate
for all ages between the declines in the death-rates from phthisis

ARE THE STATISTICS TRUSTWORTHY ?

and from tabes mesenterica, this is not consistently so at the
earlier ages, and we must regard the statistics of this disease
as on a plane of trustworthiness much inferior to that occupied
by the statistics of phthisis.

COMPARISONS OF DECLINE IN DIFFERENT TUBERCULOUS DIS-
EASES.— Mention has been made of the parallelism of movement
of the death-rates from each of the forms of tuberculosis which
are tabulated separately by the Registrar-General. This point
is worthy of further study, in view of the side-light thrown by it
on the trustworthiness of the statistics.

TABLE XVI L— ENGLAND AND WALES

Annual Death-rate per 100,000 Persons of both Sexes from each of the chief

Forms of Tuberculosis

Period.

Phthisis.

Tuberculous
Meningitis.

Tabes
Mesenterica.

Scrofula.

5 years, 1850-54

281

5 1855-59
5 1860-64

265
257

39
37

26
27

5 1865-69

253

5 i 870-74 J

228

3°

6 1875-80

208

5 1881-85

183

5 1886-90

164

5 1891-95

146

5 1896-1900

132

4 1901-04

123

If we reduce the four columns of death-rates to the same
scale by giving each rate for 1901-04 as 100, and state all the
other rates in each column in proportion to this, a more exact
comparison can be made. The result is shown in Fig. 7. Evi-
dently the somewhat less trustworthy rates for tabes have
not consistently followed the law of decline which is shown
to an almost equal extent by phthisis and tuberculous menin-
gitis.

1 The figures up to 1879 are taken from the Annual Report of the Registrar-
General of Births and Deaths for 1880, p. Ixxix. The classification was altered
in 1 88 1, and the returns for scrofula before and after 1880 are not comparable.
After 1880 the last column in Table XVII. includes lupus, tubercle of other
organs, and general tuberculosis as well as scrofula.

THE PREVENTION

228

216

9 //>

2IO

ZOE

-09

****

**v

206

ooo

203

/Qn

180

160
ISO
14-0 \
I3O
120

100

"18

*
I

S"°

I73S

•e*
^

(73

169

•\

Ifi2

\ «.

154

•5.

i4-8

149^

a
•3

144

^C4

cfe

J38

129

sl34

\%>

I08

R
1

i
S

5
5

FIG. 7. — Relative Death-rates from different Tuberculous Diseases from 1850-54
to 1901-04, the Death-rate in the most recent period in each instance being
stated as 100

CHAPTER IV
THE HISTORY OF PHTHISIS

UNTIL the eighteenth century medical men confused under
the names of phthisis or its English equivalent consump-
tion all the acute and chronic diseases of the trachea,
bronchi, lungs, pleurae, and lymphatic glands when these were
accompanied by progressive debility and emaciation. In read-
ing the old descriptions of phthisis it is not difficult, however,
to recognise that pulmonary tuberculosis formed a large portion
of this congeries, and it is not without interest to trace, however
sketchily, the views as to the nature of phthisis which have been
held in different generations. We may deal first with what may
be described as the PR.E-ANATOMICAL PERIOD, in which post-
mortem dissections were rare, and in which views as to the nature
of phthisis were based almost solely on the symptoms recognised
during life.

Hippocrates (460-377 B.C.) described the disease, ascribing
it to [a suppuration [of t the ^ lungs, which may arise in various
ways. Galen (130-200 A.D.) also described it, and believed it
so infectious that it was dangerous to pass an entire day in the
company of a phthisical person (Walshe, p. 459). Hippocrates,
Galen, Aretaeus (circa 50.3.0.), and Celsus (circa 30 B.C.) all de-
scribed the disease, but not one of them appears to have recognised
the existence of the tuberculous nodules which form its char-
acteristic lesion. With the discovery of these we arrive at the

ANATOMICAL PERIOD. — Franciscus D. Sylvius (1614-1672
A.D.) was the first to recognise the causal relation of these
nodules to phthisis, so that the first step towards accurate know-
ledge of its pathology may be said to have been due to the
making of autopsies, which became fairly frequent in the
seventeenth century. Sylvius thought the nodules to be the
lymphatic glands of the lungs, and thus to be analogous to
scrofulous growths. Much speculation was devoted to these

36 THE PREVENTION OF TUBERCULOSIS

nodules ; and in the year 1700 Magnetus first described the
more minute nodules known as miliary tubercles, comparing them
to millet seeds, and showing their presence in the kidneys, liver,
and spleen, as well as in the lungs. Morgagni (1682-1772) dis-
puted the glandular nature of tubercles ; Thomas Reid (1778)
wrote of them as being not enlarged glands, but the products of
exudation. Matthew Baillie (1793) gave the following de-
scription of tubercles : —

Tubercles are firm white bodies interspersed through the substance
of the lungs, and apparently formed in the cellular structure ; for nothing
like a gland is to be discovered in the cellular membrane of the lungs in a
healthy state; and the follicles of the bronchi are not converted into
tubercles ; they are first very minute ; the clusters probably unite and
form larger masses ; the most common in size is that of a garden pea ;
they are firm in their consistence, and often contain a portion of thick
curdy pus. . . .

Thus Baillie recognised that the large nodules in tuberculosis
are produced by fusion of smaller tubercles. He described the
cheese-like substance of these large nodules as scrofulous matter,
recognising it and pus as the two characteristic products of
advanced tuberculosis. At the same time he attempted to
distinguish between caseating pneumonia and tubercles in a
condition of caseation.

Bayle (1774-1816), the precursor and teacher of Laennec,
published in 1810 the records of 109 autopsies on tuberculous
patients, and traced the minute tubercles through the subse-
quent stages of suppuration and caseation. He was of opinion
that phthisis was a disease not peculiar to the lungs but de-
pendent on a tuberculous diathesis or special constitutional
tendency.

Laennec (1781-1826) made investigations and published
teaching on tuberculosis which has been well described as a
tour-de-force of objective analysis. He taught that every
phthisis develops from tubercles, and that phthisis and tuber-
culosis are interchangeable terms, the tubercle being a new
product which appears either in isolated nodules or infiltrated
through the tissues. In both forms, he showed that it was
first grey and hyaline, gradually becoming opaque and very
dense, and later softening and discharging its contents through
the bronchi, leaving cavities in the substance of the lungs.

THE HISTORY OF PHTHISIS

5 ' Scrofulous glands were merely tuberculosis confined to the
lymphatic glands. Laennec denied the inflammatory origin of
tuberculous matter, and especially the transformation of pneu-
monia into tuberculosis. He was very sceptical also as to the
causation of tuberculosis by bronchial catarrh. In these respects
modern pathology has in the main confirmed his marvellous
insight.

Although Laennec's views were adopted by Louis in France
and by Hughes Bennett and others in Great Britain, the tyranny
of error gradually overshadowed Laennec's teaching, and what
is known as the dualist theory prevailed. According to this
theory, which cannot even now be said to be entirely abandoned,
most of the lesions of tuberculosis are not due to the tubercles,
but are primarily inflammatory in origin, the tubercles being
secondary to the inflammatory changes. Niemeyer formulated
this view in the words, " The greatest danger to which a
phthisical patient is exposed is that of becoming tuberculous/'

The PERIOD OF MICROSCOPICAL INVESTIGATION began about
1840, and although it did not solve the problem of the patho-
logical unity between caseous pneumonia and miliary tubercu-
losis, it was not fruitless. In 1844, Lebert thought he had found
distinctive tubercle corpuscles in the tubercles. Rokitansky,
whose book on Pathological Anatomy first appeared in 1842,
declared that tubercles were new growths composed of inspissated
proteins. The doctrine of " dyscrasia " or evil constitutional
conditions was then to the fore, and as a follower of this teaching
Rokitansky considered the " tuberculous habitus " to be very
important.

In 1847, Reinhardt showed that the so-called tubercle
corpuscles may originate from pus cells, thus diminishing their
importance. In the same year Virchow did much to buttress the
dualist theory by teaching that the process of caseation is not
peculiar to tuberculosis. In 1852 he limited the term " tubercle "
to miliary tubercles, which he described as new growths sub-
sequently changed by caseation, calcification, or fatty degenera-
tion followed by absorption. He is chiefly responsible for the
dualist theory which has done much to hinder the progress of
investigation.

A step towards unlearning this erroneous teaching was taken
when Buhl in 1857 showed that in at least 90 per cent, of his

38 THE PREVENTION OF TUBERCULOSIS

cases of tuberculosis in the lungs pre-existent caseous masses
were present somewhere in the body. He attributed the tuber-
culosis to these cheesy foci, infective products from which had
gained admission to the blood and then formed tuberculosis in
the lungs or disseminated miliary disease in various organs.
Here we have the first clearly expressed conception of miliary
tuberculosis as a self-infection caused by the absorption and
distribution of infective material derived from older foci in the
patient himself.

Although Laennec's teaching led to scrofula being commonly
regarded in France as the same disease as tuberculosis, in other
countries the belief in its separate origin has only recently dis-
appeared.

The adoption of EXPERIMENTAL METHODS OF INVESTIGATION
of tuberculosis led to further advance towards precision of
knowledge. From remote times the view that phthisis was
an infectious disease had occasionally been taught (p. 55).
Some early attempts at producing artificial infection were not
successful, and Klencke's successful experiments in 1843 were over-
looked. He injected tubercle cells taken from miliary tubercles
into the jugular vein of rabbits, and twenty-six weeks later
at the autopsy found widespread tuberculosis of liver and lungs.

Villemin's epoch-making experiments were published on
December 5, 1865. He inoculated rabbits subcutaneously
behind the ear with matter taken from grey and yellow human
tubercles, and found that (i) animals thus inoculated developed
pulmonary tuberculosis, (2) control animals which had not
been inoculated remained free from tubercle, and (3) other
animals similarly inoculated with pus from non-tuberculous
patients did not develop tuberculosis. Later on he obtained
results similar to those given under (i) by inoculating with
caseous material from tuberculosis, with the sputum of con-
sumptives, and with tuberculous material from a cow. Villemin
summed up the contents of his note to the Academic de Medecine
in the following words : " (i) Tuberculosis is the effect of a
specific causal agent, in short of a virus. (2) This agent must
reside like its congeners in the morbid products formed by its
direct action on the normal elements of the affected tissues.
(3) Introduced into an organism susceptible to its action, it must
continue to reproduce itself, and at the same time to reproduce

THE HISTORY OF PHTHISIS

5 I the disease of which it is the essential principle and the determin-
ing cause. Experiment has confirmed these results of induction."
He added : " Tuberculosis is a specific affection, caused by an
inoculable agent. Tuberculosis belongs then to the class of
virulent diseases, and in the nosological scheme must take its
place beside syphilis, but closer still to glanders."

The Academic de Medecine was not convinced. During the
following year Villemin worked continuously on new experiments,
and on October 30, 1866, reopened a discussion on the same
subject. Having been accused previously of experimenting on
rabbits already tuberculous, he took in his new experiments
animals of different species. His inoculation experiments suc-
ceeded in nine out of nine rabbits, in two guinea-pigs, in a dog
and in a cat. A sheep, a cock, and a pigeon remained immune.
Having by his extended basis of operation eliminated the element
of chance, he reaffirmed his conclusions, and a commission under
Colin was appointed by the Academic to investigate his results.
In its report of July 1867 it refused to accept Villemin's con-
clusions.

They were true notwithstanding ; and to Villemin belongs
the immortal fame of being the first to show the essential dis-
tinction in tuberculosis between the virus causing the disease and the
lesions produced by it. In 1868 he published his Etudes sur la
Tuberculose in which he further answered objections, vigorously
defended the idea of contagion, and argued against the existence of
a special tuberculous diathesis, a view which at that time domin-
ated and still influences medical minds to a great extent.

Villemin's experiments were repeated by others with varying
results. Progress was retarded by the fact that in some experi-
ments tuberculosis followed the inoculation of pus, particles of
sponge, and other apparently non-tuberculous materials. Burdon
Sanderson in 1868-69 confirmed Villemin's work, and the follow-
ing extract from one of his reports to the Medical Officer of the
Privy Council shows the stage to which he had brought the
investigation : —

As regards the question of a specific contagium of tubercle, we
think it very important to note that this is not as yet disproved by the
facts of traumatic tuberculosis. It still remains open to inquiry whether
or not injuries which are of such a nature that air is completely excluded
from contact with the injured part are capable of originating a tuberculous

40 THE PREVENTION OF TUBERCULOSIS

process. The results of the following experiments undertaken at the
instance of Mr. Simon, with special reference to this question, seem in-
deed to suggest that they may not be so. Setons steeped in carbolic
acid were inserted in ten guinea-pigs on the 24th of September 1868,
each animal receiving two. At the same time extensive fractures of both
scapulae were produced on five others, care being taken not to injure
the integuments. No tuberculosis or other disease of internal organs
resulted in either case : these facts certainly point to the necessity of further
investigation in this direction.

In 1876, when Simon ceased to be Medical Officer to the
Local Government Board, the specific infectivity of tuberculosis,
and the question whether this infectivity was dependent on a
specific organism, were matters which occupied the attention of
pathologists in all parts of the world ; but neither question had
been settled experimentally. Further trials by Chauveau and
Klebs in 1873 and by Baumgarten and Cohnheim in 1880 showed
that the discrepant results referred to above were caused by
faulty experimentation involving accidental infection of the
animals. Thus Cohnheim had in the first instance concluded
that tuberculosis is not a specific process. In a second series of
experiments, however, he inoculated animals in the anterior
chamber of the eye. By this means he was able to follow each
stage of evolution of tuberculosis of the iris and cornea, and to
establish fully its specific character. On the strength of these
experiments he foretold the early discovery of the parasitic
agent of tuberculosis. Before this discovery was made H.
Martin showed that the nodules produced by foreign bodies
were not inoculable in other animals, whereas true tubercles were
re-inoculable without any diminution in virulence. Thus the
specificity of tubercle was further demonstrated by its con-
tinuous inoculability in a series of animals. William Marcet
repeated Villemin's results by inoculation of guinea-pigs with
tuberculous sputum, and failed to produce similar results with
bronchitic sputum ; and he stated rightly that an inoculated
guinea-pig might thus serve as a means of diagnosis in doubtful
phthisis.

Before the final proof of the specificity of tuberculosis was
given, much advance was made in our knowledge of its methods
of spread. Chauveau proved that tuberculosis could be pro-
duced by eating meat, etc., containing tuberculous material, and
concluded that human and bovine tuberculosis were identical.

THE HISTORY OF PHTHISIS 41

He also showed that it was the particulate part of morbid secre-
tions which was capable of spreading infection. Villemin had
made the statement that tuberculosis could be spread by inhala-
tion of the virus, and pointed out the r61e of dried expectoration
in its dissemination ; and Tappeiner was the first to demonstrate
on dogs the possibility of dissemination of infection in this
way.

Pasteur's work rendered it likely that tuberculosis was due
to bacteria. It was found that the basic aniline dyes had a special
elective affinity for bacteria, staining them deeply. Ordinary
staining by this means failed to show bacteria in the morbid
growths of tuberculosis, but after various attempts Robert
Koch succeeded in staining the bacilli of tuberculosis by first
adding a small quantity of an alkali to the aniline stain, and thus
rendering it capable of penetrating the resistant outer membrane
of the tubercle bacillus. Other means of obtaining the same
result were subsequently discovered ; and a distinctive fact of
great importance was discovered, when it was found that even
strong mineral acids, which decolorised other stained bacilli,
failed to discharge the colour from the tubercle bacillus.

On the 24th March 1882, Koch contributed to the Physio-
logical Society of Berlin his note on " The Discovery and Cultiva-
tion of the Bacillus of Tuberculosis." He isolated, cultivated
outside the body, described, and differentiated the infective
organism of tuberculosis, and proved that it could continue to
produce the same lesions indefinitely. By a method of double
coloration, he showed the bacilli coloured blue on a brown
ground of vesuvin. He showed their presence in all known
tuberculous lesions and in tuberculous expectoration, and demon-
strated the virulence of the tubercle bacillus in expectoration
which had been dried for eight weeks.

Having thus traced the steps by which the crowning demon-
stration of the inf ectivity and of the infective agent of tuberculosis
was obtained, it will be convenient to summarise briefly the
pathological and clinical features of the disease produced by the
bacillus, and next to describe its biology, before dealing more
fully with the questions of infectivity and the conditions govern-
ing the spread of infection.

In looking back on the history of tuberculosis, three names
stand out pre-eminently — Laennec, Villemin, and Koch. It is

42 THE PREVENTION OF TUBERCULOSIS

chiefly to these three men, — the last of them aided by the wonder-
ful work of Pasteur and his followers, — that we owe the discovery
that tuberculosis is an entirely preventable disease. On their
work is based our exact knowledge of the nature of tuberculosis,
and the more accurate means for its prevention which we now
possess.

The history of phthisis since statistics became available is
given in the course of the argument of Part II. (p. 212*^ seq.).

CHAPTER V
THE MORBID ANATOMY AND SYMPTOMS OF PHTHISIS

THIS work deals solely with tuberculosis from the point
of view of preventive medicine and public health. Even
when we come to consider the sanatorium treatment
of consumptives, this will be chiefly considered as a means
of preventing others from becoming consumptive. Notwith-
standing this intentional limitation, the subject cannot be
discussed fully unless a short description of the pathology
and symptoms of tuberculosis is given. Such a description
is necessary not only before we can estimate the value of
sanatorium treatment, but also in order that the means of
spreading and preventing the spread of infection, and particu-
larly the phenomenon of latency, may be understood and their
importance appreciated.

Pulmonary tuberculosis is caused by the invasion of the
lungs by the tubercle bacillus. The terminal bronchioles end-
ing in the minute air vesicles or alveoli have a diameter of from
3 to 4 tenths of a millimetre, while the tubercle bacillus measures
from ij to 3 thousandths of a millimetre in length. So far
therefore as size is concerned, there is no difficulty in the tubercle
bacillus being drawn by inspiration into the air vesicles, where
it produces its evil results.

THE COMMENCEMENT OF THE INVASION. — The method by
which the bacilli reach the alveoli, whether by inspiration, by
spread from the lymphatic glands near the root of the lung,
by the blood circulation, or in all these ways at different times,
will be considered subsequently.

From the very commencement of the attack the tubercle
bacillus meets with resistance. Its opponents are some of the
wandering or patrol cells of the body ; in the earlier stages
they consist almost entirely of amoeboid cells or leucocytes,
derived from the blood and the marrow ; at a later stage larger

44 THE PREVENTION OF TUBERCULOSIS

wandering amoeboid cells are produced by the rapid prolifera-
tion of ordinary non-wandering connective tissue cells and
of the cells lining the alveoli or air vesicles. Both kinds have
the power of ingesting foreign substances, and are called
phagocytes.

Attracted chemically by soluble substances produced by
the bacilli, phagocytes migrate into the invaded area, and there
attack the invaders in two ways, (i) Under the irritation due
to bacterial toxins they throw off into solution complex sub-
stances called antibodies. These may act either by neutralising
the toxins, in which case they are called antitoxins ; or by
destroying the bacterium itself. (2) The phagocytes push
delicate fingers of protoplasm round the bacteria, which are
thus enveloped and afterwards absorbed. The importance
of this process of phagocytosis was first emphasised by
Metchnikoff. Sir Almroth Wright has recently shown that
phagocytes cannot absorb bacteria unless the latter have been
acted on previously by specific substances present in the fluid
part of blood. These substances he has called opsonins. They
are, like other antibodies, produced by the tissue cells and
leucocytes. In normal blood they are present in approxi-
mately constant proportion, but great variations occur in
disease. The bacillus therefore is opsonised by the surround-
ing exuded plasma ; its vitality is not affected, but it is in some
unknown way rendered absorbable by phagocytes.

THE PROGRESS OF THE INVASION. — If the invasion is small
and the leucocytes lusty, the invaders are vanquished. But
otherwise the invasion progresses. Leucocytes are killed by
the bacterial toxins, and their dead bodies accumulate as pus.
The leucocytes may even be a source of danger to the body.
They may pass with their load of bacteria into the surround-
ing tissues, and here, owing to their supply of intracellular
antibodies being insufficient, they may be destroyed by the
living bacteria within them, so that the bacteria are again free,
like the Greeks from the wooden horse in the siege of Troy.
It is at this point that we have to take up our description of the
lesions produced by tuberculosis.

THE LESIONS IN TUBERCULOSIS. — The tubercle bacilli have
entered the body and the leucocytes have failed to kill them.
The earliest and most characteristic lesion produced is the

MORBID ANATOMY AND SYMPTOMS OF PHTHISIS 45

grey tubercle. Its size varies from a pin's point to a pin's head,
or occasionally it may be as large as a small pea. It is grey
and slightly translucent. Under the microscope it is seen to
consist of a group of small and large cells containing tubercle
bacilli. The grey tubercles gradually become converted into
yellow tubercles, which are opaque, slightly granular, dry and
friable. They increase in size by coalescence, and then further
changes occur. Both grey and yellow tubercles are destitute
of blood vessels, but their presence causes inflammatory
changes in the surrounding vascular tissues. This often ends
in suppuration with the formation of an abscess, whose contents
find their way into the nearest bronchiole and are expectorated.
The cavity thus produced in the lung may go on discharging
muco-pus for years. It may join with other cavities to form
larger cavities ; the discharge from which produces gradual
exhaustion of the patient, while the toxic products absorbed
from them into the circulation produce the characteristic hectic
temperature of phthisis. Occasionally severe haemorrhage
(haemoptysis) occurs owing to the bursting of a blood vessel.
The cavity, if single, may gradually contract and heal. Many
consumptives with such cavities in their lungs have under
favourable conditions survived and worked for many years.

The change from grey to yellow tubercle is due to caseation,
a process so called because the diseased part has a cheesy appear-
ance and consistence. In chronic cases the caseous material
may become calcified, and at this stage the process may stop.
In small tubercles fibrous changes may occur, the diseased part
being converted into fibrous tissue.

Three figures in Hughes Bennett's Lectures on the Principles
and Practice of Medicine (ed. 1859) so clearly illustrate the
three stages of tuberculosis of the lungs that I have reproduced
them here. Fig. 8 shows the formation of grey tubercles and
some yellow tubercles. At the apex of the lung some of the
latter have broken down into an imperfect cavity.

In Fig. 9 a lung is shown in a more advanced condition
of disease. Tuberculosis is extensively infiltrated in the upper
lobe, and a considerable cavity has formed.

In Fig. 10 the third or last stage of pulmonary tuberculosis
is shown. The upper half of the lung is occupied by an
enormous cavity, and a smaller cavity has been excavated

46 THE PREVENTION OF TUBERCULOSIS

in the lower lobe. Very often the patient does not survive
long enough to show such extensive disease. Happily the
history of a large number of cases of pulmonary tuberculosis
is not correctly depicted in Figs. 8 to 10. There may be
only one or a few of the white dots (grey tubercles) shown in
Fig. 8, and these may completely heal by calcification or fibrosis.
In fact in very few cases of phthisis is the destructive process
continuous. As Hughes Bennett (p. 715) puts it :—

It is continuously checked, and for a time slumbers ; and all morbid
anatomists have recognised, even in the worst specimens of tubercular
lungs, numerous cicatrices and evidences of attempts to heal. These
attempts are more or less perfect, and when ineffectual, it is owing to the
circumstance that as one portion of lung cicatrises, another becomes the
seat of recent tubercle.

In Fig. ii, taken from the same source, the upper portion
of a right lung is shown, in which are calcareous masses occupy-
ing the place where formerly was active tuberculous disease.

As a rule, except in children, the top of the lung is first and
chiefly diseased. The explanations given of this fact are not
altogether satisfactory, but it is probable that the anatomical
distribution of the bronchial tubes gives the key to the problem.
The apical bronchi take a very steep direction upwards ; and
this implies that in expiration there is a dead point here, and
that in coughing a backward air current may easily drive
foreign matter into these relatively inactive regions. The
fact that in children apical phthisis is less common may be
due to the fact that in them the upper part of the lung is rela-
tively short and the apical bifurcation of bronchi less steep ;
but it is also explicable on the supposition that in children
invasion of the lungs from the lymphatic glands at their root
is more common than in adults.

How TUBERCULOSIS SPREADS IN THE LUNGS.— This occurs
often (a) through the air passages. When a cavity is formed and
its contents are being expectorated from any one point, it is
easy to understand how some of the semi-purulent expectoration
can be drawn into the tubes of healthy parts of the lungs. Here
it sets up caseating broncho-pneumonia, the lesion which pre-
dominates when animals are rendered artificially tuberculous
by the inhalation of tuberculous spray. Such cases in man
usually^progress rapidly. Disease also commonly spreads (b)

MORBID ANATOMY AND SYMPTOMS OF PHTHISIS 47

by infection of the lymphatics. Phagocytes ingest tubercle bacilli
from the yellow tubercles, and then pass on into the neighbouring
lymphatic vessels. In these vessels or in the glands fed by them
such phagocytes as perish release the contained tubercle bacilli,
and thus infect neighbouring parts. Hence around a caseous
mass are often seen more recent grey and yellow tubercles. The
lymphatic glands at the root of the lung are also involved early.
(c) If the infective material gains access to the blood vessels, as
when a tuberculous growth erodes the coat of a vessel, bacteria
are disseminated by the circulation of blood either to other parts
of the same lung or throughout the body, producing general
tuberculosis.

SYMPTOMS OF PHTHISIS. — From the preceding description of
the lesions found in fatal cases of phthisis the symptoms of the
fully established disease may be gathered. An irritating cough,
accompanied by abundant expectoration of muco - purulent
material, in which tubercle bacilli can usually be found ; hectic
fever ; copious cold sweats at night ; and rapid emaciation.

The symptoms of onset are commonly very insidious. The
patient is languid, suffers from increasing weakness, and is often
thought to be suffering from " anaemia." Anaemia with a dry
cough in most instances means early phthisis. Sometimes
profuse haemoptysis is the earliest symptom recognised, and it
is often the first symptom which induces a patient to consult
a doctor. This symptom always means that an already formed
tubercle, usually a caseous mass, has ulcerated into a blood
vessel, and indicates therefore older tuberculous disease.

At certain stages of phthisis there may be no expectoration,
and this does not always imply that active mischief is in abey-
ance. Cases with expectoration are described by German
doctors as " open," those without as " closed" ; the distinction
is important, as the latter are relatively non-infective. Pro-
gressive cases all become " open " sooner or later.

VARIETIES OF PHTHISIS. — The great majority of cases belong
to the chronic variety. Some are very acute, the whole case
only lasting from a few weeks to three or four months. Such
cases often resemble pneumonia, and some are so acute as to
simulate enteric fever. Of the chronic form of disease, some
show progressive deterioration, ending fatally in from six to
twelve months; others have repeated acute attacks, with in-

48 THE PREVENTION OF TUBERCULOSIS

tervals of apparent recovery and quiescence, the intervals be-
coming shorter as time progresses ; in others a sharp attack
occurs, and the patient then permanently recovers. To these
must be added a large number of unrecognised cases, in which
recovery occurs, and in which it is difficult to obtain any history
of lung disease. The patient may have been " off colour " for a
time, may have been anaemic, and may have had a slight cough.
He then " recovers by encapsulation, unaware that the shadow
of the black hawk's wing had rested upon him " (Allbutt, p. 1152).
Further particulars of such cases are given, pp. 82 to 84.

THE CURABILITY OF PHTHISIS. — The vast majority of
attacks of phthisis are followed by recovery. This fact cannot
be too strongly emphasised. It is not a new fact discovered
since the open-air treatment of the disease came into vogue, but
has been known to pathologists and physicians from time im-
memorial. Hippocrates taught that " phthisis, if treated early
enough, gets well." In modern times Cars well (quoted by
Brouardel, p. 66) wrote in 1838 : " Pathological anatomy has
never, perhaps, given a more decided proof of the cure of a disease
than it gives in cases of pulmonary phthisis."

Hughes Bennett (p. 716) says : —

In 1845, I made a series of observations with reference to the cretaceous
masses and puckerings so frequently observed at the apices of the lungs
in persons advanced in life. The conclusion arrived at was, that the
spontaneous arrestment of tubercle in its early stage occurred in the pro-
portion of from one- third to one-half of all the individuals who die after
the age of forty. The observations of Rogee and Boudet, made at the
Salpetriere Hospital in Paris, amongst individuals generally above the age
of seventy, showed the proportion in such persons to be respectively one-
half and four-fifths.

According to Charcot, " phthisis is susceptible of cure com-
pletely and definitely even at the period of cavities." Brouardel
quotes Laennec, Nat. Guillot, and Letulle as showing that in
more than half the post-mortem examinations made by them
old healed tuberculous lesions were to be found.

Commenting on these results Dr. Ribard says : —

These figures, from the similarity even of their results, are striking.
They show very clearly that half the men, said to be well and non-tuber-
culous, dying of old age or fortuitous causes, have at a certain time in
their life been attacked by tuberculosis but have recovered. Many are

MORBID ANATOMY AND SYMPTOMS OF PHTHISIS 49

therefore affected, and many recover, if half the human race have tubercle
and go on living without discovering them. Such is the truly reassuring
result of autopsies.

Dr. Thomas Harris of Manchester (1889) taking the deaths of
persons over 20 years of age who died in the Manchester Royal
Infirmary found healed phthisis (" involuted tuberculosis ") in
about 38 per cent, of the post-mortem examinations made by him.

Coates (1891, p. 351), after giving an account of 131 consecutive
autopsies at the Glasgow Royal Infirmary, says : "It appears
that, taking even the most serious forms of internal tuberculosis,
such as consolidation of lungs, tuberculous disease of the verte-
brae, tuberculosis of the peritoneum, there is evidence that
spontaneous recovery takes place in a proportion equal to that
in which death occurs."

Austin Flint (1882), after analysing 670 cases of phthisis
in his practice, concluded that "in a certain proportion of
cases this disease ends favourably irrespectively of any ap-
preciable extrinsic agencies." He draws attention as follows to
the self -limitation which is exemplified in the majority of fatal
cases (p. 617) : —

The disease, as a rule, advances not by a continuous progress, but by
a series of successive invasions separated by variable intervals. After
each invasion, or as it has been termed tuberculous eruption, there is a
temporary self-limitation of the disease.

The continuous advancement of the disease as an exception to the
rule is the pathological feature of the so-called " galloping consumption "
or phthisis florida.

DURATION OF PHTHISIS. — From the preceding pages it is
evident that the duration of phthisis is very variable. It is
interesting to note the estimates of its duration given by different
authors. According to Austin Flint it may vary from three
weeks to forty years. Similarly Portal said " eleven days to
forty years." Laennec gave its average duration excluding
miliary tuberculosis as 24 months, Louis and Boyle on the
strength of 314 cases said 23 months, Audral 24 months, Sir J.
.Clark (from patients in private practice) 36 months. C. J. B.
Williams and C. T. Williams (Quain, 1894) give an average
duration in 198 fatal cases of 7! years, and in 802 living cases of
8J years. All these cases had been over a year under observa-
tion, which necessarily excludes some acute cases ; but with
4

50 THE PREVENTION OF TUBERCULOSIS

this exception they state that these figures " may be taken as a
correct average for the duration of the disease among the upper
classes under modern treatment, especially as 72 per cent, of the
living had recovered sufficiently to pursue their usual avocations,
and many among them had already lived upwards of 20 years
since their first attack."

Walshe (1871) gives the average duration for hospital cases
in Paris as 23*5 months. He speaks of a case lasting 22 years,
and of cases frequently lasting from 5 to 10 years. Dettweiler
gives the average duration of life of the middle-class consumptive
as 7 years, but Cornet (p. 250) says that the average duration in
adults cannot be -placed higher than 3 years, and in children
even less. He also quotes Leudet's data, which comprise 48
cases, among whom the average duration was 5 years for those
in good circumstances, 3^- years for those in hospitals. All the
figures show a shorter duration among the poor than among the
well-to-do.

If the; average duration of phthisis could be worked out
separately for patients whose illness started at different ages,
some light would probably be thrown on the varying estimates
given above. As a general rule, it is a more acute illness in the
young, and becomes more chronic with advancing years, though
there are many exceptions to this rule. A further point doubt-
less has affected the estimates of its duration quoted above. It
is well known that in the less acute cases the course of the disease
is not uninterrupted. There are attacks of " bad colds," of
" influenza," or of pleurisy, or of actually recognised phthisis,
and then occur intervals in which all symptoms are in abeyance ;
these intervals shortening if the case progresses. The intervals
may sometimes extend over many years. Is the duration of
such cases to be reckoned from the first occurrence of recognis-
able symptoms to the end of the case ? If so, many months and
> even years in which the patient is apparently well will be in-
cluded. Until these points are settled, statements as to average
duration of phthisis should only be accepted when accompanied
by information as to the intervals during which symptoms were
in abeyance.

CHAPTER VI
THE TUBERCLE BACILLUS

THE tubercle bacillus (or bacillus of tuberculosis) is a non-
motile organism, rod-like in shape, with rounded ends.
Its length is from 2 to 5 ^ (^ = one-thousandth of a milli-
metre), that is, from one-half to one-third the diameter of a
red blood corpuscle, whilst its width is about one-sixth of its
length. When stained with aniline dyes the bacilli often show
a beaded appearance, which Koch regarded as indicating the
presence of spores ; but this point is doubtful. We have already
seen (p. 41) that Koch succeeded in staining the bacilli after
long soaking of cover-slip preparations in alkaline methylene-
blue and then using vesuvin as a brown contrast stain. Ehrlich
soon made known a more certain and more convenient procedure.
He first stained for fifteen to twenty minutes with an aqueous
solution of aniline methyl violet or fuchsin, and then decolorised
with dilute nitric acid, which eliminated the colour from everything
except the tubercle bacilli. Other methods have been since
devised, of which the following is the most convenient, especially
for the examination of suspected sputum.

THE ZIEHL-NIELSEN METHOD OF STAINING. — A small solid
bit of sputum is taken, spread on a clean cover-glass and allowed
to become dry. The cover-slip, held in a forceps, is then passed
three times through the flame of a spirit lamp, holding the sputum-
spread side uppermost. This fixes the film. A watch-glass is
partially filled with a solution composed of fuchsin i part, absolute
alcohol 10 parts, and carbolic acid (5 per cent, aqueous solution)
loo parts. The cover-slip is placed film downwards on this
solution, which is heated until it steams slightly. The cover-slip
after three to five minutes is removed, the excess of dye washed
off with water, and the slip then dipped in a i in 4 solution of
sulphuric acid. As soon as all visible colour has disappeared from
the film, it is rinsed with several portions of a 60 to 70 per cent.

52 THE PREVENTION OF TUBERCULOSIS

alcohol, and finally with water. The film is then counter-stained
with a i per cent, aqueous solution of methylene-blue. On
miscroscopic examination the specimen thus prepared shows the
red bacilli on a blue background.

The above staining reaction is almost specific for the tubercle
bacillus, since the leprosy bacillus and the few others which act
somewhat similarly in resisting the decolorising effect of acids
are very rarely found under circumstances in which confusion
would be likely to arise. When bacilli in human expectoration
answer to the above test it is practically certain that the ex-
pectoration is derived from a tuberculous patient.

It must be remembered that negative results from single
examinations of suspected sputum carry little weight. Three
specimens at least should be mounted from each sputum, and in
each of these a large field, spread over the slide in preference
to the cover-slip, should be examined before a negative certificate
is given.

BIOLOGY OF THE TUBERCLE BACILLUS.— For more complete
study of the biological relations of the tubercle bacillus it is
necessary to cultivate it on or in artificial media in the laboratory.
Koch ascertained that it would not grow on the ordinary labora-
tory media, gelatine, agar, etc., because these did not remain
unaltered at the body temperature. He finally hit on coagulated
blood serum as a suitable medium, because it remained solid
and moist at the body temperature. Having obtained tuber-
culous material from newly killed animals suffering from recent
tuberculosis, he successfully grew tubercle bacilli by rubbing
this material thoroughly on the blood serum by means of a
platinum loop, and then placing in an incubator at blood heat
(37° C.)« After the fifth day dull white specks appeared on the
surface of the serum, and these gradually increased in size, pro-
ducing small dry scales, which subsequently became confluent,
forming a greyish-white covering to the serum, the latter not
being penetrated or liquefied. He subsequently succeeded in
obtaining similar growths from the cavities of tuberculous lungs,
from lupus, etc. From observations on such cultures, and on
cultures in glycerine bouillon agar, have been deduced certain
facts as to the persistence of the life of the tubercle bacillus
which have important bearings on the prevention of tuberculosis.
RANGE OF TEMPERATURE. — The tubercle bacillus of mam-

THE TUBERCLE BACILLUS 53

malian tuberculosis ceases to grow below 29° C. and over 42° C.,
of avian tuberculosis below 25° C. and over 45° C. The best
temperature for the growth of the mammalian tubercle bacillus
is 37°-38° C. As these temperatures are not common in the
external world, it is important to note that, as Cornet (p. 42)
remarks, the tubercle bacillus does not meet with the conditions
of growth " except solely and exclusively within the animal
organism with its constant and equable temperature of 37°-39° C."
Or as Dr. Moxon (1885) put it : " The life of the bacillar parasite
is difficult, easily discouraged by unfavourable circumstance,
like an aphis by an eastern wind." Beevor, Delepine, and
Kanthack have succeeded in obtaining growths of the tubercle
bacillus on potato at room temperature ; but this is difficult,
and there is no evidence that it occurs frequently. Extreme
cold does not kill the bacillus. There is considerable discrepancy
in the evidence as to the thermal death-point of the tubercle
bacillus. Probably different strains of bacilli vary in this respect,
and much will depend on the medium surrounding them. Further
details on this point will be found on page 409. Generally the
tubercle bacillus is destroyed after 4 to 6 hours' exposure to a
temperature of 55° C. ; after 15 minutes at 65° C. ; after 5
minutes at 80° C. ; after 2 minutes at 90° C. ; and in a less time
at the temperature of boiling water. In a dried condition its
vitality may survive higher temperatures than the above.

The RESISTANCE TO DESICCATION shown by the tubercle
bacillus is its most significant biological feature. It appears to
owe this resistance to the fact that it contains more fat than
other bacilli. Koch found that phthisical expectoration which had
been allowed to dry and been kept at room temperature for five
to eight weeks was still virulent at the end of the time. Schill
and Fischer found dried expectoration still virulent on the
95th day, dead on the lygth day. Toma found dried expectora-
tion virulent up to ten months (Cornet, p. 43). The duration of
vitality is much less when the tubercle bacilli are exposed to
SUNLIGHT. Koch found that in direct sunlight they died after
an exposure varying from a few minutes to several hours, accord-
ing to the thickness of the layer exposed. Diffuse light has the
same effect after an appreciably longer time. Strauss found
that flourishing cultures of mammalian tubercle bacilli perished
completely on exposure for two hours to the rays of the summer

54 THE PREVENTION OF TUBERCULOSIS

sun, while cultures dried in thin smears on glass plates had lost
their virulence under similar conditions in half an hour. More
recently Mitchell and Crouch (quoted by Lartigau, p. 29) from a
study of the influence of sunlight on tuberculous expectoration at
Denver concluded that the tubercle bacillus as expectorated on
a sandy soil is still virulent after thirty-five hours' exposure to
the direct rays of the sun, the virulence becoming lost soon
afterwards.

Where there is no free access of air or sunlight the retention
of virulence in deposited tubercle bacilli has been observed at
the end of 130 days by Ransome and of 184 days by Fischer.
It may be added that Cadeac and Malet have produced positive
results by inoculation of material from tuberculous lungs which
had previously been buried for 167 days.

It must be noted that the fact that a tubercle bacillus takes and
retains the specific stain, does not prove it to be alive. A bacillus
heated to the temperature of boiling water will take the stain
equally well. This remark is important in view of the enormous
numbers of tubercle bacilli daily expectorated by consumptives
(p. 104). It is probable that the majority of them are non-
virulent, though in phthisis generally the infectivity probably
is proportional to the total number of bacilli discharged. The
infectivity although great must not be exaggerated. The
tubercle bacillus grows with exceptional slowness both inside
and outside the body. It has a feeble vitality under both con-
ditions, and is easily rebuffed. The one circumstance under
which the extra-corporeal life of the bacillus is prolonged is
desiccation in places not exposed to sunshine. Such dry ex-
pectoration will contain numerous living bacilli.

CHAPTER VII

INFECTIVITY OF TUBERCULOSIS: A. HISTORY OF
VIEWS HELD

THE belief in the infectivity of phthisis is as old as any
extant account of the disease. Hippocrates said that
it was of all diseases the most dangerous, and fatal to
the greatest number of mankind. Galen believes it to be
dangerous to pass a single day in the company of a consumptive.
Avicenna the Arabian (A.D. 1037) referred to diseases which are
" taken from man to man like phthisis." Ballonius, a physician
of large practice in Paris in the fifteenth century, noted the fre-
quent occurrence of phthisis in those who tended consumptives.
Both Morgagni and his teacher Valsalva (seventeenth century)
asserted that they objected to conduct autopsies on con-
sumptives on account of the danger of infection. In Italy
the belief in the infectiousness of phthisis took practical form
in legislative enactments. In 1746, Ferdinand vi. issued to
the medical men in charge of the various districts an instruction
which ran as follows : —

Experience having shown how dangerous is the use of linen, furniture,
and articles which have been used by persons afflicted with, or who have
died of hectic, phthisical, or other contagious diseases, we enjoin on all
physicians to give notice of those persons who are sick with or who have
died of phthisis, so that the Alcade may cause the linen, clothing, furniture,
and other objects used personally by the patient, or which have been in
his department, to be burned ; so that the Alcade may also order the
apartment in which the patient died to be replastered and whitewashed,
and the flooring or flagging of the room or alcove in which the patient's
bed was placed to be changed. Besides, a registration must be kept of
places from which clothing found in the shops of second-hand clothes
dealers comes, with information as to the names and residences of the
vendors, as well as the persons who have used the linen and garments,
and dealers in old clothes ordinarily doing business in infected clothes.
The Alcade shall issue a paper attesting that the said goods are free from
contagion; this paper shall be the sole authorisation by which dealers

56 THE PREVENTION OF TUBERCULOSIS

in second-hand goods will be allowed to keep or sell such goods. Any
physician who will not give notice of consumptive patients, or those who
have died of consumption, to the Alcade of his quarter, shall incur, for
the first offence, a fine of 200 ducats and suspension from the practice
of his profession for one year ; and for repetition of the offence a fine
of 400 ducats and the punishment of exile for four years. All other
persons (infirmarians, domestics, attendants on the sick) who will not
report the case shall incur a penalty of thirty days in prison for the first
offence, and four years in the galleys for the second offence. Civil, re-
ligious, and military authorities shall cause to be burned in civil and
military hospitals all linen which shall have been used by phthisical
civilians or soldiers.

In 1754 the members of the College of Physicians of Florence
pronounced themselves as on the whole favouring the conclusion
that phthisis is communicable. In 1782 the city of Naples,
warned of the infectivity of phthisis by the Medical College
of its University, enforced a law for the isolation of consumptives
and the disinfection of their homes and belongings.

Nor were such views confined to Italy. In a letter to the
Lancet by Dr. Stretton (December 17, 1898), the following
quotation is given from a book written by Gideon Harvey, M.D.,
about 1660, in which consumption is described as an endemic
and epidemic disease : —

And considering withal its malignity and contagious nature, it may
be numbered among the worst Epidemicks or popular diseases, since next
to the Plague, Pox, and Leprosie, it yields to none in point of contagion ;
for it's no rare observation here in England, to see a fresh coloured lusty
young man yoake to a consumptive female, and him soon after attending
her to the grave. Moreover nothing we find taints sound lungs sooner, than
inspiring or drawing in the breath of putrid ulcered consumptive lungs ;
many having fallen into consumptions, only by smelling the breath or
spittle of Consumptives, others by drinking after them ; and what is
more, by wearing the Cloaths of Consumptives, though two years after
they were left off.

In The Expedition of Humphry Clinker, written by Smollett
in 1771, the same notion of infectiousness finds laughable ex-
pression. Writing from a fashionable inland health-resort,
he says : —

I wish I had not come from Bramble tonhall, after having lived in
solitude so long. I cannot bear the hurry and impertinence of the multi-
tude ; besides, everything is sophisticated in these crowded places.
Snares are laid for our lives in everything we eat or drink ; the very air
we breathe is loaded with contagion. We cannot even sleep, without

INFECTIVITY OF TUBERCULOSIS 57

risk of infection. I say, infection. This place is the rendezvous of the
diseased. You won't deny, that many diseases are infectious ; even the
consumption itself is highly infectious. When a person dies of it in
Italy, the bed and bedding are destroyed ; the other furniture is exposed
to the weather, and the apartment whitewashed before it is occi pied
by any other living soul. You'll allow, that nothing receives infection

sooner, or retains it longer, than blankets, feather-beds, and mattresses

'Sdeath ! how do I know what miserable objects have been stewing in
the bed where I now lie ? I wonder, Dick, you did not put me in mind
of sending for my own mattresses ; but, if I had not been an ass, I should
not have needed a remembrancer. There is always some plaguy reflec-
tion that rises up in judgment against me, and ruffles my spirits; there-
fore let us change the subject.

The experience of George Sand is also interesting. In 1839
she wrote from Spain as follows concerning Chopin, her
travelling companion, who was already consumptive, although
he did not die until ten years later : —

Poor Chopin, who had a cough since leaving Paris, became very ill.
I called in a doctor — two doctors — three doctors, each more stupid than
the other, and soon it was spread abroad that he was in the last stage
of consumption. There was great alarm, phthisis being rare in these
climates, and regarded as contagious. We were regarded as pest-breeders ;
and furthermore as heathens, as we did not go to Mass. The owner of
the small house which we had rented turned us brutally out of doors,
threatening furthermore to bring an action against us compelling us to
limewash his house, which he said we had infected. We were plucked by
the law like chickens.

At Barcelona later on the landlord demanded to be paid for
the bed on which Chopin had slept.

Medical men gradually tended towards the opinion that
tuberculosis was non-infectious, and began to explain it as a
manifestation of a special constitution or diathesis, while public
opinion in many countries still regarded it as infectious, this
belief being carried in some instances to foolish extremes. The
histories of cholera and influenza present similar anomalies.
Thus the Royal College of Physicians of London in 1854 reported
that " the theory that cholera is propagated and diffused by
means of human intercourse, receives no support from the facts
relating to variations in the intensity of cholera epidemics,
and the circumstances determining these variations." In
another part of their report they quoted the extraordinary
rapidity of the increase of cholera in a town as "an additional
reason for believing that the diffusion of cholera in a town is

58 THE PREVENTION OF TUBERCULOSIS

independent of contagion/' In the same report they record
their impression that " the share borne by human intercourse
in the dissemination of the disease is larger " than the statis-
tical facts seem to indicate. A joint inquiry was made by the
Provincial Medical Association of England into the contagious-
ness of influenza in the epidemic of 1836-37, the medical answers
to the questions on this point being "of an almost uniform
tenour, the • opinion of nearly all those who had the most ex-
tensive opportunities of investigating the disease, and the best
means of arriving at a definite conclusion, being that there is
no proof of the existence of any contagious principle by which
it was propagated from one individual to another."

And yet more exact information and more accurate medical
investigations have proved that infection is the sole means
for the spread of these two diseases. Tuberculosis differs from
them in infectivity chiefly in its longer latency and more pro-
tracted course.

CHAPTER VIII

INFECTIVITY OF TUBERCULOSIS : B. EXPERIMENTAL

EVIDENCE

INFECTION BY INOCULATION. — Villemin's experiments
(p. 38) gave the first positive evidence of infectivity.
Previous conclusions to this effect were of the nature of
surmises, and naturally liable to exaggeration and misconception.
Villemin's experiments undoubtedly did much to popularise
the idea that tuberculosis is an infectious disease. Koch's
experiments demonstrated this fact, and placed Villemin's
induction on a solid foundation.

Koch's experiments may be briefly summarised, as they
illustrate admirably the process used to prove the causal
relation between a given microbe and the specific disease caused
by it.

(i) He took as seed material the tuberculous lymphatic
glands from freshly killed guinea-pigs which had been inoculated
about three or four weeks previously with tuberculous material.
(2) This material was smeared on blood serum and incubated
at 37° C. until a sufficient growth of tubercle bacilli had slowly
occurred. (3) From this test-tube cultivation other tubes of
blood serum were similarly smeared, by rubbing some of the
small scales from the first tube over the serum in them. Koch
cultivated the tubercle bacilli in test tubes in this way through
as many as seventy generations. (4) The inoculation of guinea-
pigs and other susceptible animals with such cultures was
followed by the appearance of tuberculous nodules and other
lesions identical with those found in the animals which pro-
duced the original tuberculous material. (5) Tubercle bacilli
were found in these experimentally produced lesions as in the
original lesions of the first animals, and these tubercle bacilli
showed the same cultural characters, and when inoculated
into animals produced similar lesions to those of the original

60 THE PREVENTION OF TUBERCULOSIS

disease. Similar experiments made with tuberculous expectora-
tion from human consumptives, and with tuberculous meat
and milk, gave the same results. The proof is rendered complete
by the further fact that tubercle bacilli are not found in any
diseased conditions other than tuberculosis.

This is a convenient point to revert to the instances in
which apparently tubercles had been experimentally produced
by non-specific inoculation (p. 39). Klebs suggested that ex-
traneous infection was the cause of these anomalous results,
and Frankel and Cohnheim showed that this was the correct
explanation. Watson Cheyne (1883) proved the same thing
by a series of carefully checked experiments on rodents.
Wilson Fox in 1867-68 had apparently produced tuberculosis in
twenty-three out of 117 animals inoculated with such materials as
pus, putrid muscle, seton, etc., which were supposed to be non-
tuberculous. At his suggestion the experiments were repeated
some years later by Dawson Williams, under conditions which
prevented the occurrence of external infection, and in each case
a negative result was now obtained.

The evidence that tuberculosis is infective is not confined
to experimental inoculation. Were it so, it might still be reason-
ably contended that tuberculosis is only communicable like
tetanus or hydrophobia by introduction of the infective material
(contagium) under the skin. Experimental observations, how-
ever, have proved that it can be spread either by the inhalation
or the ingestion (swallowing) of tuberculous material.

INFECTION BY INHALATION. — Tuberculosis has frequently
been induced in guinea-pigs by making them breathe in an
atmosphere containing dust contaminated by tubercle bacilli.
The lungs in such animals become tuberculous in two or three
weeks, the extent of the lesions depending on the duration of
life before the animal is killed or dies. The lungs present the
same appearances of caseous pneumonia as do the lungs of man
in ordinary phthisis. The liver and spleen of the infected
animals also become tuberculous, and the bronchial glands
appear to be affected as soon as the lungs.

INFECTION BY INGESTION. — Experimental tuberculosis of
the intestine has been produced in guinea-pigs, rabbits, dogs,
cats, calves, sheep, monkeys, etc., by feeding them with tuber-
culous material. Pigs are readily susceptible to such infection,

INFECTIVITY OF TUBERCULOSIS 61

and frequently become infected through being fed on skimmed
milk derived from tuberculous cows. In these cases the small
lymphatic follicles in the wall of the intestine are commonly
infected first, followed about four weeks later by the mesenteric
and caecal glands. Out of twenty animals examined after experi-
mental feeding with tuberculous material Sidney Martin found
the small intestine to be involved in all but one, and the
caecum in all but three. Intestinal lesions may be absent,
especially when the dose of infection is small ; and in this
case the first lesions are in the lymphatic glands. From the
mesenteric and caecal glands the infection passes to the cceliac
glands, the liver and spleen, the bronchial and posterior medias-
tinal glands, and the lungs. Baumgarten, Fisher, and others
have shown that tubercle bacilli can pass through the mucous
membrane of the intestine without producing any local ulcer.

CHAPTER

INFECTIVITY OF TUBERCULOSIS : C. STATISTICAL
AND CLINICAL EVIDENCE

ON the strength of the statements given on pp. 8 and 49, it
has been assumed by some that phthisis is so common
and so often a non-fatal disease that everyone is more
or less exposed to infection, and that consequently infection
can play only a very minor part in its causation. The evidence
as to the percentage of the total population (say roughly one in
every two) showing evidence of old tuberculous lesions is derived
from hospital practice. Persons belonging to this type possibly
form a majority of the total population, and although the pro-
portion probably is smaller in other grades of life, we may assume
for present purposes that the same proportion holds good for
the general population in England and Wales. But it by no
means follows that one-half of the total population at any
given time is actively tuberculous and discharging tuberculous
material. The fact that recovery has occurred and the patients
have died from other diseases or from accident, shows the absurd-
ity of such an assumption. It is highly probable that the vast
majority of those showing post-mortem these healed lesions
were "closed" cases, in which the micro-organisms could not
escape ; so that the patients were not infective even during a
few months of their life. Further light is thrown on the point
by a comparison between the deaths from phthisis and the popu-
lation at each five- or ten-yearly period of life. In Table XVIII.
the deaths from phthisis have been multiplied by three, on the
commonly accepted supposition that for each death from phthisis
during a given year, three other patients have been constantly
ill with the same disease. On this basis it will be seen that the
proportion of consumptives in the general population is i for
every 263 persons, varying from i in 1881 at ages 5-10 to i in 141

at ages 35-45. At the working years of life, 20-65, on the same

INFECTIVITY OF TUBERCULOSIS

assumption it is i in 168. Probably the number actually
phthisical at any given time exceeds this proportion ; but it is
equally probable that the number at any given time capable of
imparting infection is not greater than these figures would

TABLE XVIII

On the Assumption that

Population

Deaths from

each Annual Death from

of England

Pulmonary

Phthisis means the Presence

At Ages—

and Wales
at the

Phthisis
in England

of three Consumptives in
the Population, the

Census

and Wales

Proportion of Consumptives

1901.

in 1901.

in the General Population

at each Age-period was —

Under 5

3,716,708

1,171

in 1006

5-10

3,487,291

623

1881

10-15

3,341,740

987

1129

15-20

3,246,143

2,917

371

20-25

3,120,922

4,590

227

25-35

5,255,840

9,922

177

35-45

3,996,005

9,451

141

45-55

2,902,191

6,653

145

55-65

1,943,250

3,459

187

65-7 «

1,076,006

1,260

285

75 and upwards

441,747

193

763

Total — All Ages

32,527,843

41,226

I in 263

NOTE. — The above proportions are based on an average duration of
three years for each case of phthisis. On page 360 I have assumed an
average duration of ten years, which would include also a large number
of cases that are never fatal. These estimates must be carefully distin-
guished from the estimated numbers discussed on page 15, which are
concerned with ascertaining in a life- table population traced to death,
how many total consumptives there are.

lead one to suppose. On this point the considerations detailed
in Chapter XIII., and particularly on page 101 need to be borne
in mind.

CLINICAL EVIDENCE OF INFECTIVITY. — Underlying all in-
vestigations of the history of individual cases for evidence of
infectivity are certain fundamental data, which may conveniently
be summarised here : —

1. Tuberculosis is due to a specific bacillus.

2. Tuberculosis has been produced experimentally in animals
by the introduction of this bacillus, in inspired air or with food.

THE PREVENTION OF TUBERCULOSIS

3. Man is subject, e.g. in infected households and workshops,
to the conditions which have been proved experimentally to
produce tuberculosis in animals.

It is in the light of these general considerations that the
following instances of probable infection are to be judged.
They are given as typical of the form of reasoning which in the
light of wider investigations is now known to be applicable to
such cases, and of the kind of evidence which without such wider
investigation could not be regarded as possessing great weight.

All the following cases have been taken from local investiga-
tions of notified cases.

CASE i. — Domestic infection. Father to son

C. P., set. 25, admitted to sanatorium October 5, died
November 22, 1906, of acute phthisis. Was unmarried, and
lived with his parents up to the time of his illness. His mother,
two sisters, and a brother are alive and well. No tuberculosis
known in the family except the father.

Domestic Influences.

Age.

Year.

Extra-domestic Influences.

1881

1895

Worked as a labourer, "odd

jobs," up to 19.

Probably father was ill from this

1900

Has worked as a general lab-

date.

ourer, generally in the shops

of the railway works.

In June 1903, father died, set. 41,

1903

death being returned as due

to "pulmonary and laryngeal
tuberculosis, 15 months."

C. P. says he had no cough until

1906

a few weeks before admission

to sanatorium.

Comments on Case i. — There was protracted infection from
the father ; also possible industrial infection, but this would
be only casual. A latent period of at least three years occurred
between his father's death and his first symptoms.

CASE 2. — Domestic infection. Father to son, or brothers to brother.
Action of auxiliary influences

W. O., admitted to sanatorium September 18, discharged

INFECTIVITY OF TUBERCULOSIS 65

October 25, 1906 ; had advanced tuberculosis both lungs. Had
cough for four years before admission. Four sisters and one
brother have escaped tuberculosis.

Domestic Influences.

Age.

Year.

Extra-domestic Influences.

1871

Mother died of phthisis

1884

Brother died of phthisis, set. 22,

M
19

1885
1890

Apprenticed as a gasfitter.

in Brompton Hospital ; had

previously lived at home.

Another brother died of phthisis,

1892

set. 27, at W. Until a few

months previously had lived

with present patient. Domestic

infection ceased in 1892.

28-31

1899-1902

Served in the Boer War. Had

enteric fever. Began to

Lived in lodgings after returning
from South Africa.

1904

cough while in South Africa.
Had pleurisy , and was aspir-
ated.

33-35

1904-06

Gasfitter.

Comments on Case 2. — In my opinion the protracted domestic
infection which ceased nine to ten years before he began to cough
caused this patient's tuberculosis, the sickness, exposure, and
privations of the Boer War serving to light up latent trouble.
The alternative is that more recent infection in South Africa
caused his illness.

The same question of domestic or industrial infection is raised
in Case 3.

CASE 3. — Domestic infection from brothers and sisters

A. G., aet. 32, admitted to sanatorium August i, discharged
September 12, 1906. The main facts are set forth in the scheme
on next page.

Comments on Case 3. — The patient's father and mother are
alive and well, and there is no family history of phthisis in past
generations or among uncles or aunts. The patient has been
exposed to home infection from childhood until he was 15 years
old. His first symptoms of phthisis occurred ten years later.
Several possibilities of casual extra-domestic infection present
5

THE PREVENTION OF TUBERCULOSIS

themselves — (i) when a railway shunter ; (2) when a black-
smith. His work at a music hall was after frequent cough had

Domestic Influences.

Age.

Year.

Extra-domestic Influences.

1874

Excepting the years 1897-1902,

has lived at home and been ex-

posed to the following chances

of acquiring tuberculosis : —

One brother, set. 19, died at home

1879

of phthisis.

One sister, set. 21, died at home of

1881

phthisis.

One sister, set. 5, died at home of

1882

" congestion of lungs."

1887

Left school.

1889

I In an auctioneer's office.

1891

A railway porter in E. (shunting

and lamps— did not clean out

carriages).

A sister, set. 21, died in an asylum

i893

of phthisis, 3 years after leaving

Has worked during these ten

riomc*
Patient left home and went into
lodgings for 5 years.

1897

years as a striker and black-
smith in the railway works.

Began to have a slight cough from

190x3

this time.

Patient returned home. No known

1902

domestic infection from 1890

(the date sister left for an

asylum) up to 1906.

A brother, set. 42 (married, with

1904

4 children), attended Brompton

Worked as an attendant at a

Hospital for a few months with

music hall.

one lung affected. No chance

Is somewhat alcoholic.

of infection between the two

brothers.

In July, severe hemoptysis .

1906

occurred, and probably the same remark applies to his alcoholic
habits.

CASE 4. — Protracted domestic injection from parents and brothers

and sisters

Florence S., aet. 24, admitted to sanatorium September 3,
discharged November 24, 1906 ; early phthisis, with tuberculous
cervical glands.

nfluences.

Age. 1 Year.

Extra-domestic Influences.

ed to domestic

1882

ably from early

died, set. 45, of

1890

After the father's death in 1890,

the mother began a small

laundry, and the patient and

her two sisters have helped

in it. The patient is chiefly

T (• rllaA rvf

T T

TCrt-

engaged at needlework.

INFECTIVITY OF TUBERCULOSIS

The main facts are set forth below.

Has been exposed

infection p

childhood.
Father, a wa

phthisis.

A brother, set.

phthisis.

j A sister died of phthisis
Mother died, aet. 48, of phthisis .
First noticed enlarged cervical

glands. Axillary glands soon

afterwards inflamed and sup-

pttratedfor two years.

A brother died, aet. 28, of phthisis .

A brother, set. 34, died of
phthisis.

A sister, then aged 27, was
notified as phthisis in June
1903. Tub. bac. present. Was
in sanatorium Aug. -Sept. 1903.
Is now (Dec. 1906) quite well.

Cmigh developed a few weeks
before admission to sana-
torium. Tttb. bac. found.
Cervical glands still large
and indurated, axillary glands
the same.

12
16

21
21

1894

I899
1903

1903

24 j 1906

Comments on Case 4. — The patient can scarcely be said to have
been free from the possibility of infection during her whole life.
She showed tuberculous glands at the age of 16, and signs of
pulmonary disease eight years later.

CASE 5. — Doubtful whether domestic or industrial infection

operative

Clara R., aet. 29, admitted to sanatorium July 16, discharged
August n, 1906.

Comments on Case 5.— This is a good illustration of a large
number of cases in which several points are open to doubt. Was
the patient infected from her mother, the industrial conditions

THE PREVENTION OF TUBERCULOSIS

merely breaking down her resistance ? Were the " constant
colds " only bronchial attacks on which phthisis was eventually
engrafted by infection from some of the other work-girls ; or,
as is more likely, did she have phthisis from 18 years of age
onwards ? If the latter view is taken, two possible sources of

Domestic Influences.

Age.

Year.

Extra-domestic Influences.

1877

Mother died after "breaking a

1887

blood vessel."

Patient does not know if the mother

had a cough previously.

1892

Patient went as a dressmaker.

Worked with the firm A. for

2^ years, with a friend who

died in 1901 of pulmonary

tuberculosis, and was delicate

in 1892, but is doubtful if she

then had a cough.

Began to have " constant cold's."

1895

Patient went to firm B. for 3

years

Father died of "emphysema."

1898

Went to firm C. for 4 years.

1899

Workroom containing 12 girls

was overcrowded.

1901

Often visited the bedridden

consumptive friend mentioned

above.

1902

Went to firm D. for 2 years

Large workoom here.

At Christmas had a bad cough.

1904

In May 1905 was in bed 5 days,

1905

Went to a smaller dressmaker's

and since then always cough

place for 9 months ; room

and expectoration.

underground, stuffy and dusty.

Admitted to sanatorium 3 months

1906

Went to firm F.; large work-

after tub. bac. were found

room.

in sputum.

infection are still known, the dressmaker friend or her mother.
If the mother died of phthisis, I should lean to the view that she
was the probable source of infection, because domestic exposure
is generally more intimate and more protracted than occupa-
tional exposure to infection.

CASE 6. — Domestic infection from a non-relative. Influence of

industrial fatigue

Geo. S., set. 39, admitted to sanatorium with extensive tuber-
culosis both lungs, September 8, and discharged November 2,
1906. Father and mother, five brothers, and two sisters all

INFECTIVITY OF TUBERCULOSIS

healthy. No tuberculosis known in his or his wife's family
except that the latter's father died over twenty years ago of
this disease. The main facts of his illness are summarised below.

Domestic Influences.

None.

Married . . .

Three children living : two died
stillborn, one as shown below.

Has occasionally sublet part of
his house, but not, so far as he
or his wife knows, to people
with bad coughs except as
shown below.

Had a man named P. and his
family occupying part of the
house for about a year. P.
was then ill with phthisis, and
died at Easter 1903.

G. S.'s child died of "consump-
tion of the bowels " in August
1903. This child was born
November 1902, became ill
when 4 months old ; never
had diarrhoea.

Cough began late in this year

Age.

35-36

37
39

Year.

[867

1887

Extra-domestic Influences.

Has been a house-painter all
his working life.

1902-03 i

1904
1906

No extra-domestic or family source of infection could be
detected. The facts as to the P. family need to be stated in some
detail. G. S., his wife, and four children had two rooms for them-
selves and let off the rest of the house to P. and family, who lived
here for nine months. The two families were not very friendly,
but there was a common scullery and w.c. P. was very dirty in
his habits, and spat about. His spit-cups were often left in the
scullery. About six weeks after P/s death, patient and his
family left this house. It should be added that Mrs. P., her son
and two daughters were quite well in November 1906.

Comments on Case 6. — It seems likely that G. S. and his child
were both infected by P. The escape of the P. family, and of
the other members of the S. family, does not exclude this ;
similar experiences of escape are not uncommon in the acute
infectious diseases. G. S. while living in the same house as P.
was working very long hours, and it is likely that this made him
more open to infection.

THE PREVENTION OF TUBERCULOSIS

CASE 7. — Possible public-house infection

W. W., aet. 42, admitted to sanatorium August 23, discharged
October 4, 1906. No family history of phthisis.

Domestic Influences.

Age.

Year.

Extra-domestic Influences.

1864

No infection known.

1876

Began work as an errand boy.

Father died of asthma, set. 44

1882

Married .....

1887

^ Worked as a butcher's assistant.

[Daily frequented various public-

1 houses, and has been a free

J toper from this time onwards.

1894

\ Worked as an outside salesman

| at various butchers' shops.

1903

Had a bad cough at Christmas

1905

\ Worked in a baker's shop.

time, which got well again.

)

In June severe haemoptysis^ which

1906

recurred on four occasions.

Comments on Case 7. — There is no evidence of family or other
domestic infection, and none of industrial infection. The public-
house is the most likely source of infection.

CASE 8. — Possible occupational injection

W. H., aet. 33, admitted to sanatorium October 3, discharged
October 31, 1906. There is no family history of tuberculosis.
Father and mother, three brothers, and three sisters all alive and
well. Married for seven years ; two children, both well.

Domestic Influences.

Age.

Year.

Extra-domestic Influences.

No infection known.

19
26

1873
1892
l8qq

No definite infection known.
A soldier from 1892-1904, in
India and South Africa.

' Has had a cough for a year, and
expectoration for about 6 months
before admission to sanatorium.
Never pleurisy or blood- spitting.

3i
33

1904
1906

On returning from South Africa
was engaged as a cleaner in
the P.O. One of chief duties
is to sweep out the rooms.

INFECTIVITY OF TUBERCULOSIS 71

Comments on Case 8. — The patient's present occupation —
sweeping out public offices — is a possible source of infection,
but he may have been infected while a soldier or elsewhere.
The case is one of a class in which a probable statement of
infection is impracticable.

GENERAL CONSIDERATIONS ON THE STATISTICAL STUDY OF
HISTORIES OF INFECTION IN PHTHISIS. — The preceding cases
illustrate the types of history often obtained in investigating
cases of phthisis. The difficulties in tracing the source of infec-
tion in a given case are much greater than in the acute infectious
diseases. There is an extremely variable period of latency, and
the symptoms of the initial stages of the disease may pass unre-
cognised. The study of latency is so important a part of the
problem that the next chapter is devoted to it ; and all histories
of infection should be viewed in the light of the facts there set
out. In view of the great prevalence of the disease, there is the
further difficulty that the patient probably has been exposed to
several sources of infection ; and one has to attempt to balance
quantitatively the probability of these as the active agent in
producing disease. They may, in fact, have all been co-operating
in overcoming the patient's powers of resistance.

For many years past I have carefully investigated the history
of all cases of phthisis notified in Brighton. From 1902 on-
wards (p. 341) a large proportion and during the last year over
half of these patients have been treated in the Borough Sana-
torium. It has been possible in this way to obtain fuller informa-
tion as to the patients than would have been otherwise practicable ;
and this information has convinced me that histories obtained
at a single interview with phthisical patients cannot be trusted.
My experience is that the inquiries made at the first interview set
up trains of thought and recollection, which when followed up
at a later interview may completely alter the opinion formed at
the first interrogation. For this reason I have preferred to state
below a summary of a hundred consecutive sanatorium cases
investigated very carefully by Dr. H. C. Lecky, at the
Brighton Sanatorium, in preference to a very much larger
number, in which less complete information had been obtained.
These hundred cases had been exhaustively studied, and for that
reason the results obtained respecting them are stated in some

72 THE PREVENTION OF TUBERCULOSIS

detail. The conclusions based on the less completely exhaustive
investigation of a much larger number of cases coming under
my observation during a series of years, confirm the view that
prolonged latency of already existing disease is not so rare as it
is often supposed to be.

In the following table a hundred patients, thus fully investi-
gated, are classified according to the history obtained : —

TABLE XIX

No. of
Patients.
(A. Definite limited domestic infection and definite onset . 20

B. ,, ,, ,, ,, indefinite onset

C. Possible continuing domestic infection and definite onset

D. „ ,, extra-domestic infection and definite onset

E. ,, ,, public-house ,, ,, ,,

'F. ,, ,, domestic ,, indefinite onset

G. ,, ,, extra-domestic ,, ,, ,,

H. No exposure known and definite onset
I. Suspicion of temporary exposure and definite onset .
J. No exposure known and indefinite onset
K. Suspicion of temporary exposure and indefinite onset
L. History incomplete after every effort made .

7
7

100

By limited infection is meant that the exposure to infection
is known to have ceased at a given date, as, for instance, at the
death of a consumptive mother.

The difficulties of classification of histories of infection are
very great ; and the above headings have been adopted after
much consideration. Thus it has been necessary to separate
cases where the date of onset of symptoms could be definitely
stated from others in which this was dubious ; and to separate
cases where a definite limit to exposure to infection could be
stated from others in which exposure may have continued up to
the date of onset of the patient's illness.

The table shows that in 32 per cent, of the cases definite
infection could be traced. In a further 23 per cent, there was a
possibility of such infection, but the history was not so precise
as in the previous group. In 25 per cent, of the total cases no
exposure to infection could be traced. In a further 18 per cent,
there was suspicion of temporary exposure to infection, but the
history was defective or indefinite.

The fact that in 25 per cent, of the cases no source of infection
could be discovered is instructive. Even though a considerable
number of these are explained probably by the fact that many

INFECTIVITY OF TUBERCULOSIS 73

patients having open tuberculosis are never seen by a doctor
and do not die of this disease, it appears likely that in an
uncertain proportion of cases of phthisis, — probably among the
most susceptible members of the community, — effective infection
may be received from merely casual sources of infection.

STATISTICAL STUDY OF LATENCY.— In Table XIX. out of 100
total cases 20 had a definite history of infection ceasing at a
known date, followed after an interval by phthisis in persons
who had been exposed only, so far as could be ascertained, to
this limited infection. Of these 20 patients n were men and
9 women. The duration of latency in these cases was as
follows : — Under i year, i ; 1-2 years, 4 ; 2-3 years, o ; 3 years, i ;
5 years, 2 ; 6 years, i ; 9 years, i ; 10 years, 3 ; 13 years, i ;
15 years, i ; 17 years, i ; 20 years, 2 ; 22 years, i ; 27 years, i.

Thus of the 20 cases 6 only appeared to have had a latency of
less than 5 years ; in 4 the latent period varied from 5 to 10
years ; and in 10 there was a latency of over 10 years.

In the same table 12 additional cases are noted in which
infection ceased at a given date, but the date of onset of phthisis
in the person exposed to this infection could not be definitely
ascertained, though always after the cessation of exposure. In
4 of these the duration of latency could not be stated even
approximately ; in one it was probably 2 years ; in one, 3 years ;
in one, 5 years ; in one, 7 years ; in one, 8 years ; in one, 10
years ; in one, 14 years ; and in one " many years/'

In each of these cases it is possible that more recent casual
infection, and not the more remote protracted infection, was
responsible for the tuberculosis. The view I have taken through-
out is that given one patient in a family the protracted and
intimate relationships of domestic life are much more likely
than casual extra-domestic infection to be the chief means of
spreading tuberculosis ; and that this is so even when the history
indicates a period of latency of many years. The possibility of
long latency and the importance of protracted duration of
exposure in producing efficient infection will be better appre-
ciated when the next chapter and Chapters XIX. to XXVI.
have been read.

CHAPTER X

LATENCY IN TUBERCULOSIS

ANALOGY WITH ACUTE INFECTIOUS DISEASES.— Certain
features characterise all diseases due to the reception
into the body of specific infective material from with-
out. They may be illustrated by the case of small-pox. A
person inhaling the contagion or microbes of this disease, unless
protected by a previous attack of small-pox or by vaccination,
goes through the following stages. There is first a period of
incubation, or latent period, of about twelve days, in which no
symptoms of disease can be detected. Then occur severe initial
symptoms which usually consist of vomiting, severe headache
and backache, with fever, followed seventy- two hours later by the
characteristic skin eruption. After an illness of several weeks, all
the symptoms have disappeared, the patient is no longer infectious
to those coming into contact with him, and if again exposed to
infection he is himself as a rule immune against further attack.
That is a typical instance of the course of an infectious disease,
and such diseases as whooping-cough, measles, scarlet fever, and
typhoid fever conform more or less to the type. Some acute
infectious diseases conform less completely to it. Thus in
diphtheria the immunity conferred by one attack appears to be
less complete than in the diseases just mentioned, and in erysipelas
one attack appears to predispose to rather than to protect against
a second attack. It is not necessary to enter into the possible
causes of lack of immunity in these instances. In diphtheria,
and possibly also in erysipelas, it is sometimes associated with
the persistence in the patient's body of the bacteria causing the
disease. Thus Gresswell in 1886 brought forward certain facts
which appeared to show that " diphtheria in certain individuals
may become a chronic disease, and from time to time enter upon
an active and infectious phase." I have elsewhere collected
similar evidence (1904) of cases of diphtheria, and occasionally

LATENCY IN TUBERCULOSIS 75

also of scarlet fever, in which the infection persisted for very
long periods, and subsequently reappeared after intervals of
considerable length. The analogy between these exceptional
conditions and tuberculosis is obvious. In both there is per-
sistence of infection in a more or less latent form, and in both
a partial failure to secure by one attack immunity from further
attack.

INCUBATION PERIOD OR FIRST PERIOD OF LATENCY.— In the
acute infectious diseases this is usually a fixed and somewhat
short period, seldom exceeding a few days. In tuberculosis it
may be a few weeks or many months, or even many years. There
are not wanting illustrations of similar prolongations of this
period in other diseases. Thus in pebrine, the silkworm disease
investigated by Pasteur, the egg when laid contains the germs
of the disease. These do not increase in number in the winter
in the eggs, even though the latter are kept at a favourable
temperature ; but in spring, with the growth and development
of the egg, the disease again becomes fully established. In
leprosy, a disease having close affinities to tuberculosis, two to five
years is given as the common period of incubation, but a case of
probable latency of forty years is described by Abraham (1896).
Hydrophobia usually develops, if at all, within six weeks from
the time of the bite of a rabid dog. It has been known, however,
to remain latent for eighteen months and possibly for several
years.

The following illustrations from my case-book illustrate
prolonged latency between the last known exposure to infection
and the occurrence of an attack of pulmonary tuberculosis. In
speaking of minimum latent periods in these cases, it must be
understood that every other ascertainable possibility of infection has
been investigated with negative result, and that so far as could be
ascertained the patient had only been exposed to the source of
infection which is detailed, and to those minor casual infections
(P- 73) to which everybody may be exposed.

CASE 9.— Mrs. E. S., aet. 32, was admitted to the sanatorium
with phthisis August 13, 1906. She had been exposed to pro-
tracted infection as shown in the following scheme, having
nursed her father, mother, and two brothers while they were ill
and dying with phthisis : —

THE PREVENTION OF TUBERCULOSIS

Domestic Infection.

Age.

Year.

Extra-domestic Infection.

The history makes it probable

1874

that during the whole of her
childhood her mother and one

brother were suffering from

chronic phthisis.

No evidence of any obtainable.

In 1891 patient's brother, set. 21,

1891

and mother, set. 61, died of

phthisis at home. Both nursed

by E. S. In the same year

another brother, set. 30, died of

phthisis in another house. He

also was nursed by this patient.

Married .....

1893

Patient's father died of phthisis

1894

at E. S.'s house. He was very

ill for a year, and in bed for a

month before death.

In Nov. E. S. in bed for a week

1903

with "influenza and left

pleurisy." Some cough ever

since.

E. S.'s boy, aged i| year, died

1904

of acute tuberculosis.

Admitted to sanatorium

1906

Comments on Case 9. — First exposure to infection was probably
in infancy (1874). The last known exposure was twenty years
later. First symptoms of tuberculosis occurred in 1903. The
maximum latent period is therefore twenty-nine years, the
minimum latent period nine years.

CASE 10. — H. E. G., aet. 25, was admitted to the sanatorium
May 28, discharged July 30, 1906.

During his holidays H. E. G. visited his home, but there were
no opportunities of protracted infection from the age of 15 to 21,
when his cough began. Probably the latent period was much
longer than six years, but possibly it was less. There was no
family history of tuberculosis on the paternal side ; but the
mother's two sisters had died of pulmonary tuberculosis, and the
evidence pointed to her having suffered from the same disease at
or before the time of her marriage.

The evidence in the preceding cases is purely circumstantial,
and when stated in skeleton and apart from a knowledge of the
intimate detail of each case is relatively unconvincing. The
conclusion, however, that the majority of such cases are really

LATENCY IN TUBERCULOSIS

Domestic Infection.

Age.

Year.

Extra-domestic Infection.

1881

Lived at Ba. until 15 years old,

and there all the following

cases of tuberculosis occurred.

Father died

of phthisis one year

1895

before H

E. G. left home at

the age ol
Probable

•
'Brother died of
phthisis.

15
16

1896
1897

Apprenticed to a draper in
London.

mini-

Mother died of

1898

m u ml

tuber c u 1 o s i s of

primary
latent
period.

kidney.
Sister died of tuber-
^ culosis of intestine.

1899

Became a draper's assistant
in C.

1902

Cough began this year, and in

consequence he went to sea

1903

as a ship's steward.
Had to give up sea-life, owing
to an attack of pleurisy.

1904

Began to expectorate.

1906

Has not been working for the

last two years.

cases of prolonged latency and not of yielding to casual and
undetected more recent infection, is supported by converging
lines of evidence, which may next be considered.

(1) There is pathological and experimental evidence of pro-
longed latency, primary and secondary, both in tuberculosis and
in other infective diseases, both in adults and children.

(2) The clinical occurrence, both in tuberculosis and in other
infective diseases, of prolonged secondary latency — i.e. of a period
during which symptoms of diseases previously present are in
abeyance — confirms the occurrence of a similar latency before
the first clinical symptoms appear.

PATHOLOGICAL AND EXPERIMENTAL EVIDENCE OF PROLONGED
LATENCY IN TUBERCULOSIS. — Attention has already been drawn
to the frequency with which small tuberculous lesions are found
post-mortem in those who have died from diseases other than
tuberculosis (p. 49). Thus Stengel (p. 255) says :—

The lesion may become encapsulated and so remain for years without
producing manifest clinical symptoms. This encapsulating membrane
may subsequently be penetrated and widespread infection occur. Such
latent tuberculosis is particularly frequent in the post-bronchial glands.
These are often found diseased in autopsies in which no tuberculosis is
found elsewhere. In a notable proportion of such cases emulsions of such

78 THE PREVENTION OF TUBERCULOSIS

glands produce tuberculosis in guinea-pigs, showing true latent tuberculous
disease. Such lesions explain sudden miliary tuberculosis, in which no
primary focus is found during life.

Cornet (p. 449) says :—

It has been shown, by means of inoculation tests, that if these (en-
capsulated) foci contain caseous material, virulent bacilli are always
present. Only absolutely fibroid scars, as well as thoroughly calcified
nodules, proved to be sterile (Kurlow, Green). The consumptive may
be said to sit upon a volcano. Until the capsules have become absolutely
perfect and impervious barriers, every event which tends to weaken them,
or to open up the defects in their architecture, may become the occasion
of a further dissemination of the bacilli, of a lighting up of a fresh attack.

It should be added that after giving the above evidence of
continued virulence of tubercle bacilli incarcerated in old caseous
lesions, Cornet makes the following remarks, which appear to
be contradictory to his statement quoted above, and, unlike it,
are not supported by experimental evidence : —

It seems to me a little far-fetched to attribute a fresh outbreak of
the disease, after a quiescence of years, to the resurrection of the bacilli
imprisoned in the old focus, since we know that the life period of the
bacilli is bounded by certain definite and narrow limits.

The latter statement is based apparently on the assumption
that the bacillus will find as great a difficulty in surviving in
caseous nodules at the body temperature, as it experiences after
having been expelled with the expectoration. Cornet emphasises
Kitasato's demonstration that most of the bacilli in the expectora-
tion are already dead ; but such expectoration is still commonly
extremely virulent ; and bacilli in the expectoration imply
destructive changes of tissues carried to a much further point,
than those manifested in chronic caseous nodules. Cornet asks
the question (p. 315) : " What biological facts entitle us to
assume that the bacillus is capable of remaining latent through
decades, for forty or sixty years, in the human body ? " He
is answered partially by the preceding quotations, including his
own statement. Other experimenters have furnished similar
evidence, which, although not absolutely direct and certain,
renders very probable the continuance of latency over many years.
Thus J. K. Fowler has shown that recrudescence of human
phthisis coincides in certain instances with active changes in the
old lesions. In one instance the latency had lasted a period of
forty years. Haemoptysis generally indicates fresh mischief lit

LATENCY IN TUBERCULOSIS 79

up in an old focus of disease. It was the association of recent
general tuberculosis with recurrence of active trouble in an
old focus which led Buhl to his great generalisation as to the
origin of general tuberculosis by self-infection (see p. 37).
Debove and Achard (p. 271) speak of these old foci as "le feu
qui couve, qui peut s'etendre" under the influence of protracted
overwork, fatigue, sorrow, or of an acute inflammatory attack.

In the preceding remarks it has been assumed that naked-
eye evidence of old disease was to be found in the cases in which
old foci produced acute tuberculosis. It may be noted, however,
as having a possible bearing on the problem of latency, that
lymphatic glands may contain living tubercle bacilli without
showing naked-eye signs of implication. The duration of life
of tubercle bacilli under those conditions is unknown. Loomis
(quoted by H. Walsham, p. 6), on examining thirty cases in which
there were no signs of old or recent tuberculous lesions, found
that in eight cases the bronchial glands were infective to rabbits.

A. Macfadyen and MacConkey (1903) took mesenteric
glands from the bodies of children who, dying of other diseases
than tuberculosis, at the autopsies showed no evidence of tuber-
culosis. From these glands they injected material into guinea-
pigs, and tuberculosis was produced in 25 per cent, of these.
How long these bacilli had been in the tissues without producing
evidence of disease cannot be said, nor can it be said how much
longer they would have survived had the children lived ; but these
interesting observations open up the possibility of prolonged
latency of tubercle bacilli in the absence of naked-eye lesions.

Tuberculous lesions may have long periods of latency in
animals, as well as in man. Thus Baumgarten (quoted by
Washbourne), inoculated tubercle bacilli into the anterior
chamber of the eye of a rabbit. A tubercle formed ; this was
arrested and converted into cicatricial tissue under treatment
by tuberculin. Nine months later the apparently cured tubercle
started once more into activity. The active phase subsided for
the second time, and there was apparent healing. A year later
it again became active and now spread rapidly, general tuber-
culosis being produced. This instance, in which the bacilli re-
mained alive during latent periods of nine months and- a year,
was carried out under conditions avoiding the possibility of
fresh infection from without.

8o THE PREVENTION OF TUBERCULOSIS

Miiller (1906) states that he re-tested with tuberculin two
sets of cows which when calves had been fed with infected milk,
and which owing to their positive reaction to the first test had been
fattened ; the interval between the two tests in one set was a
year, in the other two years. During the interval the cows had
been isolated. In the first set the whole of the ten cows reacted
again ; in the second twelve out of fourteen reacted. Other
cases have been observed where calves which reacted to tuber-
culin first showed symptoms of tuberculosis i-J- to 2,\ years later.
In one batch of twenty cows the animals were 4 to 5 years old
before symptoms appeared. Then they suddenly in quick succes-
sion became ill and had to be slaughtered. In all of them an
advanced and apparently very old abdominal tuberculosis was
found, the lesions being large and showing caseation and ex-
tensive calcification, with recent tuberculosis of the lungs and
other organs. Miiller adds : a few other cases of the same kind
have been observed in which entire years elapsed before the
symptoms were exhibited, and in which there had been observed
a tuberculosis of the udder at the critical time.

LATENT TUBERCULOSIS IN CHILDREN. — Ganghofner of Prague
(1905) has recorded as follows the results of 1800 autopsies on
children dying in that city from causes other than tuberculosis,
and presenting no symptoms of tuberculosis : —

Out of 460 deaths of children in the 1st year of life

latent tuberculosis was found in 33= 7*1 per cent.

„ 536 „ „ aged 1-2 „ ,, „ 86=16-0 „

„ 476 „ „ „ 2-4 „ ,, „ 117 = 24-5 ,,

„ 271 „ „ ,, 4-6 „ „ „ 73 = 26'9

„ 123 „ „ „ 6-8 „ „ „ 33 = 26-8 „

English statistics give somewhat similar results. It has
further to be noted that the absence of tuberculous lesions visible
to the naked eye does not completely prove the absence of tuber-
culosis. Ganghofner in the paper referred to above gives in-
oculation experiments proving the presence of latent tuberculosis
in children in whom ordinary macroscopic and microscopic
examination had failed to prove its presence, and similar obser-
vations have been made by others (p. 79).

Unless it can be shown to be an exceptional event for living
tubercle bacilli to be present in old tuberculous nodules, the
facts narrated in this and the preceding paragraph give a primd

LATENCY IN TUBERCULOSIS 81

facie case in favour of the view that adult tuberculosis may often
be due to the recrudescence of the disease established in small
foci within the body in early life. This view was emphasised by
Marfan (1905), whose conclusions were that (i) the infant is
most exposed to tuberculosis at ages 1-6 ; and that (2) in a
considerable number of cases showing evidence of tuberculosis
at or after adolescence, the disease has not been caused by recent
infection, but by an infection acquired in early life and remaining
latent in the interval.

The same conclusion is confirmed by the facts relating to pro-
longed secondary latency as given below.

PROLONGED SECONDARY LATENCY IN DISEASES OTHER THAN
TUBERCULOSIS. — There is, as already mentioned, abundant
evidence that diphtheria bacilli may in exceptional cases persist
in the throat for months, or rarely even for several years, without
any evidence of disease, a second attack being then produced
without any known external re-infection. The clinical evidence
of this phenomenon in tuberculosis and in diphtheria is strongly
confirmed by bacteriological evidence concerning other diseases.
Thus Washbourne (1896) states that the spores of the hay
bacillus have been found alive in the organs 78 days after sub-
cutaneous injection. He quotes an instance given by Schafer
in which diphtheria bacilli persisted in the throat for six months
after the attack. I have published (1904) instances of diphtheria
in which infection persisted 102 and 170 days after the patient
was apparently well, and cases of scarlet fever in which similarly
persistent infection was shown.

The typhoid bacillus sometimes persists in the gall bladder,
the bones, etc., for a long time after an attack of typhoid fever.
Hinze (quoted by Washbourne) gives a case of a periosteal node
appearing four months after an attack of typhoid fever ; six
months later this became an abscess, which when opened and
cultivations taken from it, showed typhoid bacilli. Buschke
found living typhoid bacilli in an abscess seven months and
Chantemesse and Widal fifteen months after an attack of typhoid
fever. A most remarkable case for this disease is recorded
by Dudgeon and Gray (1906), in which the discharge from
a bone sinus three years after the patient's attack of typhoid
fever gave pure cultures of typhoid bacilli, and appeared to be
the cause of the same disease in the patient's wife.
6

THE PREVENTION OF TUBERCULOSIS

Syphilis has many points of resemblance to tuberculosis,
especially in the slow evolution of its phenomena and the long
intervals during which symptoms are absent. In this disease
recrudescence of symptoms frequently occurs, when fresh ex-
ternal infection can be excluded with certainty, after twenty or
thirty years of freedom from symptoms ; and in such cases it is
occasionally noted that, as in tuberculosis, recognisable initial
symptoms may have been entirely absent.

CLINICAL EVIDENCE OF PROLONGED SECONDARY LATENCY
IN TUBERCULOSIS. — The following cases are typical of a large
number in which long intervals elapsed between the first
attack of tuberculosis and later attacks, and in which, I think,
there is strong reason for believing that the later attack was
caused by changes in the old foci of disease, freeing the bacilli
from their incarceration and disseminating disease to other
parts.

Domestic Infection.

Age.

Year.

Extra-domestic Infection.

1852.

No exposure to infection known.

1861

M. D.'s schoolmistress at the

National School fell ill.

1864

About this year the schoolmistress

died of phthisis, after being ill

for 2 to 3 years, during the

whole of which time M. D.

saw her nearly every day, sit-

Probable^

ting in her room, and generally

mini-

helping her.

mum

Family removed from

1867

Frimary

Ch— m to C— n.

atent

M. D. had no symptoms of

period.

phthisis for about 4 years after

the death of the teacher.

M. D. was treated for phthisis at

1869

theBrompton Hospital 6 months

as an out-patient and 5 months

as an in-patient.

'P'ather killed in an

accident.

M. D. married .

Secondary

No cough or expector-

latent-

ation for 33 years,

period.

although delicate.

Came to Brighton

Cough and expector-

1902

ation began again.

Admitted to sanatorium

1906

LATENCY IN TUBERCULOSIS 83

CASE ii.— Mrs. M. D., aet. 55, was admitted to the sanatorium
August 20, 1906, with chronic phthisis. Her family history shows
a complete absence of this disease. Her personal history is
presented in the scheme on the preceding page.

Comments on Case n. — The above facts show in this case a
primary latent period of about 4 years, followed by an illness last-
ing about a year ; and then a secondary latent period of 32 years.

CASE 12. — Mrs. A. W., aet. 24, admitted to sanatorium May 12,
discharged June 9, 1906. Increase of weight from 8 st. 5J- Ib. to
9 st. 3 Ib. Signs of disease at left apex. Has been married 6
years, and done only domestic work since that time. Has had
two children, one well, one died aged 3 years of " bronchitis."
Husband healthy. Patient was a domestic servant before
marriage, and did not work for any consumptive family. Patient's
father died of phthisis 2j years ago after an illness of 4 years.
Patient and her husband lived with the father until 3 years ago.
In 1904 she had " influenza," and afterwards was fairly well
until March of the present year. When aged 14 had (in 1896)
a gland removed from the left side of the neck, and in 1904 a
gland was removed from lower down on the same side of the
neck.

Comments on Case 12. — If it be assumed that the first tuber-
culous gland was the focus of infection of the lung, there was a
secondary latent period of about 10 years. It is possible that
the father of the patient had infected her more recently. This
would make the new primary latent period about 2 to 3 years.

CASE 13. — A. B., aet. 49, a policeman, was notified on
September 2, 1905. Tubercle bacilli had been found in his sputum
on August 31 . Had right pleurisy 16 years ago. His cough dates
from October 1903, and he had some haemoptysis early in 1905.
He was said by his doctor to have had " bronchial catarrh" in
October 1904. Had been in the police service 23 years, and
before that had been a seaman. Is an alcoholic subject. He
was admitted to the sanatorium September 8, discharged
October 6, 1905, and died March 26, 1906.

Comments on Case 13. — If, as is probable, the pleurisy was
tuberculous, there appears to have been a latent period of 13
years between it and the subsequent development of cough.

THE PREVENTION OF TUBERCULOSIS

On this supposition, we must assume an earlier infection to
which the pleurisy was secondary. The source of infection
is undetermined. The opportunities of infection both in his
occupations and in connection with alcoholic indulgence were
numerous, and the latent period may therefore have been shorter
than given above, there being numerous infections at frequent
intervals.

CASE 14. — J. M., aet. 29, admitted to sanatorium March 24,
discharged April 20, 1906. Has been a house painter for 6 years,
before that a soldier for 7 years, of which 6 were spent in India.
Has been married 5 years, but has had no children. His wife
is healthy. He has had a cough as long as he can remember,
and he had haemoptysis before going to India. The cough ceased
while he was in India, but reappeared on his return, and he has
gradually deteriorated in health. His father died of phthisis
when he was 10 years old. His brother M. M. was admitted
to the sanatorium with J. M., having phthisis and renal disease.
The brother's first symptoms date from about 4 years ago. The
two brothers have not lived together for 6 years, and then only
for a short time.

Comments on Case 14. — The father probably infected both
these patients more than 19 years ago. In M. M.'s case there
was an initial latent period of about 15 years. In J. M.'s case
symptoms of phthisis appeared much earlier ; but an interval
of 6 years followed, in which all symptoms were in abeyance.

Domestic Infection.

Age.

Year.

Extra-domestic Infection.

1885

None discovered.

Was treated in Brixton for disease

1892

of the right lung, being under

a doctor for several months.

Was then sent into the country

for three months, and has

been well from that time until

Easter 1906, when she again

1906

began to suffer from cough.

Was sent to Brighton on account

of this cough ; and when exa-

mined shortly afterwards, was

found to have a cavity at the

right apex.

LATENCY IN TUBERCULOSIS 85

CASE 15. — Jessie R., set. 21, was admitted to the sanatorium
July 26, and discharged October 25, 1906. She had disease,
including cavitation, of the upper part of the right lung. See
scheme on preceding page.

Comments on Case 15. — The first attack 14 years ago was
diagnosed as phthisis. From this date to her present attack,
the patient had been well. There was no family history of
tuberculosis, and the patient, who is in fairly good circumstances,
has not been exposed to any known infection.

CHAPTER XI
SOURCES OF INFECTION

SINCE tuberculosis is an infective Ldisease, its prevention
evidently must depend upon an accurate knowledge of
the sources from which infection is derived. With rare
exceptions, tuberculosis in man has been attributed solely to
infection derived from other human patients, or to infection
from food animals, especially cattle or pigs. The possibility
of infection by animal food-stuffs raises the large question of
the inter communicability of human and bovine tuberculosis,
which is discussed in Chapters XVI. to XVIII. Tuberculosis
from lower animals is only likely to be conveyed to man to any
considerable extent by the ingestion of infected foods, especially
milk. From human patients infection may be direct, e.g., in kiss-
ing or during coughing accompanied by the projection of particles
of expectoration into another person's mouth or nostrils ; or
indirect, as when the dried expectoration of a consumptive is
inhaled. The chief possible means of infection are thus—

1. The inhalation of dried expectoration.

2. The inhalation of particles of wet expectoration.

3. The ingestion of tuberculous milk or other foods.

Of these three it is agreed by most hygienists that only a
relatively small part of the total human tuberculosis is due to
tubercle bacilli of bovine or other animal origin, though opinions
differ as to the size of this proportion. Very few agree with
von Behring in considering bovine infection as the sole or
even the chief source of human tuberculosis.

Both i and 2 named above are concerned with coughing
and expectoration, which are the main means of tuberculous
infection. Other discharges from tuberculous patients, as
from the bowels in tuberculous enteritis, — or even without such
enteritis, when tuberculous expectoration has been swallowed,
— from the skin in tuberculous abscesses, by the urine in renal

SOURCES OF INFECTION

tuberculosis, are doubtless infective, but for fairly obvious
reasons they seldom have the same opportunities to cause
infection as the expectoration.

Expectoration can, as indicated above, spread infection
in two ways. Either it is inhaled after having become dried
and powdery, or it is inhaled directly in the form of spray or
small pellets expelled as the patient coughs. These two chief
modes of infection are fully considered in Chapter XII. In
this chapter will be considered briefly certain other modes
of infection, less important than the inhalation of infective
dust or spray, but conveniently disposed of at this stage. These
methods consist in (i) inoculation with tubercle bacilli, (2) in-
fection by kissing or by other means of conveying infected saliva,
and (3) infection by contaminated hands or by flies.

INOCULATION WITH TUBERCULOUS MATERIAL. — The sub-
cutaneous injection of tubercle bacilli in experimental animals
produces tuberculosis which, following the lymphatic tracts,
may soon become general. Such a result is rare in ordinary
life, probably because the dose of infection received through
cuts or abrasions of the skin is usually small. Lupus, a disease
eventually causing a disfiguring ulceration of the skin, is a
local form of tuberculous infection. It rarely occurs in covered
parts of the skin, and is probably caused by accidental inocula-
tion of tubercle bacilli. Local tuberculosis has occasionally
been produced at the seat of local injuries, received, for instance,
while making autopsies on tuberculous patients. Such cases are
rare, and the resulting tuberculosis seldom extends beyond the
next chain of lymphatic glands ; but in a few instances general
tuberculosis has followed.

The possibility of inoculation with tuberculosis during
vaccination with bovine lymph has been asserted. It must
be regarded as a very remote and almost negligible possibility ;
and as non-existent, when, — as is always the case in well-regu-
lated vaccine establishments, — the calves from which the lymph
has been obtained are killed and minutely examined for tuber-
culosis, and the lymph never distributed unless complete absence
of tuberculosis can be certified.

INFECTION BY SOILED HANDS. — Obviously a phthisical
patient who is not cleanly in his or her habits might easily infect
hands and fingers during expectoration, and articles of food

88 THE PREVENTION OF TUBERCULOSIS

might thus become infected. Baldwin of Saranac Lake (quoted
by Lartigau, p. 121) examined the hands of fifteen consumptives,
and of this number ten were found to be contaminated with
tubercle bacilli. These facts emphasise the importance of
care in the use of handkerchiefs and spitting-cups, and the
need for washing the hands after they have become fouled.
This source of infection must, however, be regarded as of much
less magnitude than others to be considered subsequently.

INFECTION BY THE SALIVA. — Drinking-cups, spoons, etc., used
in common may be a source of infection, and so likewise may
kissing, if tubercle bacilli are present in the saliva. On this point
divergent statements are made, Cornet (Cornet, p. 187) saying
that the saliva is ordinarily germ free ; while several observers have
confirmed the frequent presence of tubercle bacilli in the saliva
(Lartigau, p. 121). Cornet himself (Cornet, p. 166) minimises
the value of the preceding statement by urging that even if
the saliva " should contain bacilli, they would be carried into
the mouth and the digestive tract of the other person, and
not into the lungs " ; although he says that " with children
the case is different. Their mucous membranes are far more
susceptible to the bacteria, and it may be that kissing is not
infrequently of moment in producing scrofulous cervical glands."
With his statement that " so far as we are able to judge, this
danger does not play an important r6le among adults," I am
inclined to agree. Dosage would probably be small in infection
by kissing or by drinking-cups, etc., and it is unlikely that a
serious amount of infection is often produced by this means
alone.

INFECTION BY FLIES. — It is obvious that flies having fed on
or having been fouled by tuberculous expectoration might
contaminate food and thus convey infection. This possibility
has been proved experimentally. Thus Spillmann and Haus-
halter (Cornet, p. 82) found tubercle bacilli in the abdominal
cavity and in the faeces of flies which had sucked at the sputum
cloths of consumptives. These observations have been confirmed
by others. In measuring the relative importance of this method
of spreading infection, it has to be remembered that the faeces
of flies and the amount of material capable of being carried
on their limbs are extremely minute as compared with the
material in a single expectoration.

CHAPTER XII

SOURCES OF INFECTION (Continued)— DUST AND SPRAY

VILLEMIN (p. 38) appears to have been the first authority
to recognise the importance of dried tuberculous expectora-
tion as a vehicle of infection, most previous writers having
laid stress on the supposed dangers of direct personal com-
munication, or even of handling tuberculous corpses (p. 35).
The deaths from phthisis of Bayle, Laennec, Louis, and several
other French physicians who practised much among con-
sumptives, doubtless favoured the view of direct infection from
consumptive patients.

Even in recent years the idea that AIR QUIETLY EXPIRED by
a consumptive may contain tubercle bacilli has been entertained,
and some experiments by Ransome (1882) and by Williams
(1883) appeared to confirm it. It is probable, however, that in
these experiments insufficient precautions were taken to exclude
the possibility of spray or droplets ejected during coughing
gaining access to the experimental apparatus. Tyndall has
supplied the experimental proof that in quiet breathing expired
air is absolutely sterile.

For the rest of this chapter it will be assumed that inhaled
dust can penetrate to the air cells of the lungs. The evidence
for this statement, and the discussion of the relative share of
this and other methods of infection will be given in later chapters.
In this chapter we shall discuss the operation of infection by
dust and by spray, as far as possible in the historical order of the
most important experiments that have been made.

KOCH'S EXPERIMENTS AND CONCLUSIONS. — Koch describes
his procedure in experiment 26 of his classical paper as follows : —

A very roomy box, having on one side an opening for the orifice of
the spray apparatus, was placed in a garden at a good distance from
any habitation. The spray apparatus was placed outside the box, with

its orifice projecting into the interior. By means of elastic tubing and

90 THE PREVENTION OF TUBERCULOSIS

a suitable length of lead pipe passing through the woodwork of a closed
window, the apparatus was connected with an indiarubber bellows, and
so could be worked from the room beyond the region of the spray.

A pure culture taken from a phthisical lung in the human subject,
No. i, and carried through twenty- three generations in fifteen months,
was rubbed up with distilled water, and the fluid diluted to such an ex-
tent that it looked almost clear. Any visible fragments present in the
fluid subsided after standing a short time ; the upper layer, which showed
hardly any opacity, was poured off and used for inhalation. Fifty c.cms.
were sprayed in the course of half an hour on three successive days, and
inhaled by the following animals in the box : 8 rabbits, 10 guinea-pigs,
4 rats, and 4 mice. After the inhalation, the animals were kept in
separate roomy cages and well looked after. In some of the animals,
dyspnoea appeared after ten days, and 3 rabbits and 4 guinea-pigs
died in the course of fourteen to twenty-five days. All the remaining
animals were killed twenty-eight days after the last inhalation. All the
rabbits and guinea-pigs had numerous tubercles in the lungs, the size
of the tubercles being proportionate to the length of time the animals
had lived after inhalation.

In this experiment Koch was spraying cultures made from a
tuberculous lung, but in his comments on it he says :—

There can likewise be no doubt as to the manner in which the tubercu-
lous virus is carried from phthisical to healthy subjects. By the force of
the patient's cough particles of tenacious sputum are dislodged, discharged
into the air, and so scattered to some extent. Now numerous experi-
ments have shown that the inhalation of scattered particles of phthisical
sputum causes tuberculosis with absolute certainty, not only in animals
easily susceptible to the disease, but in those also which have more power
of resisting it. It is not to be supposed that man would be an excep-
tion to this rule, but, on the contrary, we may surmise that any healthy
person brought into immediate contact with a phthisical patient, and
inhaling the fragments of fresh sputum discharged into the air, may
thereby be infected. But probably infection will not often take place in
this way, because the particles of sputum are not small enough to remain
suspended in the air for any length of time. Dried sputum, on the
contrary, is much more likely to cause infection, as, owing to the negligence
with which the expectoration of phthisical patients is treated, it must
evidently enter the atmosphere in considerable quantity. The sputum
is not only ejected directly on the floor, there to dry up, to be pulverised
and to rise again in the form of dust, but a good deal of it dries on bed-linen,
articles of clothing, and especially pocket-handkerchiefs — which even
the cleanliest of patients cannot help soiling with the dangerous infective
material when wiping the mouth after expectoration — and this, too, is
subsequently scattered as dust.

It is evident from this quotation that Koch regarded dried
sputum as the most fertile source of infection. This view has

SOURCES OF INFECTION 91

been confirmed by the experiments of Cornet, Strauss, and many
others. We must next consider the experiments and views of
the school of Fliigge.

FLUGGE'S EXPERIMENTS AND CONCLUSIONS.— The following
summary is made from Fliigge 's well-known paper (1898). He
quotes results previously obtained by Sticker, who failed to infect
animals by making them inhale tuberculous sputum mixed with
fine sand, and showed that the failure was owing to the fact that
although the conglomerate of sputum and sand was driven into
the apparatus by a rapid current from bellows used in the experi-
ment, yet it failed to be inhaled by the feeble inspiratory suction
of the animal. On the other hand, Cornet succeeded in pro-
ducing tuberculosis by inhalation in guinea-pigs, by discharging
the loaded air direct into the animals' mouths, or by holding
them over a carpet while it was swept, so that the sputum par-
ticles with which it had been strewn were raised. These experi-
ments in which the sputum is artificially dried and powdered,
and the air currents are more rapid than those occurring natur-
ally in a room, are, according to Fliigge, not comparable to normal
conditions of life.

The important point to settle is whether under natural
conditions sputum, as for instance in a handkerchief, ever
assumes the degree of dryness requisite for the dust to escape
from it and become the source of infection. Experiments were
made on this point by Beninde. He showed that weak currents
of air would not disperse bacteria from handkerchiefs which had
been deprived of 60 per cent, of their moisture by being kept in
the pocket for one day. Fliigge also states that

sputum on the floor very rarely is left long enough to reach the neces-
sary degree of dryness ; each washing of the floor lessens the danger. In
ordinary dwelling-houses it is next to impossible to find dried sputum
in the dust, though in workshops, etc., where men may spit on the floor,
tuberculous dust can quite well become sufficiently dried to be blown
up into the air.

He goes on to say that

sputum is difficult to pulverise finely, and the coarser particles are not
dangerous. It is true that sweeping and dusting disturb the coarser
particles, but these do not often reach the respiratory passages, and fall
so quickly again on to any flat surface, that it is not possible for much to
be inhaled ; and as the finer particles, capable of suspension for a long
time, are very rarely and sparsely present, the danger is very slight.

92 THE PREVENTION OF TUBERCULOSIS

Fliigge summarises the results of his experiments in the
following words : —

Infection from pulverised dried sputum is doubtless possible, but it
occurs relatively seldom, because particles fine enough to be conveyed
readily by air can only be formed from completely dried sputum, and
then only in very limited quantities.

In his view that the danger from dried sputum has been
exaggerated, Fliigge in certain particulars was anticipated by
Cornet, who, although he is the chief advocate of the view that
tuberculosis is spread by infective dust, minimises its operation
in the following words extracted from his first work : —

Any one who has himself tried to rub well-dried sputum into particles
and to pulverise it very finely will agree with me that it is no easy task to
prodbce a really fine powder which remains suspended in the air for some
time. The strong statements that have been made up to now — that
one has only to rub with the foot on the dried sputum to raise immediately
a cloud of infectious germs — are absolutely false.

EXPERIMENTAL EVIDENCE OF SPRAY INFECTION. HEYMANN.
— Leaving aside experiments under artificial conditions, we may
consider those made with the natural spray produced by coughing,
sneezing, and speaking. Laschtschenko, after washing his mouth
with broth containing Bacillus prodigiosus, was able to recover
these from agar plates dispersed over a room. Sneezing was
most efficient in dispersing the bacteria, coughing next most
efficient. He made consumptives cough on to glass, and from
four patients he thus obtained abundant tubercle bacilli.

Heymann (1901) carried this further. He first made experi-
ments to determine the local dispersion and limitation of the
sputum drops. A patient was placed for i£ hour in an experi-
mental chamber in which plates were arranged in different
positions to receive droplets. After the patient had left the
chamber it was carefully closed and protected from sunlight for
some hours. The deposits on the plates were then examined by
inoculation experiments. In the case of a patient who used a
handkerchief before his mouth when coughing, it was found that
out of 36 animals inoculated with material from plates taken out
of the chamber after its use, n, or 30^5 per cent., were infected ;
and that of 34 animals inoculated from plates taken out of a
chamber where the patient did not use a handkerchief, 24, or
70-5 per cent., were infected.

SOURCES OF INFECTION

Most of the spray droplets when coughed up by the patient
were of a size which made them fall directly on to the glass plates
at short range. Some of the finer droplets, however, were easily
carried behind the patient by currents of air.

Six experiments were then made, handkerchiefs being held
from 5 to 10 cms. (2 to 4 inches) away from the mouth of the
coughing patients. Nearly half of the animals inoculated from
plates exposed under these conditions escaped infection.

The experiments showed that infective particles are rarely
carried more than i metre (39-4 inches) beyond the person
coughing, so that protection against spray infection is easy to
secure by keeping a distance of about an arm's length from the
patient, and by the latter using a handkerchief when coughing or
sneezing.

Experiments were also made by Heymann on the duration
of suspension in air of droplets containing tubercle bacilli. A
consumptive was made to cough into an experimental chamber
containing twelve covered plates, the covers of which were then
by mechanical means removed and replaced at definite intervals.
By these and other experiments it was proved that the duration
of suspension in air is not great, and consequently the amount of
infection thus received — except under conditions of the closest
intimacy — must be very small. The larger size of many of the
droplets diminishes the duration of suspension in the air. Hey-
mann next draws attention to the adhesiveness of such droplets
as have settled. He says : —

If these drops are allowed to dry for a short time on acid plates, they
can be rubbed fairly energetically with rough rags without the drops
being entirely removed. This fixation would become more definite if
the drops had settled on a fairly thick layer of dust ; and with the cleaning
methods, e.g. damp dusters, etc., employed in sickrooms, it is improbable
that much danger exists of infective particles being again raised into the
air.

He then investigated the duration of vitality of tubercle
bacilli in spray deposited on plates from sputum ejected by an
artificial spray apparatus and by patients in coughing. In
all, 96 plates were prepared, and were kept from 12 hours to
90 days. It was proved that of the tubercle bacilli from the
natural spray those kept in the dark lost their virulence within
18 days at the most, and those exposed to the light within

94 THE PREVENTION OF TUBERCULOSIS

3 days. The artificially sprayed tubercle bacilli kept in the
dark were virulent only for 7 days at the most.

The formation of pulverised sputum and its power of remain-
ing suspended in the air were next investigated. Experiments
were made showing that in quiet air after carpet -beating, etc.,
the suspension of bacilli in the air was very short. In moving
air, dust could not be detected ten minutes after the cessation
of the beating and brushing. Heymann indicated the defects
in Cornet's researches on dust infection. The number of experi-
ments in which droplet infection could be excluded with certainty
was, according to Heymann, not great ; and Cornet's technique
allowed of the inhalation of coarser particles and of adherent
droplets, as well as of the fine dust, which alone would be inhaled
under natural conditions. Heymann narrates a number of ex-
periments, in which he claims that these possibilities of error
were excluded. The number of tests made was 59, and 5 of
the inoculated animals, or 8-5 per cent., were infected with
tubercle. Heymann concludes : —

It is consequently demonstrated that dry dust containing tubercle
bacilli is only present in slight quantity in rooms of consumptives. The
low percentage in his results in comparison with Cornet's was striking,
so that a repetition of the experiments using Cornet's spongelet method
was thought worth making.

The adoption of this method of collecting the dust gave
a greater proportion of positive results, 15-8 per cent, in private
rooms, and 403 per cent, in hospitals. These results showed
Heymann that infective particles may be transported by contact
and dust, and deposited at a considerable distance from patients,
but that, as a rule, they fall and adhere, being generally too
heavy to be blown about.

Adding together Heymann's two sets of dust experiments,
the total results were as follows : —

Of a total of 239 dust samples obtained from the sickrooms of con-
sumptives, 44 contained virulent tubercle bacilli (=18-4 per cent.). In
the 123 obtained from hospital wards, 30 contained the bacilli ( = 24*
per cent.). In the 116 from private houses occupied by consumptiv<
only 14 ( = 1 2 per cent.) were infective. The hospital incidence was greate
than that in the homes of consumptives, whereas in Cornet's experiment
the incidence in the two was nearly equal.

In summing up the conclusions to be derived from his ex-

SOURCES OF INFECTION 95

haustive investigation, Heymann is of opinion that spray and
dust infection are equally important, one form taking pre-
cedence over the other, according to circumstances. When
spray infection persists for a considerable time, the patient's
environment must contain much infective material, but obviously
it varies with the stage of disease, and has the limitations
of vitality elsewhere indicated (p. 104). As a rule, infective
material is not sprayed further than an arm's length. The
duration of suspension of droplets in the air is limited, but
they have been found as long as half an hour after the last
attack of coughing ; droplets floating for so long a time as
this contain only a few tubercle bacilli. Heymann adds : —

Under natural conditions droplet infection is only operative in cir-
cumstances of closest intimacy, in the close intercourse of married people
and of mother and child ; among attendants on the sick, and in factory-
rooms, workshops, and offices.

Tubercle-containing dust particles are produced by the escape of
sputum droplets, and by remnants of sputum which may adhere to the
hand, pocket-handkerchief, bed-linen, carpets, and furniture, and especi-
ally to the floor as the result of spitting. I differ from Cornet in that I do
not attribute a greater power to this dust than to spray in producing
infection, because to enable infection to be produced the particles of
dust should possess an exceedingly fine consistency, enabling them to be
moved by even slight air currents. This they do not possess. The closely
adhering dust precipitated in sickrooms was found to contain only a few
tubercle bacilli ; and it may have settled down there in the course of
some days, so that these scanty positive results of investigation of the
dust afford no positive measure of the danger of inhalation of infective
dust. Under special conditions, in factories and workshops and on rail-
ways where numbers of human beings crowd together and cause con-
siderable agitation of the air, fine dust is formed, which may produce
infection derived from long deposits of phthisical sputum.

INFECTION DURING SPEAKING.— Fliigge and others, after
rinsing out their mouths with broth cultures of B. prodigiosus,
have found that the bacilli could be caught on culture plates
in different parts of the room, some of the plates which
were placed behind the speaker giving positive results. It
would be improper to infer from these experiments that similar
dissemination of tubercle bacilli occurs when consumptives
are speaking. As Cornet has said (p. 501) : —

When Fliigge takes cultures of the prodigiosus into his mouth,
determines that the germs are distributed in talking and coughing, and

96 THE PREVENTION OF TUBERCULOSIS

from this argues that the same occurs in the case of the tubercle bacilli
he neglects the most important link in his evidence, the tertium compara-
tionis, namely, the proof that the saliva of consumptives contains any-
thing like the same number of germs as when the mouth is filled with
a culture of prodigiosus. Researches upon this point show that the
saliva is either free from the bacilli or contains them in rare cases and in
small numbers.

Tubercle bacilli are few in number or absent from the mouth
of a consumptive except when coughing. Furthermore, the
viscous expectoration is much less easily scattered than watery
saliva.

CORNET'S EXPERIMENTS AND CONCLUSIONS. — According to
Cornet (Cornet, p. 98), Tappeiner first showed conclusively that
infection occurs by means of dust. Tappeiner infected dogs by sub-
mitting them to the inhalation of powdered tuberculous expec-
toration. Koch, Cornet, and others repeated these experiments,
substituting pure cultures of the tubercle bacillus for dried
expectoration. Other investigators with similar methods failed
to infect the animals experimented on. Hence Baumgarten
and more recently Fliigge have minimised the importance of
infection by the inhalation of dried expectoration. Their failures,
however, in the opinion of Cornet (Cornet, p. 102) were due
to a technique, faulty in departing from the natural conditions
governing infection by inhalation. The animals in their experi-
ments had been placed in closed cages, in the air of which dried
powdered expectoration was made to circulate by mechanical
means. But, as pointed out by Cornet, expectoration is very
hygroscopic, and at once under the above conditions absorbs
the respiratory moisture, becomes heavy, and is no longer borne
along in the inspiratory current of air.

In a series of experiments made in 1898 Cornet set himself
to imitate experimentally the conditions which would be found
in the dwelling of an unclean consumptive. In a room contain-
ing about 99 cubic yards of space he scattered over the carpet
dried tuberculous expectoration mixed with dust, and placed
guinea-pigs, some on the floor, and others upon stages 2 to
inches, 16 inches, and 4 feet above the floor. Then the flooi
was swept in the usual way with a stiff broom, so that a den<
dust was produced. Cornet protected himself by wearing
overall coat, and over his face a complete hood with protect e<
glass openings. A second group of animals was subjected t<

SOURCES OF INFECTION 97

direct inhalation of infected dust. Of 48 guinea-pigs used,
46 became infected. Neisser (Lartigau, p. 130) showed by
other experiments that mild currents of air can carry tubercle
bacilli from place to place, and that dried tubercle bacilli can
be held for some time in the suspended dust of ordinary rooms.

Cornet (Cornet, p. 502) quotes B. Frankel's proof that the
number of bacilli disseminated by coughing is insignificant as com-
pared with the number released by the drying of expectoration.
He let a number of consumptives wear masks for twenty-four
hours at a time, and with 219 of these masks he caught 2600
tubercle bacilli in 32 days. Compare with this the 300 million
bacilli which Heller estimates to be present in a single pellet
of expectoration. This would mean 7200 million in one day,
assuming the expectoration to occur only once an hour. Thus
one consumptive in one day may discharge in expectoration
7,200,000,000 bacilli ; a number of consumptives in 32 days
discharged by coughing 2600 bacilli. It does not follow that
the relative danger from dust and from spray is in the pro-
portion of these figures ; the proportion of each which, while
still virulent, reaches the mucous surface of a susceptible person
has to be considered. There are no means of stating this ;
it will vary with circumstances. Probably dust infection is
greater than spray infection in industrial and social life ; dust
infection bears a smaller proportion to spray infection in domestic
than in extra-domestic life ; but the evidence does not show with
certainty that under either set of circumstances spray infection
operates to a greater extent than dust infection. Whatever
be the proportion between the two, practical precautions must
take cognisance of both methods of spread.

IMPORTANCE OF DUST INFECTION. — Whatever be the pro-
portionate share of infective dust and infective spray, it is
certain that dust plays an important part in spreading tuber-
culosis. There is abundant evidence that the dust in the
vicinity of consumptives contains frequently, while that from
other localities seldom contains, tubercle bacilli. Cornet in 1888
(Cornet, p. 86) first clearly established these important facts.
Having carefully excluded the possibility of infection from other
sources, he inoculated guinea-pigs with the dust obtained from
the walls and floors of sickrooms occupied by consumptives.
His results were as follows : —

98 THE PREVENTION OF TUBERCULOSIS

In 7 hospitals 38 tests were made, 94 animals being inoculated
with dust. Of this number 52 died from diseases other than
tuberculosis, 22 remained healthy, and 20, or 21*3 per cent.,
became tuberculous. In 3 asylums n tests were made, 43
animals being employed, of whom 16 died from other diseases,
14 remained healthy, and 13, or 39*4 per cent., became tuber-
culous. In 2 prisons 5 tests were made on 14 animals, all
with a negative result as to tuberculosis. In the dwellings
and workplaces of consumptives 62 tests were made, 170
animals being employed, of whom 91 died from other diseases,
45 remained healthy, and 34, or 20 per cent., became tuber-
culous. In a surgical ward 3 tests were made, 8 animals
being employed in each instance, with a negative result a<
to tuberculosis. In certain streets 14 tests were made ; 4:
animals were employed, and here again a negative result
to tuberculosis was consistently obtained.

The dust of rooms occupied by consumptives was regularb
virulent in the instances in which the patient had been in th<
habit of spitting into his handkerchief or on the floor ; it showe<
no evidence of virulence when the spittoon or spit-bottle
been regularly used. Cornet also found virulent tubercl<
bacilli in the dust of a room in which a consumptive had di<
six weeks previously. It should be carefully noted that the
samples of dust were taren by Cornet from places where
they had settled by gravity from the air, and in which direcl
pollution by tuberculous master, either coughed up or expee
torated, or by means of dirty fingers, cups, cloths, or otherwise,
was practically impossible.

Other observers have confi rmed these results. Dr. H. Coates'
researches, carried out under the direction of Professor Delepine
at Owens' College, are especially valuable. He found that
only two out of a large number of film preparations of dusl
prepared by him were tubercle bacilli discoverable. Cultiva-
tion methods were obviously out of the question, as othei
organisms grow so much more quickly than the tubercle bacilli.
Cornet's inoculation test was therefore used. Samples oi
dust were collected from situations in which dust had settlee
naturally from the air, and where there would be no likelihooe
of direct contamination with expectoration or by infecte
articles. Samples of dust were taken from each house from the

SOURCES OF INFECTION

floor, skirting-boards, walls, shelves, mantelpieces, etc. Three
classes of houses were examined.

I. Houses which were in a dirty condition, and in which
a consumptive patient was living who was taking no precautions
to dispose of his expectoration so as to prevent infection of
the atmosphere, but who spat freely on to the floor, or into his
pocket-handkerchief, etc.

II. Houses which were in a very clean condition, but in
which a consumptive patient was living who was not sufficiently
careful as to the disposal of his expectoration.

III. Very dirty houses, in which there had been no case of
tuberculous disease for at least three years past.

The following table shows the results obtained : —

TABLE XX

Class I.
Dirty Houses containing Consumptives
who Used no Precautions.

Class II.
Clean Houses con-
taining Consumptives
not sufficiently
Careful.

Class III.
Dirty Houses
in which
Consumptives
had not Lived.

The number of houses from which
dust was examined was . . . 23
The number to be excluded because
the inoculated animals died rapidly
after inoculation was ... 2
The number found infective by inocu-
lation (one by microscopic examina-
tion only) was . . . . 14
Thus the percentage of infected houses

10
0

5
50 'o

10
0
0

The average size of the infected rooms

336 c. ft.

The average size of the non-infected
rooms was .... 368 c. ft.
The lighting and ventilation was —
Good in 5 positive and
7 negative cases
Fair in I positive and
I negative case
Bad in 8 positive and
I negative case
Samples were taken at different levels
in 16 houses
Of these samples the number found
infective was . . . . 13
Of the infective samples the number
near the floor was ... 9
4 to 6 feet above the floor was . . 13

506 c. ft.

In I positive and
5 negative cases
In 2 positive and
O negative case
In 2 positive and
o negative case

...

The preceding results indicate that there is no necessary

ioo THE PREVENTION OF TUBERCULOSIS

relationship between cubic space and the number of tubercle
bacilli in a room. The second series shows that ordinary
cleanliness does not alone suffice to prevent the accumulation
of infectious material in the rooms occupied by a consumptive.
The third series shows, so far as a short series of experiments
can, that tubercle bacilli are not present except in the immediate
environment of consumptives. The results obtained in further
experiments are interesting.

Five specimens of dust were collected at various elevations
from the walls of the waiting-room of the out-patients' depart-
ment of the Hospital for Consumption in Manchester. This
waiting-room is a lofty, well-lighted, and well- ventilated hall,
used by 180 patients every morning. Ten guinea-pigs were
inoculated and killed five weeks afterwards. None of them
showed any signs of tuberculosis.

Five samples of dust were also examined from the waiting-
room of one of the large general hospitals, and here also the
results were negative.

Dust taken from railway carriages failed to produce tuber-
culosis, but two samples taken from a general waiting-room
at a railway-station both produced tuberculosis.

Tubercle bacilli have been frequently found in the dust of
railway carriages, omnibuses, and tram-cars.

CHAPTER XIII

CIRCUMSTANCES LIMITING THE AMOUNT OF INFECTION
BY DUST AND SPRAY

i. T IMITED OPPORTUNITIES FOR INFECTION.— We have seen
\^_j that on the assumption that each annual death from
phthisis implies the constant presence in the general
population of three infective cases of the same disease, one in
every 263 of the population of England and Wales is infective, the
highest proportion being at ages 35 to 55 (p. 63). Even if we
assume that ten instead of three infective phthisical patients
are constantly present in the population for every death from
phthisis, the proportion will only be i in 79 of the total popula-
tion. Probably from the point of view of active infectivity three
years is a much more likely duration than ten years.

There is little if any foundation for the loose statements as to
the ubiquity of the tubercle bacillus. It is true that one-twelfth
of the total deaths from all causes are due to phthisis (p. 8), and
that at certain ages as many as half the bodies of persons having
died from other diseases have been found to present old healed
or latent tuberculous lesions (p. 48). One cannot, however, argue
from these data that at any given time a large proportion of the
population are capable of infecting others with tuberculosis.
The figures need to be considered, not in relation to deaths from
other causes, but in relation to the total population ; and when
this is done, the proportion of phthisical persons, on the three
years' basis stated above, is only i in 1881 of the children aged
5-10 years, i in 1129 of the children aged 10-15, and i in I4i'pf
adults aged 25-35.

2. NOT EVERY CONSUMPTIVE is INFECTIOUS, AND A CON-
SUMPTIVE is NOT INFECTIOUS THROUGHOUT THE WHOLE OF HIS
ILLNESS. — Careful patients do not endanger those with whom
they live or work. The experiments recorded on pp. 98 and 100
show that in rooms where consumptives use the simple pre-

102 THE PREVENTION OF TUBERCULOSIS

cautions required, the dust is free from infective material. (On
this point see also pp. 91 and 92.) The experience of hospitals for
consumptives appears to confirm the same conclusion. Those
patients who habitually swallow their expectoration — and this
includes nearly all children and lunatics — are relatively harmless
except to themselves, assuming that the excreta are properly
disposed of.

Many consumptives again have no expectoration during a
large part of their illness ; and in many others repeated examina-
tion fails to detect tubercle bacilli. Thus of 326 undoubted
cases of phthisis treated in the Brighton Borough Sanatorium
during the three years 1903-05, 195, or 59-8 per cent., had tubercle
bacilli in their expectoration during their stay in the sanatorium ;
80, or 24*5 per cent., had throughout expectoration showing no
tubercle bacilli; and 51, or 15*7 per cent., had no expectoration
at all. Most of these cases had either consolidation or cavitation
of the lungs. Of course the failure to find tubercle bacilli in the
expectoration of one-fourth of the total patients does not prove
their entire absence in these cases ; and it is likely that in some
of these cases inoculation experiments would have given positive
results. It is almost certain, however, that a considerable pro-
portion of the total cases, in addition to the sixth part who had
no expectoration, were not a source of infection while under
treatment, and probably not in a large part of the rest of their
illness. It should be added that in nearly all the above cases
three specimens of expectoration were examined before a negative
return was made. On the other hand, Sir Hugh Beevor (1905),
when examining the expectoration of 100 cases of phthisis
(32 cavity cases and 68 without discoverable cavity), found that
tubercle bacilli were absent in only about 15 per cent.

The annual report of the Mount Vernon Hospital for Consump-
tion for 1907 contains valuable data as to examination of sputum
of patients, from which the table on the following page has been
prepared.

Thus of the total 678 patients 10 per cent, had no sputum
while in the hospital, and of the 608 who had sputum 33 per cent,
while in the hospital had no tubercle bacilli on repeated examina-
tion.

3. CONSUMPTIVES DIFFER GREATLY IN INFECTIVITY.— It has
already been mentioned that when there is no expectoration

LIMITING THE AMOUNT OF INFECTION 103

the danger of infection is absent, whilst when the expectoration
is swallowed the danger is only to the patient himself. It may be
taken as a rough guide, that (i) the danger varies with the amount
of expectoration. This is not certainly true, and not always
true. Abundant purulent expectoration may show no tubercle
bacilli, and scanty expectoration may teem with them. The rule
may, however, be taken as a useful practical guide, and it follows
that advanced cases of phthisis in which expectoration is abundant
present greater possibilities of infection than early cases (see also
p. 394) . It appears probable that the danger from advanced cases
may be greater than is implied by the above rule. Advanced
patients are weak and may be bedridden, and under these circum-
stances are less able carefully to control the hygienic disposal

TABLE XXI

(See p. 1 02 for Reference to this Table,}

Condition of
Patients.

Number of
Patients.

ercentage of Total
aving no Sputum.

umber of Patients
having Sputum.

Number whose
putum showed no
Tubercle Bacilli
on Repeated
Examination.

Percentage of
Expectorating
Patients in whom
o Tubercle Bacilli
were discovered.

PM-0

Infiltration of one

lobe only .

198

163

Infiltration of more

than one lobe,

but no cavitation

277

252

Cavities present .

203

193

of their expectoration than if they were less enfeebled. The
importance of careful and cleanly nursing at this stage needs to be
emphasised.

(2) The danger is great in proportion to the frequency of
expectoration, infrequent expectoration being much more likely
than frequent to be carefully deposited.

(3) The number of tubercle bacilli in the expectoration is
not a certain guide as to degree of infectivity. Dead tubercle
bacilli take the stain for microscopic examination as well
as living bacilli. Kitasato (quoted by Cornet, p. 83) proved
experimentally that the majority of tubercle bacilli in expectora-
tion or in cavities are already dead. When therefore Cornet
gives a calculation showing that a single patient may expectorate

104 THE PREVENTION OF TUBERCULOSIS

daily 7200 million bacilli, and Nuttall that a patient with moder-
ately advanced disease and expectorating from 70 to 130 c.c. daily
may discharge daily from i£ to 4^ billions of bacilli, and when
Bellinger estimated that i c.c. (about a quarter of a teaspoonful)
may contain 810,000 to 960,000 bacilli, it must not be assumed
that these are all living bacilli. Living bacilli will probably be
present, quite sufficient to do mischief if the opportunity arises,
but the possibilities of mischief are not so great as might at first
be supposed.

4. VIRULENT BACILLI HAVE A LIMITED EXTRA-CORPOREAL
EXISTENCE EVEN WHEN LEFT ALONE. — In streets they cannot
(p. 331) be found except in expectoration itself. In dwellings they
have a more prolonged vitality, but according to Cornet infective
material has usually disappeared from a dwelling after about six
months. It is therefore, in all probability, an exaggeration to
speak of a house as being saturated with the infection of years.
One scarcely needs to add that it would be folly to trust to the
slow processes of nature for removing infection, when by disinfec-
tion and cleanliness this can be secured at once.

5. ONLY A FEW BACILLI REACH THE EXPERIMENTALLY DETER-
MINED DURATION OF EXTRA-CORPOREAL EXISTENCE. — Direct
sunlight kills them quickly (p. 53), being a disinfectant without
peer. The dispersion produced by air currents minimises any
subsequently received dose of infection, while street cleansing and
the more effective scavenging produced by rain sweep infectious
material into the sewers. It must be repeated that these factors
are mentioned, not with the idea that we can afford to rest content
with their operation without stopping indiscriminate expectora-
tion, but to prevent exaggerated notions as to the possibilities
of infection.

6. THE DISSEMINATION OF THE INFECTIOUS MATERIAL DIS-
CHARGED BY CONSUMPTIVES is LIMITED BY ITS PHYSICAL CHAR-
ACTER.— If the patient and his attendants and friends take the
simple precautions required to prevent spray infection during the
act of coughing, no immediate danger attaches to the expectora-
tion. A lump of expectoration in its wet condition is absolutely
incapable of spreading infection, except in the unlikely events
of its smearing the hand, or being carried by flies or otherwise,
and thus leading to the infection of food or of the cavity of the
mouth directly. The tubercle bacilli are as safely imprisoned

LIMITING THE AMOUNT OF INFECTION 105

in the lump of expectoration as they would be in a bottle.
Evaporation of the watery part of the expectoration is not
accompanied by any escape of tubercle bacilli. Currents of air
similarly have no effect. The bacilli cannot leave the expectora-
tion so long as it is moist. Expectoration is not only moist
but also viscid, and thus the tubercle bacilli often remain im-
prisoned, even after all moisture has evaporated ; and sweeping
or rubbing with boots, etc., is required to convert the expectora-
tion into a condition of such dryness that its dissemination as
dust becomes practicable. (On this point, see p. 92.)

Even when expectoration becomes dust, and the particulate
infective material can be scattered, it obeys the laws of gravity
and tends to sink again after being disturbed. Hence a room
which is very infective while sweeping is going on or soon after-
wards may be occupied with relative safety an hour or two later.
Tyndall's experiments demonstrating how particles of dust settle
out of quiet air have clearly shown this. It must be repeated
that it would be unreasonable to trust to the physical laws which
minimise the risk of infection, and not to insist on the cessation
of indiscriminate expectoration and on the wet cleansing of all
occupied rooms and public places.

7. THE AIR EXPIRED BY CONSUMPTIVES IN ORDINARY BREATH-
ING IS ABSOLUTELY STERILE (see also p. 89).

8. The circumstances which limit the amount of infection
by presenting opposing forces to the invading bacilli will be
considered later, (Chapters XXII. to XXVII.)

CHAPTER XIV

THE PORTALS, OF INFECTION: A. INFECTION BY
INHALATION

APART from the ingestion of infected food, to be considered
later, the predominant means of infection are the
spray produced by the consumptive as he coughs or
sneezes, and the dust of his powdered expectoration. Where
do the tubercle bacilli thus received take root, and how d<
they reach those parts of the body in which the main lesioi
of tuberculosis are found ?

They enter the body by the mouth or nostrils, and eithei
(a) are passed through the mucous membrane of the mouth 01
naso-pharynx into the adjacent lymphatics ; or (b) are swallowe<
and lodge in the intestines and the mesenteric glands connectee
with them; or (c) are inhaled into the lungs. From any one
of the points thus reached the tubercle bacilli may and common!]
do pass on to other parts of the body. Lesions thus occur a1
definite points, but there is no need for the supposition that one
part of the body is more susceptible than another to tuber-
culosis. The lungs and the mesenteric glands, so far as
know, suffer more than other parts only because they are more
exposed to invasion.

That tubercle bacilli are inhaled by persons in contact wil
consumptives, or by animals subjected to experiments witl
tuberculous dust, has been repeatedly shown. Strauss foune
tubercle bacilli in the nasal cavities of various healthy persons
frequenting the wards of the Charite and Laennec Hospital
in Paris ; of 29 persons employed in consumptive wards 9, oi
whom 6 were orderlies, gave positive results when tested b]
inoculation on guinea-pigs. St. Clair Thomson (1901) showee
that in the healthy nose most of the bacteria inhaled are immedi-
ately stopped at the nostrils (see also p. no). He quote
Liaras as having repeated Strauss's experiments under similai

106

THE PORTALS OF INFECTION 107

conditions on eighteen persons, but with precautions to secure
cultures in each case from the interior of the nose and not from
the nostril; the results were negative in each case. Notwith-
standing the discrepant result of these observations, there is
overwhelming evidence, both clinical and experimental, that
tubercle bacilli may be inhaled and find their way by direct
or indirect routes to the lungs. The subject may be conveniently
discussed under the following heads : —

1. By what means can the inhaled bacilli be checked ?

2. At what points do the bacilli enter the tissues of the body ?

3. What is the evidence that in phthisis the infection some-
times reaches the lungs by inhalation, and not always indirectly
by the lymphatic or blood circulations ?

The general rule is that at whatever spot on or in the tissues
of the body tubercle bacilli succeed in resisting phagocytic
and other inimical agencies, there or in lymphatic glands con-
nected therewith will tuberculosis develop. The usual course
is for the tubercle bacilli to pass through the surface on which
they have become deposited, and to be carried thence by the
lymph stream. The lymphatic glands may act as filters pre-
venting the tubercle bacilli from spreading to other parts of
the body ; just as glands in the armpit may prevent general
blood poisoning from a whitlow. Such carriage by the lymph
stream is slow and largely barred by the glands. Rapid trans-
port to more remote parts of the body can occur only when the
bacilli have gained access to the blood vessels and are carried
with the blood circulation. Then general or so-called miliary
tuberculosis occurs, a relatively rare and late phenomenon in
the disease.

MEANS BY WHICH THE INHALED BACILLI CAN BE CHECKED.—
i. The Complexity and Shape of the Respiratory Passages.— Angles
are met with in the nostrils, nasal cavity, pharynx, glottis,
trachea, and bronchi, and at every successive angle the inhaled
dust is filtered off. With quiet breathing, the greater part is
stopped in the nostrils.

2. The high Reflex Irritability of the Nasal and Pharyngeal
Mucous Membrane.— The irritation produced by the presence
of foreign particles may be so great as to cause sneezing and
consequent expulsion of the offending particles, together with
others too small to offend.

io8

THE PREVENTION OF TUBERCULOSIS

3. The respiratory passages are lined with a coat of mucus
and the individual cells are provided with cilia flicking
particles upwards towards the outlet. By this means a steady
flow of mucus towards the pharynx is maintained, and a similai
flow along the nose. Accumulated dust is thus swept into
position from which it can readily be ejected. Should the
bacilli, notwithstanding the preceding impediments, succ<
in obtaining lodgment in any part of the mucous membrane
they have then to do battle with the phagocytes of the
and the antibodies formed in connection with them.
victorious, the bacteria work their way into the underlying
lymphoid tissue and along the lymph channels, and establisl
a primary focus of infection.

POINTS OF ENTRY. — If infective dust or droplets have
passed the guarded portals of the mouth and nose, tuberclt
bacilli may penetrate the mucous membrane of the back
the nose, of the tonsils or larynx, of some lower part of tl
respiratory tract, or through decayed teeth. An obvioi
lesion may not develop at the point of penetration. This
been shown by Sidney Martin in the case of animals fed wit!
tuberculous milk, a local ulcer being developed only whei
massive infection has been received; while only the subjacenl
lymphatic grands showed disease when the dose of infectr
material was more minute.

ADENOID GROWTHS, so common in the post-nasal cavities oi
children before puberty, favour the occurrence of infection ; foi
they narrow the passage for air and hinder the expulsion oi
particulate matter. Naked -eye evidence of tuberculosis
adenoids is seldom seen ; but many observers have shown b]
microscopic examination or inoculation that tubercle bacilli
often contained in adenoids. Thus G. Morgan (1899) foune
tubercle bacilli in from 12 to 15 per cent, of his cases of adenoie
in the substance of the morbid structure. Thomson (1901
gives a tabular statement of 1427 microscopic examinatioi
of adenoids, in 5-1 per cent, of which tuberculosis was found.
Dieulafoy similarly found tuberculous changes in 5*7 per cent,
of his case of adenoids ; and the proportion was increased b]
inoculation experiments to 20 per cent. It seems likely, there-
fore, that tubercle bacilli may enter at this point more often tl
is ordinarily supposed.

THE PORTALS OF INFECTION 109

The TEETH possibly may also be the point of invasion. Thus
G. W. Cook (quoted by Squire, 1906) found tubercle bacilli in
the pulp of decayed teeth and in scrapings taken from and
around the teeth, especially of the young.

The TONSILS probably play a considerable r6le as a primary
site of tuberculous infection. The act of swallowing tuberculous
dust or spray or food presses infective particles against the tonsils,
in the crypts of which the infective matter may lodge. Like
all lymphoid tissue, the tonsils are " on outpost duty, to arrest
the invading bacilli," and it is rather remarkable that active
tuberculous disease of the tonsils is -so seldom seen. Tubercle
bacilli are often present in the tonsils without any naked-eye
evidence of disease. On this point Latham (1900) has confirmed
by the inoculation method the work of Woodhead and many
others. He proved that the central portions of the tonsils of
forty-five consecutive children aged from 3 months to 13 years
showed evidence of tuberculosis in seven instances. Infection
through the tonsils is common in pigs. It is probably more
common in children than is usually supposed.

The LARYNX is only exceptionally the seat of primary tuber-
culosis, laryngeal implication being more often a symptom of
advanced pulmonary tuberculosis. The trachea and bronchi
are also seldom attacked, the inhibitory influences enumerated
on page 107 rendering the infection of these parts infrequent.

INFECTION OF THE SUBSTANCE OF THE LUNGS BY DIRECT
INHALATION is usually taught to be a frequent occurrence. We
must now consider in detail the evidence for and against such
direct inhalation.

I. The Intricacies of the Respiratory Passages. — It is not sur-
prising in view of these intricacies, and of the moisture and
other influences tending to deposit dust during inspiration, that
Cohnheim (1890) describes the air passages as forming a com-
paratively long and narrow, closed and protected tube system ;
while Virchow long upheld the view that dust could not find its
way into the ultimate lung substance (quoted by Arlidge, 1892),
arguing that the black pigment found in miners' lungs was due
to altered blood pigment and not to carbon. In 1866, however,
he was convinced that his former views on this point were
incorrect.

Against these mechanical difficulties must be set the facts

no THE PREVENTION OF TUBERCULOSIS

that during hard work breathing becomes more rapid and more
laboured, and that the mouth is apt to be open; furthermore,
that inspiration takes place over 20,000 times in the twenty-four
hours, and often occurs in a very dusty atmosphere. Under
these circumstances it need not be the subject of surprise that
the defensive arrangements are occasionally overworked and fail
to prevent invasion by infective dust.

2. Experimental Evidence. — St. Clair Thomson and Hewlett
(1895) having ascertained that at least 1500 organisms are inhaled
into the nose every hour, and that in London it must be common
for 14,000 to enter in an hour of quiet breathing, nevertheless
found that the interior of the great majority of normal nasal
cavities is perfectly aseptic (p. 106). They also confirmed Hilde-
brandt's experiments made in 1888, in several instances the
trachea of animals killed in the laboratory being found on opening
to be free from bacteria.

On the other hand, Zenker (quoted by Arlidge, p. 246) pro-
duced red colouring of the substance of the lungs of animals by
causing them to inhale a red dust ; and Knauff (quoted by
Buck, p. 29), after inhaling particles of ultramarine for only ten
minutes, found that the cells of his expectoration contained blue
particles in their interior. In ultramarine workers the coloured
dust has been recognised in expectoration fourteen days after
cessation from work. Rabbits confined in a smoky atmosphere
can be shown to have fine particles of carbon in their bronchi.
Knauff (quoted by Greenhow, 1869) placed dogs for from one day
to three months in a roomy chest, into which the fumes of a smok-
ing oil-lamp were conveyed by a flue opening through the floor.
One dog killed after a single day in the smoke chest had the
whole surface of the bronchial mucous membrane even to the
alveoli of the lungs covered with a deposit of carbon mixed with
mucus. Animals kept there for some weeks showed similar
deposits throughout the lungs ; the lymphatic glands were very
early affected. In animals confined for several weeks in the
experimental chest there was almost invariably a deposit of
carbon below the pleura. Control animals showed no similar
appearances.

It must be admitted, however, that none of these experiments
is quite inconsistent with the view that the particles of pigment
had been swallowed and reached the lungs by means of the

THE PORTALS OF INFECTION m

lymph stream ; and the view that the pigment in miners' lungs
and similar diseases owes an intestinal origin has in recent years
been revived by the French school, especially by Villoret. Van
Steenberghe and Grysez fed guinea-pigs and rabbits with food
containing mixed coal dust and particles of Indian ink, finding
at the autopsy on these animals pigment in the lungs only, the
abdominal organs and mesenteric glands being free. Schultze
(1906, Munch, med. Woch., liii. 1702) repeated these feeding
experiments with similar results, but he is convinced that in
feeding experiments, even when undertaken with the aid of a
tube, inhalation cannot be excluded, and he explains in this
way the deposit in the lungs. That this may be the correct
explanation is supported by the fact that in a rabbit having
a gastric fistula, through which he introduced pigments into the
stomach daily for two months, no deposit was found post-mortem
in the lungs. The experimental evidence, in short, cannot be
said to have settled the question.

3. Microscopic Evidence.— According to Rindfleisch (1875,
p. 649), the first lesion in pulmonary tuberculosis occurs at the
angles and projections situated where

the smallest bronchioles become continu-
ous with the acini. This can be readily
understood from Fig. 12, if it be as-
sumed that the tubercle bacilli have
been inhaled into the acini. During
coughing they will become lodged in the
crannies around the opening of the bron-
chiole (a), and disease consequently may
start here. The diameter of a minute
branch of the bronchus at a is from 0-3 atais the junction of the

to 0'4 mm., as compared with -0015 to bronchiole with the acinus

•004 mm., the size of a tubercle bacillus.

4. Clinical experience supports the view that direct inhalation
of infective particles into the lung substance is at least excep-
tional. In 1868 Mr. (now Lord) Lister showed that suppuration
did not follow when air had escaped into the pleura through
injury of a lung by a fractured rib, thus indicating that the
inspired air is probably sterile. His exact words are as follows: —

Why air introduced into the pleura through a wounded lung should
have such totally different effects from that entering through a per-

ii2 THE PREVENTION OF TUBERCULOSIS

manently open penetrating wound from without, was to me a complete
mystery till I heard of the germ theory of putrefaction, when it at once
occurred to me, though we could not suppose the gases of the atmosphere
to be in any way altered in chemical composition by passing through
the trachea and bronchial tubes on their way into the pleura, it was only
natural that they should be filtered of germs by the air passages, one of
whose offices is to arrest inhaled particles of dust, and prevent them from
entering the air cells.

5. The relative infrequency of tuberculosis of the larynx is
adduced as evidence of the completeness with which nitration
of the inspired air is effected in the naso-pharynx. St. Clair
Thomson (1901) found in 100 autopsies in pulmonary tuber-
culosis that only 30 had laryngeal disease ; and in another series
that only i in 450 had tuberculous nasal disease. But, as already
explained, the relative immunity of the larynx is probably due to
the freedom of movement of its parts, the violent coughing ac-
companying local irritation in it, and the active secretion of
fluid washing away invading particles. Primary tuberculosis of
the larynx occurs sometimes, but it is the exception.

The evidence briefly summarised above is conflicting. In
view of what we know to occur in knife-grinders and in lead and
slate miners, as well as of the evidence given above, the balance
leans to the conclusion that direct inhalation of dust into the
lungs occurs. Such dust, if it carries with it the tubercle bacillus,
may be regarded as an inoculating needle, securing a firm foot-
hold for the bacillus in the pulmonary tissues.

INFECTION OF THE LUNGS OTHERWISE THAN BY DIRECT
INHALATION. — Though it be agreed that the lungs may be in-
vaded directly during inhalation, this is certainly not the only
means of infection. The lungs may also be infected secondarily
through the following channels : —

(a) Through the bronchial glands. Tuberculous material is
arrested at the tonsils or elsewhere, and the bacilli pass to the
cervical and bronchial glands by the lymph stream. Sims
Woodhead's experiments (1898) on a series of pigs fed with milk
containing tubercle bacilli throw light on this question. Th<
line of invasion could be traced in these pigs from the tonsi]
and lymphoid tissues of the throat to the neighbouring lymphati<
glands along the neck ; thence to the upper part of the chesl
to the glands at the root of the neck and the pleura. In tl
connection must be noted the frequency with which, in mai

THE PORTALS OF INFECTION 113

pleurisy precedes other signs of pulmonary tuberculosis. Wood-
head's conclusion is as follows : —

I am driven to the conclusion that this method of infection of the
glands of the neck through the tonsils must be a comparatively frequent
occurrence, especially in children under insanitary conditions, and sub-
jected to various devitalising influences.

There can be little doubt that the infection may spread
downwards to the bronchial glands and then into the lungs, and
that this is a fairly common method of infection, especially in
children. That this is so is confirmed by the fact that in children
the parts of the lungs near their roots are often most affected by
tuberculosis. According to H. Walsham (1904),

it is still an open question whether or not the lung can be infected by the
gradual extension of the bacilli downwards with the lymph stream. I
think in these cases where we find tuberculous change in the cervical
glands further advanced than in the bronchial, we may assume that
the lung has been infected in this manner.

Case 4, p. 66, is probably one of phthisis originating in this
way. It is not unlikely, however, that as in the case of intestinal
infection (p. 116) the first chain of glands, in this case the cervical,
may escape obvious involvement, the bronchial glands suffering
most.

The bronchial glands themselves may be infected from two
sources : (a) from the cervical glands, and probably from the
tonsils, as indicated above ; (b) from the alimentary canal.
Thus Woodhead has traced tuberculosis from a caseous or old
calcareous mesenteric gland through the chain of retro-peri-
toneal glands up through the diaphragm to the posterior
mediastinal and bronchial glands, and thence to the lungs.

According to Guthrie, to the above methods of access to the
mediastinal and bronchial glands must be added the possible
passage of bacilli, swallowed with mucus or food, through the
cesophageal lymphatic plexus to the posterior mediastinal
glands. Squire (1906) believes that the implication of the
bronchial glands is oftener produced in the reverse direction,
from lungs to glands, than is usually accepted.

(b) Through the alimentary canal. This will be considered
separately in the next chapter.

114 THE PREVENTION OF TUBERCULOSIS

by tubercle bacilli carried in the blood circulation. This circu-
lation of infective products undoubtedly happens in general
tuberculosis, as Buhl showed in 1857 (P- 37)- A caseous nodule
breaks down, its contents enter the blood vessels, are carried to
the heart and thence in the round of the circulation. It is likely
that a more localised distribution of infection occurs by the blood
vessels, when tuberculous material ulcerates into a blood vessel
in the lung, and the disease spreads with the blood current to
other parts of the lung. According to Volland (Cornet, p. 182),
pulmonary tuberculosis is produced by bacilli which have entered
the cervical glands and are carried thence within the leucocytes
by way of the lymph stream and the lesser circulation to the
lungs. We shall discuss later what means, if any, can be used
to determine whether a given fatal case of tuberculosis has been
caused by inhalation or ingestion ; and, if the latter, whether
through the ingestion of human or of bovine infectious material.

CHAPTER XV

THE PORTALS OF INFECTION : B. INFECTION BY
INGESTION

THE arguments for and against the direct invasion of the
lungs by inhaled particles have been given in the last
chapter. If direct infection by way of the lungs is escaped,
it does not follow that no infection occurs. As we have already
seen, the individual may be infected through the mouth, naso-
pharynx, or oesophagus. The next possibilitv of infection is
through the stomach. Little is known of this, as separate from
intestinal infection, and the subsequent course of the bacilli
would be almost the same in both instances. In passing we may
note

THE EFFECT OF THE GASTRIC JUICE ON SWALLOWED TUBERCLE
BACILLI. — Falk and Wesener exposed tuberculous material to
the action of an artificial gastric juice for some hours, and showed
that it had not lost its virulence when tested by inoculation on
animals. Strauss and Wurtz subjected pure cultures of the
avian tubercle bacillus to the action of a dog's gastric juice, and
found that at the end of eight to twelve hours the bacilli were still
able to produce local tuber culosis when inoculated on animals.
It must be remembered, however, that the fat-splitting enzyme of
gastric juice is very sensitive to its environment, and is destroyed
quickly when the juice is used in vitro. Probably the fatty
envelope of the tubercle bacillus would be more readily dissolved
within the stomach than in an experiment under artificial con-
ditions. Nevertheless in the stomach the digestive or inhibitory
effect of the gastric juice would be diminished by dilution with
food and fluid, and many tubercle bacilli would doubtless pass
on unharmed into the small intestine.

THE LESIONS PRODUCED BY INGESTED TUBERCLE BACILLI. —
Most of these follow on the passage of the bacilli through the
intestinal mucous membrane. The local effect on the mucous

n6 THE PREVENTION OF TUBERCULOSIS

membrane varies with the dose and the virulence of the bacilli,
and possibly with the age of the patient. Sidney Martin's
experiments in feeding pigs with tuberculous material showed
that there need not be a local development of tuberculosis at the
point of entry of the bacillus (see also p. 113), but that such
lesions occurred when major doses of a more virulent strain were
given.

It might be argued that in these cases infection had not come
vift, the intestine. Thus Cadeac (quoted by Miiller, 1905) believes
that in most feeding experiments, the tubercle bacilli enter in
the region of the mouth and pharynx. Having fed guinea-pigs
with material rich in bacilli, he killed them at the end of seven days,
and tested the glands of the head and of the mesentery by inocula-
tion, obtaining a negative result in the latter, a positive in the
former case. Miiller has found that in guinea-pigs fed with
infected milk the mesenteric glands may be primarily affected.

A. Calmette and A. Gu6rin experimented on young goats
suckled from their mothers' teats, which had previously been
made tuberculous by the artificial introduction of tuberculous
material into the mammary gland. They all acquired intestinal
tuberculosis, followed by mesenteric disease. Then a number of
adult goats were fed with tuberculous material by means of an
cesophageal tube. These all contracted grave and rapidly fatal
pulmonary tuberculosis, without obvious intestinal and with
only a few mesenteric lesions. They concluded that in adults
tubercle bacilli pass easily through the mesenteric lymphatic
glands to the thoracic duct, and thence through the heart and
pulmonary arteries of the lungs.

The Second Interim Report of the Royal Commission on
Tuberculosis (1907) gives the details of experiments in which
calves were fed with the milk of cows whose udders had been
made tuberculous by intra-mammary injection. It was found
that in only one out of six calves thus fed was general tuberculosis
produced, the tuberculosis in the others being confined chiefly
to the intestines and mesenteric glands. Fourteen cows fed with
tuberculous milk from various sources showed chiefly mesenteric
lesions. On the other hand, generalised progressive tuberculosis
was readily produced in monkeys by feeding them with tuber-
culous milk.

The experiments of Calmette and Guerin indicate that tuber-

THE PORTALS OF INFECTION 117

culosis of the bronchial glands and of the lungs may be the
result of feeding with tuberculous material, with or without
mesenteric disease ; but it appears likely that in human tuber-
culosis due to ingestion, implication of the mesenteric glands is
generally more abundant and more severe than that of other
parts of the body.

The age of tuberculous lesions is judged by the presence or
absence of caseation or calcification ; these signs being taken to
indicate an older lesion than tuberculous disease in which these
degenerative changes have not occurred. On the value of such
evidence in experimental animals, Professor Delepine (1898,
p. 734) may be quoted : —

There are very often clear indications in the body of the victim showing
the channels through which the bacilli have penetrated. We have seen
how the bacilli infect first the lymphatic glands nearest to their point
of entrance. The lymph coming from the intestine passes first through
the mesenteric glands. The lymph from the lungs passes in the same
way through the bronchial glands. It is therefore evident that in the
event of the bacilli penetrating through the intestine the mesenteric
glands would be chiefly affected, and in the case of lung infection the
bronchial glands would be most involved. There are cases in which
death occurs before any other glands than those first invaded have had
time to become diseased ; in such cases the state of the glands will clearly
indicate the channel through which the bacilli have entered.

In a series of over 300 experiments I have found that tuberculosis of
the mesenteric glands occurs extremely late in guinea-pigs infected through
other channels than the intestinal canal and the peritoneal cavity, and
am absolutely convinced of the value of lymphatic glands as indicators of
the path followed by tubercle bacilli in cases which have died before the
disease has become too advanced. According to Dr. Woodhead, the
post-mortem examinations of the bodies of tuberculous children who
had died before the age of five and a half years show that in the large
majority of them the intestine and mesenteric glands were affected, and
that in 14 per cent, of those cases the mesenteric glands alone were
tuberculous.

Notwithstanding somewhat discrepant results from experi-
ments, we may, I think, assume that the evidence of death-
returns and still more of post-mortem examinations, gives some
indication of the relative frequency of intestinal and of more
direct pulmonary infection.

AGE INCIDENCE OF DEATH-RATE FROM THE DIFFERENT FORMS
OF TUBERCULOSIS. — The following table, which I have calculated

THE PREVENTION OF TUBERCULOSIS

from the Registrar-General's returns, shows the age incidence of
the death-rate from the three chief forms of tuberculosis : —

TABLE XXII. — ENGLAND AND WALES, 1901
Death-rate per 100,000 Persons living at each Age-period

0-5.

5-io.

IO-I5.

15-20.

20 and

upwards.

Pulmonary Tuberculosis .

176

Tuberculous Meningitis

109

Tabes Mesenterica .

125

Even if a large deduction be made for errors of diagnosis and
certification in the returns of tuberculous meningitis and tabes
mesenterica, it still remains true that there is an inverse relation
between the age incidence of death from those two diseases
and that of death from phthisis. Whether, as adults take
much less uncooked cows1 milk than children, it may be inferred
with safety that respiratory infection is more common in adult
life and digestive infection in childhood, is still open to doubt.
An d priori probability to this effect is created ; but this is some-
what shaken by our knowledge of the different channels through
which the lungs may become infected. It is quite possible
that phthisis originating vi& the digestive tract may be more
frequent than the above table would indicate.

EVIDENCE FROM AUTOPSIES. — The evidence from autopsies
as to which are the oldest lesions is apt to be disturbed by the
fact that, no examination being possible until natural death
occurs, the bronchial and mesenteric glands may appear to be
implicated equally. Possibly also the changes may occur more
rapidly in certain lesions than in others. Thus in guinea-pigs
lesions advance more rapidly in lymphatic glands than in lungs.
Furthermore, as pointed out by H. W. Russell, lesions in lym-
phatic glands are more easily detected than equally large lesions
in a large organ like the lung. These sources of error possibly
explain[some of the discrepancies in the results of autopsies made
at different hospitals, of which the following are examples : —

Dr. L. G. Guthrie (1899) tabulated 77 post-mortem examina-
tions made on tuberculous children at the Paddington Children's
Hospital. He found tuberculosis of the various thoracic organs

THE PORTALS OF INFECTION 119

(lungs, pericardium, and pleura) in the aggregate of all the cases
examined 105 times, of the various abdominal organs (peri-
toneum, intestine, spleen, liver, kidneys, and pancreas) 102
times, of the brain and meninges 41 times, and of the bones and
joints 6 times. He notes the difficulty in determining the start-
ing-point of infection from the stage of the lesions produced ; but,
adopting the usual method of deciding the source of disease, he
found that of the 77 cases, thoracic tuberculosis was most pro-
minent and apparently primary in 42 (54'5 per cent.), and
abdominal tuberculosis in 19 (24*6 per cent.). In 7 of the re-
mainder (16) the thoracic organs were as much affected as the
abdominal. In 6 cases the origin was not discovered, and 3
single cases originated elsewhere. The thoracic glands were
found in a state of caseation 46 times, and the abdominal glands
31 times. Both sets were caseous in 15 cases ; in 3 neither set
was affected, and in 12 their condition was not noted. Thus
the glands were caseous in 62 cases, or 80*5 per cent, of the total.
Dr. Guthrie adds that he has not regarded mere caseation as
evidence of primary glandular infection, and that he could only
trace the origin of tuberculosis with any degree of certainty in
41-5 per cent, of the cases — to the thoracic glands in 17 cases
and to the mesenteric glands in 15.

Dr. Guthrie summarises other experiences as follows : —

MM. Rillet and Barthez found the origin in caseous bronchial glands
in 79 per cent, and in mesenteric glands in 46 per cent, of cases. Sim-
monds discovered caseous bronchial and tracheal glands in 73 per
cent, and caseous mesenteric glands in 46 per cent., whilst Dr. Walter
Colman attributed the origin to caseous thoracic glands in 79 per cent,
and to mesenteric glands in 66 per cent, of his cases.

He adds : —

The discrepancy between these statistics and my own may be due
to the fact that I have discarded the glands as the primary source of
infection unless they have been both obviously caseous, and also associated
with miliary, or at all events comparatively recent, tuberculosis else-
where.

Dr. Still (1899) concluded from post-mortem examinations of
269 children under 12 years of age that the most common
channel of infection in children is through the lungs ; that
infection through the intestine is less common in infancy than

120 THE PREVENTION OF TUBERCULOSIS

in later childhood ; and that milk cannot be the usual source
of infection. Dr. A. Latham, tabulating over 3000 post-mortem
results on children, says they show that in children tuberculosis
of the bronchial glands is the lesion most constantly found, and
that disease is in the majority of instances most advanced in
these glands. He deprecates the inference that infection has
necessarily been conveyed aerially, and considers that infected
milk supply plays an important r6le.

Dr. Kingsford (1904) has added further cases and tabulated
the results of previous observers in an excellent paper.

It would be easy to give further figures, but they are all
inconclusive. It cannot be regarded as settled, to what degree
human tuberculosis is due to direct inhalation into the lungs,
to entrance of infective material through the tonsils, etc.,
and to intestinal infection. Much less is this point settled for
pulmonary tuberculosis. For a large proportion of intestinal
may be and probably is secondary to pulmonary tuber-
culosis ; and tuberculous meningitis may be secondary to an
earlier focus of tuberculosis in any part of the body. Con-
versely, a large, possibly the largest, part of pulmonary tuber-
culosis may be due not to the direct inhalation of infective
material into the lungs, but to secondary implication of the lungs
from the neighbouring glands. And these glands or the pul-
monary disease itself may furthermore have been the nidus of
potential and eventually active pulmonary tuberculosis for many
years before the latter disease comes into active existence. The
evidence needs to be sifted with the utmost care in each indivi-
dual case. Even then, the final decision arrived at after a
careful balancing of all the available evidence cannot be regarded
as certain. But the same remark applies to a large proportion
of the broader problems of medicine ; and we are not relieved
thereby from the responsibility of adjudicating and of taking
practical measures based on our decisions. The obviously safe
plan is to guard against all the possible sources of tuberculous
infection that have been considered, though the greatest im-
portance must be attached to the prevention of the inhalation or
swallowing of dried expectoration or expectoration in the form
of spray.

CHAPTER XVI
RELATION OF BOVINE TO HUMAN TUBERCULOSIS1

IN the earlier attempts to diminish tuberculosis, the pre-
vention of infection by means of food bulked very largely.
The only foods which are of importance in this connection
are cows' milk and its products, and the flesh of the ox and
pig. Inasmuch as cows' milk is the chief possible non-human
source of tuberculosis in man, the question becomes in the main
one as to the relation between bovine and human tuberculosis.
The earlier view is summarised in the following remarks from
the Report of the Royal Commission appointed to inquire into
the Effect of Food derived from Tuberculous Animals (1895) : —

Par. 22. As regards man, we must believe — and here we find our-
selves agreeing with the majority of those who gave evidence before us —
that any person who takes tuberculous matter into the body as food,
incurs some risk of acquiring tuberculous disease. . . .

Par. 23. We regard the disease as being the same disease in man and
in the food animals, no matter though there are differences in the one
and in the other in their manifestations of the disease ; and we consider
the bacilli of tubercle to form an integral part of the disease in each, and
(whatever may be its origin) to be transmissible from man to animals
and from animals to animals.

In Par. 80 of the report of the same Royal Commission it
is stated emphatically that " no doubt the largest part of the
tuberculosis which man obtains through his food is by means
of milk containing tuberculous matter."

The views stated above were generally entertained by
Koch among others, judging by his statement in 1882 (Berliner
klin. Wochenschr., 1882, p. 230) that " bovine tuberculosis is

xThis and the next two chapters were written before the appearance of
the second Interim Report of the Royal Commission appointed to inquire into
the Relations of Human and Bovine Tuberculosis. Any modifications necessi-
tated by that important report are added in footnotes, or in special paragraphs.

122 THE PREVENTION OF TUBERCULOSIS

identical with human tuberculosis, and is thus a disease trans-
missible to man."

In 1901, Koch gave his famous address at the meeting of
the British Congress on Tuberculosis. In this address he said :—

This manner of infection is generally regarded nowadays as proved,
and as so frequent that it is even looked upon by not a few as the most
important, and the most rigorous measures are demanded against it.
In this Congress also the discussion of the danger with which the tubercu-
losis of animals threatens man will play an important part.

After excluding the tuberculosis of poultry, which differs
so much from human tuberculosis that it can be left out of
account as a source of infection for man, he added,

the only kind of animal tuberculosis remaining to be considered is the
tuberculosis of cattle, which, if really transferable to man, would indeed
have frequent opportunities of infecting human beings through the
drinking of the milk and the eating of the flesh of diseased animals.

After indicating the obvious impossibility of investigating
the problem by direct experiments on human beings, Koch
said : —

Indirectly, however, we can try to approach it. It is well known that
the milk and butter consumed in great cities very often contain large
quantities of the bacilli of bovine tuberculosis in a living condition, as
the numerous infection experiments with such dairy products on animals
have proved. Most of the inhabitants of such cities daily consume such
living and perfectly virulent bacilli of bovine tuberculosis, and unin-
tentionally carry out the experiment which we are not at liberty to make.
If the bacilli of bovine tuberculosis were able to infect human beings,
many cases of tuberculosis caused by the consumption of alimenta con-
taining tubercle bacilli could not but occur among the inhabitants of
great cities, especially the children.

His remarks on this point will need discussion later (p. 131),
but in the meantime we may quote his conclusion, which is that,

though the important question whether man is susceptible to bovine
tuberculosis at all is not yet absolutely decided, and will not admit of
absolute decision to-day or to-morrow, one is nevertheless already at
liberty to say that, if such a susceptibility really exists, the infection of
human beings is but a very rare occurrence. I should estimate the extent
of infection by the milk and flesh of tuberculous cattle and the butter
made of this milk, as hardly greater than that of hereditary transmission,
and I therefore do not deem it advisable to take any measures against it.

This important expression of opinion involved a re-testing
of the whole question of the relationship between bovine and

BOVINE AND HUMAN TUBERCULOSIS 123

human tuberculosis, and since Koch's address many have
been working at the problem. In England a Royal Commission
was appointed to inquire into the Relations of Human and
Animal Tuberculosis, and in 1904 it issued an interim report,
from which the following extract is taken : —

We have up to the present made use, in the above inquiry, ol more
than twenty different " strains " of tuberculous material of human
origin, that is to say, of material taken from more than twenty cases of
tuberculous disease in human beings, including sputum from phthisical
patients and the diseased parts of the lungs in pulmonary tuberculosis,
mesenteric glands in primary abdominal tuberculosis, tuberculous
bronchial and cervical glands, and tuberculous joints. We have com-
pared the effects produced by these with the effects produced by several
different strains of tuberculous material of bovine origin.

In the case of seven of the above strains of human origin, the intro-
duction of the human tuberculous material into cattle gave rise at once
to acute tuberculosis, with the development of widespread disease in
various organs of the body, such as the lungs, spleen, liver, lymphatic
glands, etc. In some instances the disease was of remarkable severity.

In the case of the remaining strains, the bovine animal into which
the tuberculous material was first introduced was affected to a less extent.
The tuberculous disease was either limited to the spot where the material
was introduced (this occurred, however, in two instances only, and these
at the very beginning of our inquiry), or spread to a variable extent from
the seat of inoculation along the lymphatic glands, with, at most, the
appearance of a very small amount of tubercle in such organs as the
lungs and spleen. Yet tuberculous material taken from the bovine
animal thus affected, and introduced successively into other bovine
animals, or into guinea-pigs from which bovine animals were subsequently
inoculated, has, up to the present, in the case of five of these remaining
strains, ultimately given rise in the bovine animal to general tuberculosis
of an intense character ; and we are still carrying out observations in this
direction.

We have very carefully compared the disease thus set up in the bovine
animal by material of human origin with that set up in the bovine animal
by material of bovine origin, and so far we have found the one, both in
its broad general features and in its finer histological details, to be identical
with the other. We have so far failed to discover any character by which
we could distinguish the one from the other ; and our records contain
accounts of the post-mortem examinations of bovine animals infected
with tuberculous material of human origin, which might be used as
typical descriptions of ordinary bovine tuberculosis.

The result at which we have arrived, namely, that tubercle of human
origin can give rise in the bovine animal to tuberculosis identical with
ordinary bovine tuberculosis, seems to us to show quite clearly that it
would be most unwise to frame or modify legislative measures in accord-
ance with the view that human and bovine tubercle bacilli are specifically

124 THE PREVENTION OF TUBERCULOSIS

different from each other, and that the disease caused by the one is a
wholly different thing from the disease caused by the other.

The preceding sketch of a few of the most prominent features
in the history of this moot point would not be complete without
noting that in 1896 Professor Theobald Smith first drew atten-
tion to certain differences between bacilli from human and
bovine sources, and in 1898 he classed human and bovine bacilli
as separate types or races. Although the evidence which he
advanced had been somewhat neglected until Koch published
the results of his limited series of experiments, the idea that there
are two types of tubercle bacillus bearing on human disease,
the Typus humanus and the Typus bovinus, is by no means new.

DIFFERENCES BETWEEN HUMAN AND BOVINE TUBERCULOSIS.
— i. Differences in Morphological Characters of the Bacilli. — The
bovine bacillus is more uniform and constant in form than the
human bacillus. It is thick, straight, and short, seldom more
than 2 p in length, and averaging less (Theobald Smith).
Human bacilli are larger from the start and tend to increase in
length at once in subculture. They are generally more or less
curved. These morphological differences tend to disappear in
the tissues of susceptible animals. The bovine bacilli stain
deeply with carbol-fuchsin, beading being nearly always absent
from young cultures and often from old ; human bacilli stain
less intensely with carbol-fuchsin, and beading is generalty
seen, even in early growths.

2. Differences in Growth in Media. — Bovine bacilli, according
to the same authority, grow more luxuriantly in artificial media
than human bacilli, especially in glycerinised broth.

3. Differences in Reaction. — Theobald Smith has

called attention to the difference in the movement of the reaction of
the glycerin bouillon in which bovine and human bacilli are multiplying.
In the case of the bovine cultures this movement leads to a final reaction,
either neutral, feebly alkaline, or feebly acid, toward phenolphthalein ;
in case of the human cultures to a pronounced acidity to phenolphthalein.
In the latter the reaction at first becomes less acid, then either much
more acid, or else it remains at a medium level.1

1 According, however, to the experimental work done by Dr. A. S. Griffith for
the Royal Commission on Tuberculosis (vol. iii. of Appendix to Second Interim
Report) these differences appear to be " differences in degree and not in kind,
and are attributable to variations in saprophytic power which have been shown
to exist on other media."

BOVINE AND HUMAN TUBERCULOSIS 125

4. Differences in Pathogenic Effect. — The bovine bacillus has
a much greater pathogenic power than the human bacillus
for all animals with which it has been inoculated ; except that
in the pig and guinea-pig the susceptibility to both types of
bacilli is so great that it is hard to distinguish between them
(Ravenel, 1902, p. 26). Koch and Schiitz in their experiments
found that in pigs also the bovine was much more active than
the human bacillus. Rabbits have been found to withstand
the injection of doses of human bacilli, when an equal dose of
bovine bacilli caused fatal tuberculosis. The difference in
pathogenic effect between the human and bovine type is even
more obvious in the case of cattle. Thus a subcutaneous
injection of 5 eg. of bacilli of the human type caused in cattle
only a local reaction at the seat of infection and in the neigh-
bouring glands, the local disease decreasing and not spreading
to internal organs, "even after protracted observation," whereas
the same dose of bovine bacilli caused disseminated tuberculosis
(Kossel, 1905). The difference in the two types is especially
marked when animals are fed with pure cultures of the bacilli.
When animals have been dosed for three months with cultures
of the human type, bacilli are found to have accumulated in
the mesenteric glands, without any change there other than
calcification, and always without that wider dissemination seen
in experiments with the bovine type.

Kossel draws attention to the necessity, in making compara-
tive tests, of taking certain precautions, the ignoring of which
may have caused some of the discrepant results published by
different experimenters — (i) Comparable material alone should
be used, — only young cultures, in which the same nutrient
material has been employed. (2) Fresh strains of bacilli must
be used, isolated recently from the animal body. (3) Faulty
results have ensued from inoculating with pieces of tuberculous
organs instead of with cultures. (4) Experiments should be
on as wide a basis as possible. Kossel inoculated 27 different
strains of bacilli of the bovine type and produced disseminated
tuberculosis in 32 out of 33 cattle ; while the inoculation of
38 different strains of bacilli of the human type into 44 cattle
produced local lesions only.

The above results can now be checked by the elaborate
and protracted experimental observations of the English Royal

126 THE PREVENTION OF TUBERCULOSIS

Commission given in their Second Interim Report (1907). The
experimental results of the work are summarised as follows by
Sidney Martin : —

GENERAL SUMMARY OF RESULTS OF THE ROYAL COMMISSION

i. Bovine Tuberculosis
(Thirty strains examined)

The bacillus of bovine tuberculosis has been shown by the experiments
to have certain characteristics as follows : —

a. It shows some variations in its growth on artificial media, and
according to these variations can be arranged into three groups or
grades (I., II., III.).

/3. When inoculated into bovines, rabbits, guinea-pigs, pigs, goats,
monkeys, and the chimpanzee in appropriate doses it produces death
by generalised tuberculosis.

y. It shows stability as regards its cultural characters, both when sub-
cultured and when passed through animals. Whether these characters
can be altered by prolonged passage in certain animals is still the
subject of experiment and cannot now be answered.

8. It shows great stability in virulence both after long subcultivation
and after passing through animals.

2. Human Tuberculosis
(Sixty cases examined)

The bacilli of human tuberculosis show a greater variety than those of
bovine tuberculosis.

Group I

(Fourteen cases examined)

a. The bacilli obtained from the virus of human beings in this group
have all the characters of the bacillus of bovine tuberculosis as regards
cultural characters, virulence for the animals previously mentioned, and
stability of cultural characters and of virulence.

The bacillus of this group is identical with the bacillus of bovine tuber-
culosis.

/3. The bacillus of these cases was a single bacillus — there was no
evidence of a " mixture " of different kinds of bacilli.

y. The bacillus was the cause of death of the individuals from which it
was obtained. This is more particularly shown by the study of Viruses
H. 32 " Y.W.," H. 59 " L.B.," and H. 64 " M.G.," in which general tubercu-
losis was the cause of death of the child. The disease started as abdominal
tuberculosis, but became generalised. Culture not only from the mesenteric
glands, but also from the bronchial glands and lungs and meninges, had
the characteristics of the bovine bacillus in cultivation and in virulence.
No mixture of bacilli was here present. The children died of an infection
by the bacillus of bovine tuberculosis.

This group includes three cases of cervical gland tuberculosis and eleven
cases of abdominal tuberculosis.

BOVINE AND HUMAN TUBERCULOSIS 127

Group II
(Forty cases examined)

The bacilli obtained from the virus of human tuberculosis in this group
differs from the bacillus of bovine tuberculosis in the following points : —

a. In culture they are more luxuriant and are distinguished as refer-
able to Groups IV. and V.

/3. When inoculated into calves and rabbits they do not produce the
generalised and fatal disease caused by the bovine bacillus.

The result of inoculation is not a negative one, but varies within
certain limits with different viruses, and in rabbits the viruses oc-
casionally kill the animal by producing a generalised disease.

They agree with the characteristics of the bovine bacillus in the follow-
ing points : —

a. They produce general tuberculosis in monkeys and the chimpanzee.

/3. The lesions produced in these animals are the same anatomically
as those produced by the bovine bacillus.

y. The lesions produced in calves and rabbits are histologically tuber-
culosis, although usually they show retrogression.

This group includes : —

Sputum Culture . . . . .2 cases

Pulmonary Tuberculosis . . . .10,,

General „ . . i case

Bronchial Gland „ . . .2 cases

Cervical Gland „ . . 6 „

Abdominal „ . . 8 „

Joint „ . . 9 „

Testicle ,, . . I case

Kidney „ . . i

The experiments show, however, that this division into two groups of
the bacilli found in human tuberculosis is not the whole question.

Group III
(Six cases examined)

The investigation of two viruses, H. 53 " D.H." and H. 49 " T.C."
shows that bacilli are obtainable from cases of human tuberculosis
which belong to neither group. The bacilli from the two viruses mentioned
showed an irregular virulence in calves and rabbits, and one of them,
H. 49 " T.C.," showed also (i) that the culture of the original material
lost its virulence after prolonged subcultivation, and (2) that the original
virus, although irregularly virulent for calves, became highly and uniformly
virulent after being passed through a calf. The culture of H. 49 " T.C."
obtained from the original material has in cultivation the characters of
the bacillus of bovine tuberculosis, belonging to Grade II. There was no
evidence of mixture in the case of either virus.

The results of the examination of the bacilli in the case of these two
viruses point to the conclusion that the bacilli were bovine in origin
and had been altered by residence in the human being.

128 THE PREVENTION OF TUBERCULOSIS

As bearing intimately on this matter, the question of the transforma-
tion of the human bacillus into the bovine as shown in the experiments
previously discussed must be mentioned.

When by passage through calves, the slightly virulent bacillus of
human tuberculosis becomes apparently modified into the bovine bacillus,
it was suggested that it was not a real modification, but that the original
virus was a mixture of bacilli, and that during the passage the bovine
bacillus alone survived. But in these passage experiments there is evi-
dence that at the time when the virus is becoming virulent, the bacilli
separated by culture are " unstable " in virulence for calves and rabbits ;
an instability similar to that of the original virus of H. 49 " T.C."

The consideration of these cases tends to bridge the gap between the
bacilli of Group I. (bovine bacilli) and those of Group II., which they
suggest may only be a form of bovine bacillus, degraded as regards viru-
lence for calves and rabbits, by long residence in the human body.

If bacilli of the bovine and human types have distinctive
characteristics, and differ greatly in their pathogenic effects on
cattle, the answer to the question, is tuberculosis in cattle pro-
duced by the bacillus of the human type, must with certain
limited exceptions be in the negative. It does not, of course,
follow from this that human tuberculosis may not be caused
by bacilli of the bovine as well as of the human type. The
results obtained by the Royal Commission as well as by German
and American observers indicate that bovine is at least an
occasional cause of human tuberculosis. There may be said
to be three schools of opinion on the subject : —

I. Human and bovine tuberculosis are totally distinct diseases,
and are not to any serious extent inter communicable. This
appears to be Koch's position, for in his Nobel Lecture (1906)
he says : —

We must attain to absolute clearness as to the manner in which in-
fection in tuberculosis takes place — i.e. as to how the tubercle bacilli
get into the human organism, for the sole purpose of all prophylactic
measures against a pestilence must be to prevent the entrance of the
germs of disease into man. Now, as regards infection with tuberculosis
only two possibilities have hitherto presented themselves — namely,
infection by tubercle bacilli emanating from tuberculous human beings
and infection by tubercle bacilli contained in the flesh and milk of tuber-
culous cattle. After the investigations which I have made hand-in-hand
with Schiitz as to the relation between human and bovine tuberculosis,
we may dismiss this second possibility, or at least regard it as so slight
that this source of infection as compared with the other falls quite into the
background. We arrived, namely, at the result that human tuberculosis
and bovine tuberculosis are different from one another, and that bovine

BOVINE AND HUMAN TUBERCULOSIS 129

tuberculosis is not transmissible to man. With reference to this latter
point, however, I wish, in order to prevent misunderstandings, to add
that in saying this I mean only those forms of tuberculosis that have
to be taken into account in connection with the combating of tuberculosis
as an epidemic disease — namely, generalised tuberculosis and above all
pulmonary phthisis. ... I wish only to add that the testing of our in-
vestigations which has been carried out with the utmost care and on a
broad basis in the Imperial Office of Health in Berlin has led to a confir-
mation of my opinion, and that, moreover, the harmlessness of the bacilli
of bovine tuberculosis to man has been directly proved by the repeated
inoculating of human beings with the material of bovine tuberculosis
by Spengler and Klemperer. In connection with the combating of tuber-
culosis, then, only the tubercle bacilli emanating from human beings
have to be taken into account.

2. The ingestion of bacilli of the bovine type is the essential
cause of tuberculosis in the human being. The chief exponent of
this view is von Behring, who, in his Cassel Lecture (1903), says :— -

Koch's assertion that there are essential differences between human
and bovine tubercle bacilli, and that these differences are not bridged over
by any connecting links . . . has since called forth observations from
all over the world which positively demonstrate the existence of inter-
mediary stages in the virulence of tubercle bacilli derived from mammals.
Generally, tubercle bacilli derived from cattle are more virulent for all
animal species thus far examined than are human tubercle bacilli. And
the opinion is constantly gaining ground that bovine tubercle bacilli are
also more virulent for man.

His own special views are embodied in the following extracts
from the same lecture : —

According to my ideas there has not yet been a single well-authenticated
case in which pulmonary consumption has originated in adults as the result
of a tuberculous infection developing epidemiologically, i.e. under con-
ditions essential for infection occurring in nature. „

His view is that in all cases in which phthisis is caused,
apparently by human infection during adult life, there has
been pre-existing tuberculosis of bovine origin, and he holds
that

considering the figures . . . showing the enormous diffusion of tuber-
culosis, the objection is surely justified that the persons thus dying of
consumption already had a tuberculous focus in the lungs, and that this
pulmonary disease, under a mode of life favourable to tuberculosis, was
converted into florid phthisis.
9

i3o THE PREVENTION OF TUBERCULOSIS

It is necessary to give further extracts from this lecture,
in order to make von Behring's position quite clear. He concedes

not only the possibility, but the actual occurrence of pulmonary tuber-
culosis going on to consumption, as a result of infection of an adult person
... in the sense that on the basis of an infantile infection a pulmonary
tuberculosis has developed, which becomes manifest only through the
agency of the additional infection.

His chief contention is contained in the following words :—

I believe I have discovered a new principle which may be expressed
thus :

The milk fed to infants is the chief cause of consumption.

3. Human tuberculosis may be and is caused by bacilli of
either the bovine or human type. This is the view most generally
and justifiably entertained, supported as it is by the balance
of all available evidence. The extracts from the Interim
Report of the Royal Commission given on p. 126 show that bacilli
of the human type are sometimes very virulent to cattle ; and
the practical conclusion given in an earlier report (p. 123) of
the same Commission as to the undesirability in the interest
of man of relaxing precautions against bovine tuberculosis,
must commend itself as reasonable. Thus Ravenel (1905, p. 147)
says : —

Theoretically, there is no reason why the bovine bacillus should not
be readily transmitted to man. It has for all other mammalia on which
it has been tried a virulence greatly exceeding that of the human tubercle
bacillus. It would certainly seem a remarkable anomaly for man,
who is one of the most susceptible of all animals to tuberculosis, to be
immune to the most powerful virus known. In the whole range of com-
municable diseases we have nothing comparable to this state of affairs,
should we admit it.

These three views will be next considered.

CHAPTER XVII

EVIDENCE OF THE OCCURRENCE OF BOVINE
TUBERCULOSIS IN MAN

THE occurrence of tuberculosis of bovine origin in man to
an extent of practical importance is, as we have seen,
denied by Koch and those who agree with him. What
evidence is there for and against this view ? Tuberculosis might
conceivably be produced in man by bacilli of the bovine type, (i)
if these bacilli were themselves able to cause active disease in
him ; or (2) if they were to survive in his tissues in a latent con-
dition for a period sufficient to enable them to become changed
into bacilli of the human type.

WHAT EVIDENCE is THERE THAT BACILLI OF THE BOVINE

TYPE CAN CAUSE ACTIVE TUBERCULOSIS IN MAN DIRECTLY,

WITHOUT CONVERSION INTO THE HUMAN TYPE ?— The only
satisfactory evidence available consists in finding, in the lesions of
human disease, bacilli which conform to all the known distinctive
tests of the bovine type, including those already given on p. 124.
This evidence has been supplied by various workers. Thus Theo-
bald Smith in 1898 made from the mesenteric glands of children
two cultures, of which one was of human while the other
was pronounced to be of bovine origin. At the same time he
supplemented the studies made by Ravenel " upon a presumably
bovine culture from a child, by applying a new reaction test "
(described on p. 124). ' This latter culture had also the char-
acteristics belonging to the bovine bacillus." Later (1904,
p. 9), he showed that the bacilli present in three cases of general
tuberculosis — a child aged eight months and two adults — did not
belong to the bovine type. In a paper published in 1905, Theobald
Smith, after giving further cases fully worked out, states that
Vagades (Zeitschr. fur Hygiene, 1898, xxviii. p. 276) found " one
culture among 28 isolated from man, which, it seems to me, was a
bovine bacillus." Tie also quotes Lartigau (Journal of Medical

132 THE PREVENTION OF TUBERCULOSIS

Research, 1901, vi. p. 156) as finding at least one bovine culture
of maximum virulence among nineteen cultures of human source ;
and he quotes Ravenel as having, like himself, isolated from
mesenteric glands two cultures, of which one was of the human
and the other of the bovine type. He emphasises (1905, p. 296)
the fact that

but few experimenters have taken the time necessary to isolate and
carefully compare cultures. The literature does not therefore offer
that precise basal information upon which far-reaching conclusions may
be built.

Since the above quotation was written, the Imperial Board of
Health in Berlin and the English Royal Commission have both
issued reports, the latter of which is quoted on p. 126. In the
former Kossel (1905, p. 1448) states :

The result of the far-reaching experiments conducted under my direc-
tion in the Gesundheitsamt at Berlin has been to show that in human
tuberculosis tubercle bacilli may exist that correspond in every respect
in their morphological, biological, and pathogenic qualities to bacilli of
cattle tuberculosis — that is, such as belong to the Typus bovinus.

Among 56 cases of human tuberculosis we found these germs 6 times —
that is, in 10 per cent, of the cases. It would, however, be erroneous
to conclude from these figures alone that 10 per cent, of all cases of human
tuberculosis in Berlin were caused by infection with tubercle bacilli of
the Typus bovinus, and that for the following reason : We included
in the number of our experiments chiefly cases in which we could assume
that the tuberculosis owed its origin to an intestinal infection, and possibly,
therefore, to food containing tubercle bacilli.

Tubercle bacilli of the Typus bovinus appear chiefly in tuberculous
lesions in children, and among our cases we found that, with one excep-
tion, it was the mesenteric glands or intestinal ulcers that contained
the bovine germs. When, on the other hand, the sputum of adults
suffering from pulmonary phthisis was examined, only bacilli of the
Typus humanus were found. That tubercle bacilli of the Typus bovinus
can, however, also enter the adult body was ascertained by our finding
them, together with those of the Typus humanus, in a case of extensive
tuberculous ulcers of the intestines in a woman.

RESULTS OF THE ENGLISH ROYAL COMMISSION.— As already
indicated, the bacilli obtained from sixty cases of human tuber-
culosis were exhaustively examined by every known method,
with the results as to type of bacillus set out on p. 126. These
results are so important from other points of view that I have
set them out in tabular form below, in a table modified from the
table on p. 72 of the report of the above Commission.

EVIDENCE OF BOVINE TUBERCULOSIS

133

TABLE XXIII

Summary of Results of Examination of Different Strains of Human
Tubercle Bacilli

8 .

C -i-j

a %

o -2

--H'ja

c 'S bfl t/j

***** '7^

3,a

H .^

»^ H G "^3

->-> •"§

.« d

;£ rt

2 w *£ c3

SaSa

>J 8 |

Nature of Case.

Part used for
Experiment.

<U QJ

•r-J <U

flfJ

J5 g

<S O

^ >

1/3 ^ !o 3

j> o

<u f cj .i;

I. Sputum (4 cases) .
2. Primary Pulmonary Tu-
berculosis ( 10 cases) .

Sputum .
Lung

...

Lung and cervical

gland .

...

3. General Tuberculosis

(i case)

Bronchial glands

...

4. Bronchial Gland Tuber-

culosis (4 cases) .

Bronchial glands

5. Cervical Gland Tuber-

culosis (9 cases) .

Cervical glands

6. Primary Abdominal Tu-

berculosis (19 cases) .

Mesenteric glands .

...

Mesenteric gland,

cervical gland,

meninges

...

Mesenteric gland,

bronchial gland .

...

Mesenteric gland,

lung, cervical

gland, meninges .

Mesenteric gland,

meninges

Mesenteric gland,

lung, meninges

7. Joint Tuberculosis (10

cases)

Scrapings from joints
Pus from lumbar ab-

...

scess .

...

8. Tuberculosis of Testis (i

case) ....

Testis

...

Tuberculosis of Kidney

(i case)

Kidney .

...

9. Lupus (i case)

Scrapings of the i

lesions .

Total

. 1

134 THE PREVENTION OF TUBERCULOSIS

It will be noted that 14 out of the total number of strains
obtained from human sources conformed to the bovine type.
Out of 19 cases of primary abdominal disease, in which infection
might be through ingestion, 10 were of bovine type ; out of
8 cases of tuberculous cervical glands, in which similar infec-
tion during swallowing might occur, 3 were of bovine type ;
whereas only i strain of the bovine type was obtained from
sputum out of 4 examined, and none from diseased lungs out
of 10 examined. Of 4 cases of bronchial gland disease 2 were
of the human type and 2 doubtful. Of 10 cases of joint
tuberculosis, all were of the human type.

It would be unjustifiable to infer from the above figures that
probably 14 out of 60, or about 23 per cent., of all cases of human
tuberculosis are derived from tuberculous cattle. If the single
sputum case be omitted, the parts affected in the above cases of
tuberculosis of bovine type are the mesenteric and cervical
glands. But primary tuberculosis of these parts causes less than
10 per cent, of the total mortality officially recorded as due to all
forms of tuberculosis in this country. If the 28 cases of cervical
and primary abdominal tuberculosis are assumed to be typical
of what similar examination on a larger scale would show, it is
noteworthy that 13 of these, i.e. about half, were of the bovine
type. This would reduce the 10 per cent, above mentioned to
5 per cent. ; and until further evidence accumulates it may be
convenient to assume that from 5 to 10 per cent, of the total
human mortality from tuberculosis is due to infection from
bovine sources.

This assumption will be subject to modification if future
investigations show that the bovine bacillus can be transformed
into the bacillus of the human type. We may next consider with
advantage the evidence at present available on this point.

WHAT EVIDENCE is THERE THAT TUBERCLE BACILLI OF ONE

TYPE CAN BE TRANSFORMED INTO TUBERCLE BACILLI OF ANOTHER

TYPE ? — Many investigators hold that the characteristics given
on p. 124 as distinguishing races of mammalian tubercle bacilli are
variable elements, and can be modified by growing the bacilli in
different culture media. Theobald Smith (1905, p. 297), however,
observes : —

This view, I think, would be rejected by all who have studied con-
tinuously bacilli from different species. Virulence necessarily declines

EVIDENCE OF BOVINE TUBERCULOSIS 135

with prolonged cultivation, and bacilli may assume slightly different
forms on different culture media. These do not overthrow, but simply
mask, racial characters.

On the other hand, Ravenel (1902, p. 45) says : —

With these facts before us I do not think we are forcing a point in
believing that it is at least possible for the bovine bacillus to become
rapidly so changed in the body of man that it will show the cultural and
pathogenic peculiarities which we find usually in cultures of human
origin.

In support of this view he quotes Nocard, who by introducing
bovine and human bacilli into the peritoneal cavity in collodion
sacs showed that in five to eight months both bovine and human
bacilli acquired the cultural characteristics of the avian tubercle
bacillus, and to a certain extent also its pathogenic action.
Holier thought that he had so changed the human tubercle
bacillus by passage through the blind worm for a year, that it
grew best at 20° C. like the bacillus of fish tuberculosis.

According to von Behring and De Jong passage through goats
is able to change the bacillus of the Typus humanus into the
Typus bovinus. Kossel is incredulous as to the transformations
enumerated above, believing the results obtained to be due to
inaccurate methods. He quotes as an analogous case the fact that
before Koch discovered a method of separating bacteria and
growing them in pure culture on solid media, examples of trans-
formation of one species of bacteria into another were described,
to be rejected on more accurate investigation. The experiments
of Weber and Taute indicate the need of great caution. They
have shown that the tubercle bacilli of fishes, mentioned above,
are really acid-fast saprophytes derived from mud. Similarly
in De Jong's experiment a goat was left for 3^ years after inocula-
tion with tubercle bacilli of the human type before cultures were
taken from it. The possibility under these circumstances of
extraneous infection by bovine tuberculosis is considerable.
Kossel, Weber, and Heuss passed bacilli of the human type
through goats and cattle, and found that after five passages
they still remained of the same type, although they had been
in the goats up to 202 days and in the cattle up to 381 days.

So far the evidence favours the view of a racial difference
between tubercle bacilli of the bovine and those of the human

136 THE PREVENTION OF TUBERCULOSIS

types. Against this view is urged the fact that it is possible — in
the words of Kossel (p. 1448) — to

immunise cattle with the aid of tubercle bacilli of the Typus humanus
against the bacilli of the Typus bovinus. This fact is adduced as proving
that one is dealing with one and the same germ, on the ground that im-
munity against a bacterium can only be produced by an identical micro-
organism. Especially von Behring and Lorenz have emphasised this
fact as conclusive.

But Kossel then remarks :—

On the other hand, it must be remembered that immunity in this
direction is not equally specific in all species of bacteria. The fact that
efficient tuberculin can be prepared from avian tubercle bacilli as well as
from mammalian by itself suggests caution in applying to tuberculosis
such experiences as have been gained with regard to immunity in other
groups of bacteria. Furthermore, Beck observed that animals injected
with acid-fast bacilli had become hypersensitive to tuberculin ; and Koch
stated that by injection of tubercle bacilli into animals a serum could be
produced which possessed agglutinating power, not only on tubercle bacilli,
but also on saprophytic acid-fast bacilli.

Finally, opinions are not wanting that the treatment of experimental
animals with saprophytic acid-fast bacilli — that is, by micro-organisms
in no way identical with tubercle bacilli — has a protective influence
against the infection by tubercle bacilli (Moeller, Friedmann). I do
not deny that the tubercle bacilli of the Typus humanus are nearly related
to those of the Typus bovinus, and that their origin may be traced to a
common stock ; but these are considerations for which sufficient founda-
tion is wanting. To-day we are dealing with two types that do not play
the same part in the distribution of tuberculosis in animals and man,
and therefore must not be confounded. It is not essential whether these
types are defined as different species, or races, or varieties, for in any case
they are not identical.

The position of the problem is well summarised in the above
extract from Kossel's report ; but on the same question the very
elaborate and important experiments made on behalf of the
English Royal Commission should also be consulted.

We are now in a position to consider

VON BEHRING'S VIEWS AS TO HUMAN TUBERCULOSIS. — These
are given in short in the extracts on p. 129. He holds that bovine
tubercle bacilli, after long residence in human tissues from infancy
onwards, become the source of adult phthisis, the chief cause of
mortality from tuberculosis. Dr. Romer showed that true
albumins penetrate unchanged the intestinal mucous membrane
of new-born foals, calves, and small laboratory animals, without

EVIDENCE OF BOVINE TUBERCULOSIS 137

being converted into peptones as in adult animals. Following
up this observation, von Behring found that similarly bacteria
passed much more easily through the alimentary mucous mem-
brane of new-born than of adult guinea-pigs. He concluded that
the penetrability of infantile mucous membrane in artificially
fed infants is the important cause of tuberculosis. He says :—

The tubercle bacilli which gain access to the system through the ali-
mentary tract in infancy constitute the important etiological factor in the
production of the tuberculous infection which leads to consumption. . . .

The virus of tuberculosis . . . creeps in most insidiously, all un-
noticed, being in this respect analogous only to the virus of leprosy, of
syphilis, or possibly of malaria in tropical countries. It may be months,
years, or decades before the infection leads to manifest disease. This
depends on the virulence of the virus . . . and on the number of bacilli
introduced.

Although von Behring states very lucidly the important fact
of prolonged latency, his view that nearly all or all tuberculosis
in man is due to a primary infection by the bovine bacillus cannot
be accepted, for the following reasons : —

(a) As inferred on p. 134, the results of the Royal Commission
can be regarded only as proving that a relatively small proportion
of human tuberculosis is of bovine type.

(b) The evidence given above points against the conclusion
that transformation of Typus bovinus into Typus humanus occurs ;
so that, until further investigations have been made, von Behring
cannot justifiably explain the relatively small proportion of the
bovine type as being due to transformation in the body into the
human type.

(c) Doubtless other observers have, like myself, collected a
number of cases of fatal infantile tuberculosis where human milk
alone had been given. The element of doubt attaching to
tubercle bacilli of uncertain or variable type (Group III. p. 127)
mentioned in the report of the Royal Commission (1907) is
summarised in the following extract (par. 63) from that report :—

Should it be proved that the cases in question were due to an ad-
mixture with the bacilli of human source of a few bacilli of bovine source,
the two kinds always remaining distinct the one from the other and
never becoming changed the one into the other, we should have no need
to enlarge appreciably our conception of the extent to which the human
body is subject to bovine tuberculosis. Such cases of admixture must
be few and their effect slight ; bovine tuberculosis in the human body

138 THE PREVENTION OF TUBERCULOSIS

would practically be limited to cases such as those which furnish Group
I. (p. 126).

Should, however, it be conclusively proved that a eugonic l bacillus
of low virulence may be modified under certain conditions into a dysgonic
bacillus of high virulence and vice versd, our views as to the relation of
human to bovine tuberculosis must be very different. Such a conclusion
would lead to the following view. Bacilli from a bovine source entering
a human body in scanty numbers may become lodged there without
immediately provoking a generalised progressive tuberculosis. During
their sojourn there they may become modified into eugonic bacilli of
low virulence ; and they may then give rise either to a limited tuberculosis
only or, under the influence of certain conditions, to a generalised pro-
gressive tuberculosis. For some time after the change they may remain
unstable and capable of reverting to their bovine character under changed
conditions, when subjected for instance to the influence of bovine tissues
as in the passage experiments. Or after a long stay in the human body
their character may become so fixed that they cannot be distinguished
from bacilli conveyed directly from man to man.

It is on account of the far-reaching bearings of the conclusion that we
are unwilling to make any statement at all premature.

We may take this opportunity of pointing out that time is an essential
factor in dealing with a disease of so chronic a nature as tuberculosis.
Some of its problems, such for instance as the possible change in virulence
and other characters of the virus obtained from one kind of animal by
repeated passage from animal to animal of another species, can only be
settled after constant observations extending over a long period of time.

From a survey of the evidence we must conclude that the
conversion of Typus bovinus into Typus humanus during the
lifetime of a single person and in his tissues is unproved. The
third of the three alternatives given in Chapter XVI., pp. 128-129
fits in best with all the facts at present known ; and we
are justified, in view of the balance of evidence, in concluding
that (i) both Typus bovinus and Typus humanus are competent
to produce tuberculosis in the human being ; (2) both forms of
the disease have been identified in man (p. 126) ; (3) the bovine
type is more common in children than in adults ; (4) the bovine
type retains its special characters even in the human subject ;
and (5) tuberculosis of bovine origin is much less frequent in the
human subject than tuberculosis of human origin.

Conclusions (i), (2), and (3) are established with certainty;
(4) and (5) are probable.

1 A eugonic bacillus is one which grows readily, a dysgonic bacillus one which
grows with difficulty on artificial media.

CHAPTER XVIII

TUBERCULOSIS FROM MEAT AND FROM MILK AND
OTHER DAIRY PRODUCTS

IN this chapter it is assumed that a certain — probably a
relatively small — proportion of human tuberculosis is
caused by tubercle bacilli of the bovine type ; and it is
proposed to consider the extent of the disease in cattle and the
frequency with which tubercle bacilli are found in milk and other
dairy products.

AMOUNT OF TUBERCULOSIS IN CATTLE.— According to the
evidence of Mr. (now Sir) T. H. Elliott, Secretary to the Board of
Agriculture, before the Royal Commission on Tuberculosis, 1898,
at least 20 per cent, of the cows in this country are tuberculous.
Delepine (1899) found that in farms which had careful sanitation
the proportion varied according to age from 20 to 31 per cent,
in milch cows, and that on some farms from three-fourths to all of
the cows were affected. MacFadyean (1901) states : " We know
that about 30 per cent, of all the cows giving milk in this country
are tuberculous in some degree." This undoubtedly implies a
most unsatisfactory state of things ; and if tuberculosis is easily
communicable to man from tuberculous cattle, the wonder is
not that the disease is common in man, but that it is not much
more common.

Tuberculous cattle might be a source of human tuberculosis
(i) by dust or spray infection from cattle suffering from lung
disease ; (2) by the eating or handling of the flesh of tuberculous
cattle ; or (3) by consuming milk or some milk-product derived
from tuberculous cows.

There is no evidence on the first point, and it may be
ignored, as an unlikely or at least an uncommon source of
infection.

TUBERCULOUS FLESH. — Butchers and others dressing tuber-
culous animals may receive accidental inoculations through

139

140 THE PREVENTION OF TUBERCULOSIS

wounds ; but the development of fatal tuberculosis after such
accidents is excessively rare.

The flesh from tuberculous cattle is undoubtedly sometimes
infective. Much evidence on this point was collected by the
English Royal Commission of 1895. It was shown that uncooked
tuberculous material given as food to guinea-pigs, calves, pigs, and
cats produced tuberculosis. In " joints " of meat it is excep-
tional to find tuberculous nodules or other evidence of disease,
though to a practised eye the " stripping " of the pleura lining
the ribs gives rise to suspicion of tuberculous " grapes" removed
in the dressing of the animal. S. Martin in his experiments
for the above Commission frequently produced tuberculosis by
inoculating or feeding animals with flesh from tuberculous cattle,
" in which no tubercle could be detected by his ocular tests."
This led him to consider the " real and considerable danger " of
the meat becoming contaminated by the butcher's hands, knives,
and cloths, which had been previously in contact with tuberculous
lesions in the animal. " The greater the amount of tubercle
there is in the cow " the more likely " is the sticky caseous
matter to get smeared over the carcass." Thus he failed to pro-
duce tuberculous disease by feeding animals on meat from cows
with mild or moderate tuberculosis, though inoculation of test
animals might be successful ; while feeding with meat from
cows with advanced or generalised tuberculosis succeeded in
producing tuberculosis.

The main tuberculous lesions in cattle are found in the organs,
membranes, and glands ; but seldom in the flesh or meat sub-
stance. Naked-eye evidence of disease has therefore usually
been removed from the dressed carcass, with the possible excep-
tion of a few pea-like tubercles internal to the ribs or about the
diaphragm, or a few small glands in certain " joints." As will be
seen subsequently, cooking processes may with certain exceptions
protect the adult (p. 404) ; but meat juice made from tuberculous
flesh is distinctly dangerous.

The fact that during a period in which the consumption of
meat has greatly increased, human tuberculosis has greatly
declined does not favour the view that tuberculous meat has
played a large part in its causation. I am not aware of any
evidence that the proportion of tuberculous cattle is markedly
less than formerly.

TUBERCULOSIS FROM DAIRY PRODUCTS 141

TUBERCULOUS MILK. — The evidence of the pathogenicity of
cows' milk to a dangerous extent is much clearer than that of
cows' flesh. Thus S. Martin reporting to the same Commission
(p. 16) found that out of 15 tuberculous cows 8 had healthy
udders ; 2 had udder disease, which was proved after slaughter
not to be tuberculous ; and the remaining 5 had tuberculous
udder disease. With the milk from these cows, tests were made
with the following results (Report, p. 16) :—

(a) The 8 tuberculous cows which had healthy udders showed him no
tubercle bacilli whatever in the milk of any one of them ; 41 test animals
fed with their milk remained perfectly free from tuberculous disease ; 28
test animals inoculated with their milk also remained quite free from
tuberculous disease.

(6) The 2 tuberculous cows which had udder disease, found post-
mortem not to be tuberculous in nature, showed him no tubercle bacilli
in their milk. Three test animals, fed with their milk and 14 other test
animals inoculated with their milk, remained, all of them, perfectly
free from tuberculous disease.

(c) Of the 5 tuberculous cows which had udder disease, found post-
mortem to be of tuberculous nature, 3 showed him tubercle bacilli
in their milk. He could not find tubercle bacilli in the milk of the
other 2. With milk from the 3 cows, 15 test animals were fed, with the
result of producing tuberculosis in every one of them. With milk from
one or other of the same 3 cows, 13 test animals were inoculated, with
the result of all 1 3 acquiring tuberculous disease. The milk of the fourth
cow (one of those which had not shown tubercle bacilli) was used to feed
10 test animals, and produced tuberculosis in 4 of them. Inoculated
into 6 test animals, all of them became tuberculous. The milk of the
fifth cow (in which also no tubercle bacilli had been seen) was used to
feed 2 animals, but without result. Yet when it was used to inoculate
2 other animals, both of them acquired tuberculous disease.

(d} It remains to note these tests as applied to the milk of the two
cows found after slaughter to be suffering under another disease, but not
tubercle. The results were : no tubercle bacilli found in the milk of
these cows ; inoculated into 17 test animals, it did not produce tuber-
culosis in any one of them ; milk from one of the cows, however, in some
test animals gave rise to various abscesses.

The Report of this Commission goes on to say (p. 17) that

according to our experience, then, the condition required for ensuring
to the milk of tuberculous cows the ability to produce tuberculosis in
the consumers of their milk, is tuberculous disease of the cow affecting
the udder. It should be noted that this affection of the udder is not
peculiar to tuberculosis in an advanced stage, but may be found also
in mild cases.

142 THE PREVENTION OF TUBERCULOSIS

All are agreed that when there is tuberculosis of the udder
the milk is found to be dangerously infectious, and so likewise
are all products of such milk, as butter, skimmed milk, butter-
milk, cheese. Thus the report of the same Royal Commission
states : " The milk of cows with tuberculosis of the udder
possesses a virulence which can only be described as extra-
ordinary " (par. 61). It is also ominous that " the spread of
tubercle in the udder goes on with most alarming rapidity."
Sims Woodhead remarks (par. 62), "I have noticed on several
occasions, during the interval between fortnightly inspections
carried on along with a veterinary surgeon, that the disease had
become distinctly developed. It may be, of course, that the
early evidence has been overlooked at the previous inspection,
but whether this is the case or not, the spread of the disease was
so rapid as to afford very good ground for alarm. The very
absence of any definite sign in the earlier stage is one of the greatest
dangers of this condition."

Professor (now Sir) J. MacFadyean (1901, p. 84) points out in
the following remarks,

not every cow that is tuberculous gives milk containing tubercle bacilli.
It is true that opinions with regard to this point are not absolutely unani-
mous, but there is ample evidence to justify the assertion that, as a rule,
the milk is not dangerous until the udder itself becomes diseased. The
experiments pointing to an opposite conclusion form only a small minority,
and the results obtained in most of them were probably due to careless-
ness on the part of the experimenter. In a few of the cases in which
the milk of an apparently healthy udder was found to be infective it is
probable that the gland tissue was in reality diseased, though not to
an extent discoverable without microscopic examination. The important
question, therefore, is not what proportion of milch cows are tuberculous,
but what proportion of them have tuberculous udders. Some authorities
have estimated this to be as high as 10 per cent., but the proportion is
certainly much less than that in Great Britain. My own experience
leads me to think that about 2 per cent, of the cows in the milking herds
in this country are thus affected. Now, the milk secreted by a tuber-
culous udder always contains tubercle bacilli, and it sometimes contains
enormous numbers of them, and when these facts are apprehended one
begins to realise the seriousness of the danger to which, in the present
state of affairs, those who drink uncooked milk are exposed. But tl
are one or two considerations that make the danger greater than tl
mere statement of the number of cows affected would at first sight in-
dicate. In the first place, the udder disease is not attended by any paii
or tenderness in milking, and the milk for a considerable time after th<
udder has become manifestly diseased may appear quite wholesome,

TUBERCULOSIS FROM DAIRY PRODUCTS 143

though in reality it is charged with the germs of tuberculosis. It there-
fore often happens that the gravity of the condition is not realised by the
milker or the owner of the cow, and the milk continues to be sold for
human consumption. There is scarcely any room for doubt that if it
were sold and consumed unmixed with other milk some of the persons
partaking of it would become infected. In practice it is usually mixed
with the milk from other cows that have healthy udders, and thus the
germs are distributed among a large number of persons. Even tuber-
culous milk that has been thus much diluted may prove infective, but the
danger to the individual consumer is in inverse proportion to the degree
of dilution. Since about one cow in 50 is the subject of tuberculosis
of the udder, and the average number of cows in the milking herds of
this country is less than 50, it follows that the majority of dairies and
farms supply milk that is free from tubercle bacilli, or at least does not
contain any derived from this source. On the other hand, when the
infected material is present, it operates with the greatest intensity in
the milk of single cows and in the mixed milk from small herds.

By other observers the percentage of milch cows with tuber-
culous udders is put somewhat higher. Thus Professor Delepine
puts it at 3*7 per cent. (1899, p. 19).

Miiller (1905) found that the udder was tuberculous in from
1*1 to 3'7 per cent, of the tuberculous cows slaughtered in Saxony
during the years 1888-97, and in r6 per cent, of the tuberculous
cows in the whole of Germany. In Denmark in 1901-02 the
number of cases in which tuberculosis of the udder was detected
and the cows subsequently slaughtered was 584, or 0*55 per 1000
of the total stock. In the experience of the East Prussian Herd-
book Society, a half-yearly examination of the herds and a
quarterly examination of the milk, implying a very thorough
control, only showed 62 cases of tuberculous udder in 15,000
cattle, or 0*4 per cent.

ARE TUBERCLE BACILLI FOUND IN Cows' MILK IN THE
ABSENCE OF TUBERCULOUS UDDER DISEASE ? — From the experi-
ments of the English Royal Commission already quoted, it
would be inferred that this question must be answered in the
negative, or that tubercle bacilli when found are too few to be
dangerous. Other experimenters have published results con-
tradictory to these, tubercle bacilli being found in milk when
udder disease was absent in cows suffering from clinical tuber-
culosis, or even when cows had no obvious evidence of tuber-
culosis but reacted to the tuberculin test. These results have
failed to be substantiated. Thus Ostertag examined 77 such

144 THE PREVENTION OF TUBERCULOSIS

cows without finding tubercle bacilli in the milk after testing
it microscopically, by inoculation, and by prolonged feeding
experiments. Ascher, M'Weeney, and Strenstrom obtained like
results. The latter concluded that tubercle bacilli found in the
milk by observers obtaining different results must have gained
access to it during milking. This may have been derived from
tuberculous milkers ; but Ebers regards the very common
fouling of the milk with particles of cow-dung as the source of
tubercle bacilli. Tuberculous cows after coughing commonly
swallow their expectoration, which would subsequently appear
in the faeces. This evidence has been more recently confirmed.

The detailed results of the East Prussian Herdbook Society
are interesting in this connection. Samples of milk were taken
from the total milk of 1596 herds, and tubercle bacilli were found
in 97 samples. In 59 of these tuberculous udders were dis-
covered, and in the other instances there was reason to believe
that contamination of the milk after leaving the animal had
occurred. The above experiment represented the milk of about
20,000 cows, and it may be assumed in accordance with average
experience that 6000 to 7000 of these were tuberculous ; and
yet in 1499 out of 1596 herds no tubercle bacilli were found in
the milk. The evidence that contamination of the milk is most
often due to udder disease is very strong, though contamination
by cows' dung or from milkers also occurs, and cannot be left out
of count.

PROPORTION OF INFECTIVE MILK IN MIXED SUPPLIES TO THE
PUBLIC. — The preceding experience of the Herdbook Society
must be regarded as exceptional, in the fewness of the herds
containing infective milk. English experience shows that a very
large percentage of ordinary mixed milk contains tubercle bacilli.
Thus Delepine (1898) found tubercle bacilli in 22 out of 125, or
17-6 per cent., of samples of milk from country dairy-farms
collected at railway stations in Liverpool and Manchester.
Kanthack and Sladen found that specimens of 9 dairies were
infected out of 16 examined. Woodhead and Wood found
virulent tubercle bacilli in 5 out of 50 specimens, and Rabino-
witsch and Kempner in 7 out of 25 samples in Berlin. Taking
these as fair samples of a much larger number of examinations,
it would appear that about 20 per cent, of the mixed milk supplied
to towns contains living tubercle bacilli.

TUBERCULOSIS FROM DAIRY PRODUCTS 145

TUBERCLE BACILLI IN OTHER DAIRY PRODUCTS. — Many
observations have been made and tubercle bacilli have been
found. It is not unlikely that the earlier observations over-
stated the facts, acid-fast bacilli simulating the tubercle bacillus
having been confused with it. There can be no doubt, however,
that when milk contains tubercle bacilli, cream, butter, cheese,
skimmed milk, and buttermilk are likewise infective. Margarine
may also contain tubercle bacilli, introduced with the milk
which is blended with it. Cream is likely to be particularly
dangerous, as the cream in rising is found to carry an excessive
proportion of the bacilli with it. The feeding of calves and
pigs on skimmed milk, buttermilk, whey, and the refuse collected
on centrifuges is a common source of tuberculosis in them. The
horse has also been shown, especially in Denmark, to be very
liable to tuberculosis when fed on milk or its products. Pigs
are rarely infected from one another, but mainly by their food.
Tuberculosis is very prevalent in pigs only when a large dairy
industry is carried on. The slaughter-house reports of Copen-
hagen for 1897 show that the proportion of tuberculous pigs
varied from 3 to 14 per cent. ; while in Bavaria, in which there
is only a small dairy industry, only 0*2 to 0*4 per cent, of the
pigs slaughtered in 1896-1900 were tuberculous. In Denmark
pig tuberculosis has become much less frequent since it has been
made compulsory to heat separated milk before it is returned
from the creameries.

CHAPTER XIX
DOMESTIC INFECTION

TUBERCULOSIS is undoubtedly caused most often by
domestic infection. Koch (1906, p. 1449) savs that
tuberculosis " has been frankly and justly called a
dwelling disease " ; while Biermer goes further and describes it
as essentially a bedroom disease. There is little doubt that its
infection is chiefly acquired in bedrooms. Industrial conditions,
although an important source of infection, probably act to an
even greater extent by removing or paralysing influences inhibi-
tory to infection, thus opening the door to infection or stirring
into activity infective material latent in the tissues.

In treating of domestic infection it is necessary to distinguish
between indirect or mediate and direct or immediate infection.
The influence of overcrowding is complex, and is concerned
partly with* infection and partly with the conditions of imperfect
sanitation usually associated with overcrowding.

INFECTION DUE TO THE DWELLING PROPER. — The experi-
mental Results of Cornet and others (p. 98) show that tubercle
bacilli are present, but only in the immediate environment of
consumptives. Given that a house has become infected through
the uncleanly habits of a consumptive who has recently lived
and possibly died in it, there are the great limitations to infection
already enumerated on pp. 101-105. Although it is in the highest
degree desirable that such a house should be efficiently cleansed
and disinfected, it is unlikely to form a large element in the pro-
duction of phthisis by domestic infection. It may be, however,
that apart from this additional source of infection, evil conditions
of housing lower vitality, diminish the resistance to infection,
and thus increase the amount of tuberculosis among the poor.
This point is further discussed on p. 192. Such influences
undoubtedly favour tuberculosis by hastening the occurrence
of infection, and no preventive measures can be regarded as

DOMESTIC INFECTION

147

efficient and complete which do not vigorously attack and re-
move housing defects. It is possible, however, to obtain some
indications of the chief agency which causes the dissemination
of tuberculosis in overcrowded quarters.

OVERCROWDING. — There is abundant statistical evidence of
the close association between overcrowding and excessive
mortality from phthisis. Thus Sir Shirley Murphy has shown
that in London the death-rate from phthisis steadily increases
with the proportion of the total population living more than two
in a room, in tenements comprising less than five rooms. This
experience is summarised in the following table : —

TABLE XXIV

Condon. — Proportion of Population living

more than Two in a Room
(in Tenements of less than Five Rooms).

London. — Average Annual Death-rate

from Phthisis per 100,000 of Population,

1894-98.

Districts with under 10 per cent.
10-15
15-20
20-25
25-30

30-35
over 35

144

161;

177

209

231

259

When the same facts are subdivided according to ages of
the patients dying from phthisis, it is found that the excess of
the death-rate from this disease in the most overcrowded
districts is greatest at the ages at which the mortality from it is
heaviest. Sirj Shirley Murphy in commenting on the table
summarised above says (Ann. Rep. 1898, p. 46) : —

There is obviously a relation between the amount of overcrowding
;and the phthisis death-rate. The figures do not, however, suffice to
show whether the overcrowding caused phthisis, or whether the disease,
by adding to family expenditure or by diminishing the wage-earning
| power, left less money available for rent and thus brought about the over-
I crowding, or whether again overcrowding is associated with some other
j condition or conditions which are favourable to disease. In all- prob-
i ability all these circumstances have tended to produce the results shown
| in the table.

There is a further difficulty in accepting the above figures
'as completely satisfactory evidence that crowding is a main
'influence in causing tuberculosis. The house where a person

14*

THE PREVENTION OF TUBERCULOSIS

dies of this disease is not necessarily the house in which he ac-
quired it. In view of the frequent changes of house among the
poor, and of the protracted duration of phthisis, the coincidence
between the two is probably exceptional. The usual course of
events is for a person who becomes consumptive to drift, owing
to his impaired working powers, from the class of skilled to that
of unskilled and casual labour ; and with each step downwards
his housing conditions deteriorate to a corresponding degree.

In Part II. pp. 220 to 229 a comparison of different coun-
tries shows that the death-rate from phthisis does not vary in
accordance with their relative position as to sanitation and
housing, whether the different countries are compared with each
other, or whether the death-rate and housing conditions of
the same country are compared at different times.

The following additional evidence, quoted from a recent
address by the writer (1907), bears on the same point. The
figures as to housing are taken from a paper by Sir W. Matheson,
Registrar-General of Ireland: —

TABLE XXIVA

T3<~

-0 0 e

w 0 c

lit

jy o "" j2

|fsS|

c S o ^.^5

% &| g

S-jf c I 1

aj i> s-i ^ o

PH f"1 r- PH

H*i

O PH ^ g

O rt £^H ^

«*H rQ 4) .5 ^_

$ S."*

*o *jQ a

"o w g 8s

° g s +3 §

Q " «3

>rj 0) "-J r_j

(D ^ o M <^f

^ O (-1 OS f-H

<u "M tJ rf

^O S "tpJ

3 ^-* ^^ ^*» o

•~ O tS O-i

S^'-S ^^

| gH

S 1 2i §
£ c g «

I i'^ "^

|gS

fcH

*^H

O Qlo

Dublin

36-70

8-69

I0'6i

329

Belfast

I -00

0-09

O'lO

313

London

14-66

0-57

0*70

171

Liverpool

6-14

0-22

0*24

190

Manchester .

1-90

0*04

0-05

208

Edinburgh .
Glasgow

16-98
26-11

I -80
4-28

2-33
5-24

I64
177

Thus in Glasgow, which has 26 times as large a proportion
of one-roomed tenement dwellings as Belfast, and 52 times as
many persons in its one-roomed tenements with 5 or more oc-
cupants, the death-rate from phthisis instead of being higher
is 43 per cent, lower than that of Belfast. This does not imply
that in a given town the death-rate from phthisis is not higher

DOMESTIC INFECTION 149

I in the smaller and more overcrowded tenements. Abundant
statistics show this to be the case. But it is clear from the
I above table that size of dwelling or even degree of overcrowding
may be overshadowed by the effect of other influences.

It may be taken as an axiom that overcrowding favours

J tuberculosis. Doubtless there is more than one modus operandi in

I! bringing about this result. Two things, however, are certain : —

(a) Tuberculosis cannot be produced, however strong may

be the favouring circumstances, unless its infection is received ;

!| and (b) although, as seen above, the death-rate from phthisis in

a given community is always greater in proportion to the

amount of overcrowding, there is, when different countries or

different cities are compared with each other, no direct relation

between the amount of overcrowding and the amount of

phthisis.

It will be subsequently seen that a given amount of over-
crowding with a large amount of institutional segregation of
consumptives is associated with less phthisis than when over-
crowding is less but accompanied by only a small amount of
institutional segregation of consumptives (pp. 224 to 295). We are
justified in concluding therefore, that the quickest way to diminish
the risks of overcrowding is to favour by every means of persuasion
\the removal of the sick from among the healthy. This should, of
course, be accompanied by strenuous endeavour to diminish over-
crowding, apart from the question of such removal.

FAMILY INFECTION. — The facts already given indicate almost
sufficiently the risks of family life when one member is a con-
sumptive, though they also happily indicate with what ease and
how simply these dangers may be avoided. The histories of
family infection given on pp. 64-68 are examples of the
conditions under which tuberculosis spreads.

It is sufficiently clear that young children are particularly

prone to be infected, partly because they are more caressed, and

i possibly also because they are more susceptible than their elders.

Girls are more exposed to infection than boys (see p. 171).

The most intimate relationship in family life is that of husband

land wife, and the evidence as to infection between these may

I therefore be examined.

INFECTION IN MARRIED LIFE. — When a married man or
•woman is consumptive, is the proportion of instances in which

THE PREVENTION OF TUBERCULOSIS

the partner is also consumptive greater than the average for
persons of the same age and sex apart from married life ? There
cannot be said to be sufficiently full evidence to settle this point.
The following table is given to show the varying percentages
stated by different collectors of statistics :—

TABLE XXV
Number of Married Couples with One or Both Consumptive

jy a;

a!)

Authority.

ITw'S

O Q CU

I||

fit

Quoted from —

III

Brehmer

159

II'9

Cornet On Tuberculosis

(Nothnagel), p. 265.

Haupt .

260

"'5

Cornet .

594

135

22 7

})

Schuyder

844

3-8

Lancet, Sept. 19, 1891.

Rivers .

7-1

K. Pearson, 1907.

Weber .

237

Weber, 1874.

Clearly figures giving such discrepant percentages cannot
be comparable. Observations of supposed infection between
married couples or its absence are trustworthy only if they
accurately state the length of the married life of the couples
under observation, and the subsequent history through life of
the surviving partner. In other words, to arrive at the truth
one must have the complete life-experience of the married
couples, and a sufficient number of these to avoid accidental
errors. I do not think that most of the observations tabulated
above will bear this test. Even when these tests are satisfied,
it has to be remembered that frequently patients having had
phthisis die as the result of other diseases. The long latency of
phthisis in a considerable proportion of the total cases is one of
the most serious difficulties in the more detailed and elaborate
investigation on this point that is needed.

Even when allowance is made for coincidence, the following
instance of apparent communication of pulmonary tuberculosis
by a husband to successive wives, given by Sir Hermann Weber
(1874, p. 144), is sufficiently striking to deserve reproduction : —

DOMESTIC INFECTION 151

A. B. lost his mother, two brothers, and a sister from pulmonary tuber-
culosis. He had haemoptysis at the age of 20. He then became a sailor.
He married when 27 years old, and was then quite well.

His first wife came of a healthy family, and had good health till to-
wards the end of her third pregnancy, and she died after her confinement.

After a year he married again, his wife being apparently healthy.
She developed a cough after a year of married life, and died of pulmonary
tuberculosis.

His third wife was 25 years old when he married her. She came of
an exceptionally healthy family. In her second pregnancy she began
to cough, and died after the second confinement.

His fourth wife, who was 23 years old when he married her, and who
had come of a healthy family, began 1 3 months later, i.e. 3 months after
her first confinement, with a cough, and died later of phthisis.

A. B. did not marry again, When examined in 1854 after the death
of his third wife he showed evidence of old pulmonary tuberculosis. He
died in 1871 of this disease, and an autopsy showed old cicatrised disease,
and recent tuberculosis.

Dr. Weber states that in 29 marriages between consumptive
wives and healthy husbands only one husband became con-
sumptive ; while in 51 marriages between consumptive husbands
and healthy wives 18 wives became consumptive.

There is, I think, in view of our general knowledge of tuber-
culosis, no reasonable doubt that the close intimacy of married
life has, in the absence of intelligent precautions, been a not
infrequent cause of phthisis when one partner is already affected.
The wife is more likely to suffer from her diseased husband, than
the husband from his wife ; as the wife has more protracted
opportunities of receiving infection, especially in the later stages
of the disease.

CHAPTER XX
INFECTION IN ATTENDANCE ON THE SICK

THE majority of consumptives, when ill enough to require
nursing, are nursed by their own relatives. The degree to
which infection occurs among them has already been
discussed (p. 149). In view of the evidence already given, and
that cited in Part II., there can, I think, be little difficulty in
agreeing that the home-treatment of advanced consumptives
in crowded dwellings, in which the necessary precautions cannot
be taken, is a predominant cause of the continued spread of
tuberculosis. It still remains to discuss the possibilities of
infection of nurses and other attendants in the institutional
treatment of phthisis, and the possibilities of infection of doctors
who attend consumptive patients at their homes or in institutions.
The most carefully investigated experiences are those of the
Brompton Hospital and of the Victoria Park Hospital for Dis-
eases of the Chest, the former investigated by Drs. Cotton and
Theodore Williams, the latter by Dr. Andrew. Wilson Fox (1891,
p. 563) summarises these experiences in the table on the follow-
ing page.

In the Brompton returns the number of nurses and servants
is given only for 20 years, the deaths for 36 years. It appears
that, so far as could be ascertained, during 36 years only one
death from phthisis occurred among the physicians, and only
five cases among the nurses during or subsequent to their work
in the hospital. The results for the Victoria Park Hospital
are somewhat similar. It is very difficult to analyse this evidence.
It is very scanty. It is not certain how thoroughly the subse-
quent history of workers in these hospitals was traced. It is
likely that such workers as had died were less completely traced
than those still alive. Again, we do not know the total dura-
tion of hospital work of the above persons. If we assume that,

including servants, it averaged two years, then among the 377

152

INFECTION IN ATTENDANCE ON THE SICK 153

workers in the Brompton Hospital the annual number of cases
of phthisis among the staff while still at the hospital (exclud-
ing deaths) was about i in 94, or including cases developing
later was i in 37, which is much higher than the estimated
number in the general population (p. 63). I do not think,
however, that the evidence as collected is sufficiently accurate
to bear such a comparison as this, and it is made only to in-

TABLE XXVI

Brompton.

Victoria Park.

Number of Cases

of Phthisis.

<-M

S
o

C/2

I-S

cJ *a

<u.2

,0 |

M
OJ

S "o

Resident Medical Officer .

Clinical Assistants

Matron .

Nurses

IOl(?)

!(?)

5(?)

Servants .

32(?)

...

j255

I(-)

I(»)

Porters

Secretary and Clerks

3(?)

3 ! i

Dispensers

Chaplain .

.. .

...

Physicians and Assistant

Physicians

...

i(?)

Total

377

402

8(7?)

dicate that the data, if completely accurate, do not centra-
indicate a considerable possibility of infection among the staff
of these hospitals, and do not, as commonly supposed, offer
any presumption of freedom from infection.

A similar remark applies to Dr. Robertson's figures for the
Ventnor Hospital for Consumption (Bulstrode, 1903, p. 76).
During the 22 years 1881-1902, 15,500 phthisical patients were
treated in this hospital, and during the same period 62 officers,
208 nurses, 407 housemaids, and i charwoman — total, 678
—were engaged in the institution. None of the officers have
contracted tuberculosis. Six nurses, of whom two died, have

154 THE PREVENTION OF TUBERCULOSIS

had phthisis, but apparently three had the disease on admission.
The records for housemaids are not very definite. Here, again,
one would wish for exact information as to the length of service
and of the subsequent period over which each member of the
staff could be traced. In view of what has been said about
prolonged latency of tuberculosis (p. 73), this is an essential
condition of an accurate investigation.

The above experiences are usually quoted as instances of
non-infection in hospitals. They should rather be described
as examples of investigations, in which the data are, possibly
owing to insuperable difficulties, incomplete and insufficient
to justify any dogmatic statement.

In attempting to ascertain the true inwardness of the statistics
of hospital staffs relating to phthisis, generally quoted, it is not
suggested that the nursing of consumptives under the hospital
conditions of to-day, including the adoption of the best pre-
cautionary measures, involves considerable risk.

All that is suggested is that the danger is to a definite extent
greater than that for the general population, though much less
so than formerly. In all well-regulated workhouse infirmaries,
hospitals, and sanatoria, absolute cleanliness is maintained ; and
soiled handkerchiefs and the contents of spittoons are pre-
vented from becoming sources of infection. The chief remain-
ing source of danger is direct infection, which the careful nurse
avoids. The conditions are altogether different from those of
the wife who attends on the consumptive breadwinner. She
is in intimate personal contact with the patient day and
night ; may have insufficient rest ; is overf atigued, and often
underfed. Mental anxiety still further lowers her powers of
resistance to infection. It is not strange, therefore, if she falls
a victim, while the hospital nurse escapes. There is little diffi-
culty in agreeing with Koch's summing up of this subject (1906,
p. 1449) :-

In hospitals for. pulmonary phthisis it is in certain circumstances
possible that no cases of infection occur among the attendants, or at
any rate so few that in former times it was thought necessary to regard
this as a proof of the non-contagiousness of tuberculosis. But if one
examines such cases more carefully there are good reasons for the apparent
non-contagiousness. It then appears that the patients in question are
people who are very cautious about their sputum, see to the cleanliness
of their dwellings and clothing, and live in copiously aired and lighted

INFECTION IN ATTENDANCE ON THE SICK 155

rooms, so that the germs that get into the air can be swiftly swept away
by the current or killed by the light. If these conditions are not fulfilled,
there is no lack of infection even in hospitals and the dwellings of the
well-to-do, as experience teaches daily. And it becomes the more frequent
the more uncleanly the patients are as regards their sputum, the more
lack there is of light and: air, and the more closely crowded together the
sick live with the hale. The danger of infection becomes especially great
when healthy people have to sleep in the same rooms with sick people,
and even, as unfortunately still frequently happens among the poor, in
the same bed. This kind of infection has struck attentive observers
as so important that tuberculosis has been frankly and justly called a
dwelling disease.

Doctors are not exposed to infection so often, or for such
long periods, as nurses. They have no difficulty in their work
in escaping direct infection from coughing, and one would not
expect to have among them any definite evidence of risks
markedly greater than those of the general community, of acquir-
ing tuberculosis. The data in Table XXVI. are too scanty
to form the basis of a sound conclusion. The official occupa-
tional figures given by Dr. Tatham in the Decennial Supplement
of the Registrar-General's Report (1881-90) offer a much wider
basis of induction. In these figures the death-rate from all
causes and from certain specified causes among males, aged
25-65, are compared in groups, whose composition as to age
is identical. In these groups the number of the general popu-
lation that would furnish 1000 total deaths from all causes
(comparative mortality figure) is found to furnish 966 deaths
among doctors, 821 among lawyers, and 533 among the clergy.
Ogle in 1871-80 found that the death-rate from phthisis and
from respiratory diseases was lower among doctors than among
the general male population. The figures for 1881-90 confirm
this result, as shown in the following table : —

TABLE XXVII

Comparative Mortality Figures of Males aged 25-65, during
in Different Occupations, from

All Causes.

Phthisis.

Bronchitis.

Pneumonia.

Influenza.

All Males

IOOO

192

107

All occupied
Males

953

185

105

• 34

Clergy .

533

Doctors .

966

IOS

1 i

156

THE PREVENTION OF TUBERCULOSIS

Doctors have a much lower death-rate from phthisis than
the average male population. It will be observed that their
death-rate from influenza is excessive, and the comparison
is interesting, illustrating as it does the much more rapid and
more intense infectivity of the latter disease.

CHAPTER XXI

INDUSTRIAL INFECTION

IN considering the possibilities of infection in various
industries, the general considerations already emphasised
must be borne in mind. (i) Prolonged exposure to
infective material is more likely to be successful than
intermittent and occasional exposure. (2) Intimate con-
tact, as between husband and wife, and still more — because
of the possibilities associated with long latency — between
parent and child, is more likely to cause infection than the
less intimate contact which characterises the usual conditions
of work.

It has to be remembered, however, that the dust inhaledln
many occupations may not only serve as a vehicle for the
tubercle bacillus; but if, as frequently happens, it is angular
or rough, may serve as an inoculating needle for the bacillus ;
and by this means it is conceivable and in fact likely that
smaller doses of infective material than in domestic life may be
made almost equally efficient.

TABLE XXVIII.— PHTHISIS

Comparative Mortality Figures of Males aged 25-65, the total Deaths
of all Males at these Ages being taken as 1000

Percentage

Among

1890-91-92.

1900-01-02.

Decline or
Increase in

Ten Years.

Occupied Males —
(a) in England and Wales as a whole

214

175

-18

321

262

-18

258

202 .

-22

(d) in agricultural districts

157

125

-2O

521

583

+ 12

157

158

THE PREVENTION OF TUBERCULOSIS

The chief available and approximately accurate statistics of
phthisis in relation to industrial occupations are those supplied
in the Decennial Supplements to the reports of the Registrar-
General of Births and Deaths. The results of the last two of
these reports, which are by Dr. Tatham, are given in Table
XXVIII. on previous page. The meaning of the words
comparative mortality figure has already been explained on

P. 155.

Unoccupied males represent a large proportion of invalids,
and we may leave them out of consideration. The excess of
phthisis in industrial over agricultural districts will be noted,

TABLE XXIX. — PHTHISIS

Comparative Mortality Figures of Males aged 25-65, the total Deaths
of all Males at these Ages being taken as 1000

Occupation.

Comparative Mortality Figure.

Percentage
Decline or
Increase in
Ten Years.

1890-91-92.

1 900-01-02. l

Occupied males ....

214

175

-18

General shopkeeper

272

344

+ 26

/i/IO

-28

Tool, scissors, file-maker

390

353

- 9

File-maker .....

467

375

-20

Copper miner ....

384

+ 30

Cutler, scissors-maker .

442

516

+ 17

Tin miner .....

586

838

+ 43

Messenger, porter ....

376

368

_ 2

General labourer (England and

Wales)

295

45°

+ 53

Costermonger, hawker .

5i6

+ o

General labourer (London)

445

+ 19

General labourer (industrial districts)

363

567

+ 56

Inn, hotel servant (agricultural dis-

tricts)

412

410

- 0

Inn, hotel servant (industrial dis-

tricts)

415

426

+ 3

Innkeeper, servant, etc. (London) .

519

443

Inn, hotel servant (England and

Wales)

552

533

- 3

Inn, hotel servant (London) .

705

669

- 10

1 The above corrected figures are supplied through Dr. Tatham's kindness,
before the publication of Part II. of the Decennial Supplement for 1891-1900.

INDUSTRIAL INFECTION 159

and the still greater excess in London. It is also noteworthy
that the decline of phthisis among occupied males is about equal
in industrial and agricultural districts.

In Table XXIX. is shown the relative position of the
chief occupations in association with which fatal phthisis is
particularly prevalent.

Among all occupied males there has been in ten years a
decline of 18 per cent, in phthisis, as compared with a decline
of 22 per cent, in the general population. The great excess
of phthisis among males in towns and the special figures in the
preceding table indicate that a most fertile line of work is
open in the prevention of industrial phthisis. The class of
occupations in which the excess of phthisis is greatest, and
in which this excess is increasing, throw much light on the
lines of preventive work which are indicated. The occupations
in Table XXIX. can be classified under three heads : (i) Those
in which the workers are exposed to irritating and injurious
dust, as scissors-makers, file-makers, tin miners; (2) those
who are particularly prone to alcoholic excess, and are
particularly exposed to infection from indiscriminate expec-
toration, as innkeepers and inn servants ; and (3) those whose
work is casual in character, and who likewise are addicted to
frequenting public - houses, as general labourers, messengers,
costermongers. The occupation of a "general labourer" in-
cludes many loafers, as well as many who have fallen from
skilled occupations owing to illness; and it is difficult to dis-
tinguish between the public-house and the industrial factors,
or to state in the case of how many the ill-health prevented the
patient securing a more stable occupation. It will be noted that
general labourers showed a marked increase, while hawkers and
messengers showed little or no decrease, of phthisis. Innkeepers
and inn servants have in some districts made their previous bad
record worse. Lead miners and file-makers show considerable
improvement, while tin miners, copper miners, and cutlers have
become worse.

The obvious indications for prevention are the diminution or
removal of dust, the substitution of wet cleansing for sweeping,
the use of fans to divert dust from the workshop. The operation
of the Workshops and Factories Acts is gradually improving the
condition of workshops and factories ; but evidence of improve-

160 THE PREVENTION OF TUBERCULOSIS

ment has not yet shown itself to a marked extent in the death-
returns for phthisis among miners and among general shopkeepers,
as is indicated in Table XXIX. Another decade will doubtless
see great advance in the directions indicated above, and will
bring nearer the realisation of the benefit from preventive- work
already being done.

CHAPTER XXII

SUSCEPTIBILITY TO INFECTION

A SPECIAL susceptibility to infection, hereditary or acquired,
is generally regarded as appertaining to those who become

tuberculous, and as being indeed necessary for the develop-
ment of tuberculosis when infection is received. In those showing
this special susceptibility vital resistance to invasion by disease
is supposed to be deficient, or the patient is said to be abnormally
vulnerable to disease. The resulting amount of disease which
will follow infection by the tubercle bacillus will vary on the one
hand according to the number and virulence of the particular
bacilli introduced into the system, and on the other hand
according to the resistance of the patient to invasion.

It is extremely difficult to resolve resistance into its con-
stituent factors, and in fact it cannot be done with exactitude.
In part it consists of innate, and in part of acquired powers, and
the resistance may prove its power after as well as at the time of
the invasion by bacilli. The difficulties of estimating resistance
are particularly great in a disease which is so prevalent as
tuberculosis. Nearly one-ninth of the deaths in the total
population result from invasion by the tubercle bacillus, and,
judging by hospital experience, as many as half of the adults
of the working classes dying of other diseases show indication
post-mortem of some degree of past tuberculous invasion, either
in the lungs or elsewhere. The latter evidence may be re-
garded as indicating either almost universal proclivity to a certain
extent, or some measure of immunity on the part of a very high
proportion of the total population. The former view appears to
me to be nearer the truth, as all degrees of lesions are found in
the above cases, and a very high proportion of the total number
of those who have suffered severely from tuberculosis recover
completely and die from other diseases. In view of the
two aspects of the case it is not surprising that the clinician

162 THE PREVENTION OF TUBERCULOSIS

G. Se"e (quoted by Cornet, p. 328) should say that " la pre*disposi-
tion est un mot pour masquer notre ignorance " ; or that , on the
other hand, J. Kingston Fowler (1898, p. 305) should say : —

Although infection must be regarded as the causa sine qud non, it is
not necessarily of most importance from a practical point of view. If
of a large number of persons exposed to infection only a few acquire
a disease, the susceptibility of the individual becomes a factor in causa-
tion of greater moment than exposure to infection.

The underlying assumption in the position taken up by those
holding the view expressed in the above quotation appears to be
that everybody " exposed " to infection necessarily receives an
efficient dose of infection. The error of this assumption can be
seen by ascertaining what happens when a given number of
persons are exposed to the infection of acute infectious diseases
like scarlet fever, diphtheria, and enteric fever. The instances
best lending themselves to such an inquiry are milk outbreaks
of these diseases, as in these the element of chance appears to
be largely eliminated, and it is reasonable to believe that the
infective material is distributed throughout the milk. In such
outbreaks the families invaded by the disease in question may
be as low as 6 per cent, of those supplied with the infected milk in
scarlet fever, n per cent, in typhoid fever, and 7 per cent, in
enteric fever (Newman and Swithinbank, 1903, p. 268). I have
known two milk outbreaks of scarlet fever in which the percentage
of families affected was considerably lower than 6 per cent.
It has to be noted, furthermore, that the percentage of persons
affected in the families supplied with milk from the infected
source would be much less than the above. The fact is that
in all these diseases a very large proportion of the persons ex-
posed either escape because they do not receive any infection,
just as in battle the majority of soldiers are not shot, or else
receive an inefficient dose of infection, like soldiers who are
touched by spent bullets. The circumstances which limit infec-
tion among those " exposed " to tuberculosis have been already
fully discussed (p. 101).

It should be noted further that in comparing tuberculosis
with the three acute infectious diseases just named, we are in
tuberculosis, with a few imperfect exceptions, restricted to
mortality statistics, while we have complete records of total cases
in the other diseases. The fact that old localised and cured

SUSCEPTIBILITY TO INFECTION

163

tuberculous lesions are so often found at autopsies does not
appear to me to indicate that the majority of the population are
naturally immune to tuberculosis ; any more than it would be
justifiable to state that the majority of the population are natur-
ally immune against the three following infectious diseases,
because in scarlet fever about 95 out of every 100 attacked, in
enteric fever about 85, and in diphtheria 80 to 90 out of every
100 attacked, recover.

When, therefore, we use Allbutt's (1899, p. 1149) phrase of
" openness to consumption," it must be remembered that the
presence of a constant and inherent " openness " in certain
individuals or in certain families is not demonstrated, however
likely it is. It is useful to assume its existence, as a reason for
additional precautions in the cases in which the family or personal
history points to such " openness " ; but in experience it is
difficult if not impossible to obtain exact evidence of such " open-
ness," in which the disturbing factor of excessive exposure to or
excessive dosage of infection can be entirely eliminated.

In Chapter XXIV. we shall deal with those personal con-
ditions, often temporary in character, which appear to diminish
the resistance to infection ; such as the state of nutrition,
alcoholism, overfatigue, and injuries. Age and sex as bearing
on the same problem are discussed in Chapter XXIII., while in
Chapter XXV. the possible influence of heredity in producing
a congenital susceptibility will be discussed.

CHAPTER XXIII
AGE AND SEX

ALL investigators agree that tuberculosis is rare in infancy,
when stated in proportion to the infantile population.
This is true, notwithstanding the national statistics as
to the number of deaths caused during infancy by tuberculous
meningitis and tabes mesenterica. Even when stated in pro-
portion to the total infantile deaths from all causes, the number
verified by autopsies is small. Thus Hervieux at the Paris
Foundling Hospital found on careful post-mortem examination
only ten cases of tuberculosis, or about i per cent, in 996 infants
who had died in the first year of life. Frebelius in ten years
had 16,581 autopsies on infants aged one to four months at
the St. Petersburg Creche, and found tuberculosis in 416, or
0*4 per cent. Schwer, in 690 infants dying under one year of
age, found 44 tuberculous, or 6*3 per cent. These were dis-
tributed as follows : —

263 infants aged I day to 4 weeks — o tuberculous = o per. cent.

123 „ „ 5 to 9 weeks— i ,, = o'8

144 ,, ,, 9 weeks to 5 months — 15 ,, =10*4 ,,

160 ,, ,, 6 months to i year — 28 ,, =I7'5 ,,

The number of deaths from tuberculosis rapidly became
more numerous in the second year of life; and, according to
Papassine, Rilliet, and Barthez, towards the age of five, half
the deaths of children which occur are due to tuberculosis.
This figure does not correspond with the figures for England
and Wales in 1901. If reference be made to Tables XIV. and
XV. it will be seen that the highest recorded death-rates from
tuberculous meningitis (109) and from tabes mesenterica (125
per 100,000) are at ages 0-5, while that from phthisis (315 per
100,000 for males) is at the age -period 45-55. Without
accepting the complete accuracy of the rates for the two first,

it is at least evident that as fatal diseases they are chiefly

164

AGE AND SEX

165

children's diseases, while fatal phthisis is chiefly a disease of
adults. Tatham has drawn attention to the fact that the
age of maximum mortality from phthisis has been postponed
in both sexes as shown below : —

TABLE XXX. — AGES OF MAXIMUM MORTALITY FROM PHTHISIS

( The age-periods in heavy type have the maximum rates > the others
being approximate]

Periods.

Males.

Females.

1851-60

20-25, 25-35, 35-45

25-35

1861-70

25-35> 35-45

25-35

1871-80

35-45

25-35

1881-85

35-45

25-35

1886-90

35-45. 45-55

25-35, 35-45

1891-95 -

35-4S 45-55

35-45

This postponement may be ascribed to a greater saving of
life at those ages formerly most liable to death from this disease,
or to a postponement of death in those who are attacked by
it. Probably both causes are at work. In the following
diagram, taken from Dr. Robertson's annual report for Birming-
ham (1905), the age distribution of the death-rate from phthisis
is shown for both males and females, in Birmingham, Sheffield,
and England and Wales as a whole.

The diagram on the next page enables us also to compare the
death-rate from phthisis in the two sexes, and to see the general
excess of the male rate. It will also be observed that the difference
between the adult death-rate of males and females respectively
is much greater in the two great urban centres than in England
and Wales as a whole, which coincides with the difference noted
on p. 221, where it is pointed out that urban life is not in England
materially less favourable to women than rural life, in respect of
phthisis. In this diagram the female death-rate from phthisis
is seen to be higher in England and Wales as a whole during
a large part of adult life than in Sheffield and Birmingham,
again illustrating the point emphasised on p. 221 as to the failure
of urban conditions of life to raise the female phthisis death-
rate. The contrast with the male death-rates from phthisis in
adult life is very striking.

In Table XII. and Fig. 6 the death-rates from phthisis among

Death Rates from Phthisis in several Age-groups.

DEATH

RATE

ENGLAND AND WALES
SHEFFIELD - - -
BIRMINGHAM • •

- 1890-99- THUS •

- 1890-99- Do. •—

- 19O5 - Do.

FIG. 13. — Death-rates from Phthisis for Males and Females at different Age-
periods in England and Wales, Sheffield, and Birmingham (Robertson)

AGE AND SEX

167

males at each age - period in 1861-70 and 1901 respectively
are compared.

For 1901, the death-rates for children under five have been
calculated in Dr. Tatham's reports for each year of life, and
these have been compared with the official figures for 1871-80
in the following table : —

TABLE XXXI.— PHTHISIS

Death-rates per 100,000 of Population living at each of the First
Five Years of Life

\ \

Period.

O-I. 1-2.

2-3-

3-4-

4-5-

All Ages
under 5.

1871-80 .

141 117

1901 ....

49 44

It has not been thought necessary to subdivide these accord-
ing to sex.

In the following table the male and female death-rates from
phthisis in four successive decennia are given for the first twenty
years of life : —

TABLE XXXII.— ENGLAND AND WALES
Pulmonary Tuberculosis

Period.

Death-rates per 100,000 of Popula-
tion living at Ages

Relative Death-rate of
Females, that of Males
being stated as 100.

0-5-

5-io.

10-15.

15-20.

0-5-

5-io.

10-15.

15-20.

1861-70 .

9.S

105

219

,Sii

173

142

1871-80 .

i34

3«

168

240

176

1881-90 .

21)

129

180

129

204

140

1891-1900

5°

IOO

129

137

215

130

Taking the first five years of life together, it will be
noted that in 1891-1900 the female is 13 per cent, lower
than the male death-rate, a difference which has not hitherto
been explained. The sex difference at ages 0-5 in the three

i68

THE PREVENTION OF TUBERCULOSIS

previous decades varied from 6 to 4 per cent. At ages 5-10
there has been throughout the forty years a greater female
than male rate. At first the excess of the male rate was 10 to
ii per cent., it then increased to 29 per cent., and in the last
decade became 37 per cent. In the next age-period 10-15,
the excess of the female rate is even more striking : in 1 861-80
it was 73 to 76 per cent, higher than the male rate, in the last
twenty years the female has been more than double the male
rate, and the sex difference has increased in the last decade.
At ages 15-20 an inverse process on a smaller scale is visible.
The female rate was 42 to 43 per cent, higher than the male
in the first twenty years, in the third decade it was 40, and in
the last decade it was 30 per cent, higher. It is most difficult
to explain these differences and the changes in the differences,
assuming that they represent actual facts. Sir Hugh Beevor
(1899) thinks that there is a true sex difference as regards this
disease at the ages of rapid growth. The growing lung " is able
to resist infection ; resistance of the growing lung effectively
accounts also for the very regular difference in the sex incidence
of phthisis up to the age of 20 /' He draws attention to the

TABLE XXXIIL— PHTHISIS

Death-rates per 100,000

Relative Death-rates

Males.

Females.

in 1891-1900, the
Death-rate for 1861-70

being stated as 100.

Ages.

1861-70.

1891-1900.

1861-70.

1891-1900.

Males.

Females.

10-

105

15-

220

100

312

129

20-

389

189

397

159

25-

411

237

440

192

35-

417 310

39i

212

45-

388 3H

287

I64

55-

331

262

208

124

65-

204

158

125

75 and upwards

All Ages

254

158

255

121

AGE AND SEX

169

earlier and more rapid general development, and particularly
of the lungs in girls. Thus growth in height in girls is com-
pleted at the age of 15 years, while boys go on growing two
or three years later, and he connects this fact with the higher
female phthisis rate at ages 10-15. However applicable this
explanation may be for the ages 10-15, it can scarcely be appli-
cable to the ages 5-10, in which the female rate is to a less extent
excessive. It is likely that the excess at all ages 5-20 among
girls is partially explicable on the ground that they live a much
less outdoor life than boys, and are much more constantly
exposed to domestic infection.

In the table on the preceding page the death-rates at
different ages from phthisis are given separately for the two sexes
at intervals of thirty years.

It will be seen that at ages 0-5 the decline in the male death-
rate from phthisis has been 4 per cent, less than that in the
female rate ; that at ages 5-15, the decline has been 10 per cent,
greater in the male than in the female rate. At ages 15-20, the
difference is only 5 per cent. At all subsequent ages the decline
has been less among men than among women, this being most
markedly so at ages 45-65.

The relation between the death-rates from phthisis in the
two sexes can be further studied in the following table : —

TABLE XXXIV.— PHTHISIS

Relation of Female to Male Mortality at each Age and in each Period,
that of Males for the same Age and Period being stated as 100

Period.

10-

15-

20-

25-

35-

45-

55-

65-

75 and
upwards.

1861-70

iii

173

142

103

107

| 1891-1900 .

§7

i37

215

130

The relations shown in this table are set out graphically in
Fig. 14.

It will be observed (a) that at the two extremes of age there
is little change in the relation which the male and female death-
rates bore to each other in 1861-70 and in 1891-1900 ; . (b) that
in adult life women have gained considerably more than men;
and (c) that they have lost as compared with boys at ages 5-15.

4>>

FIG. 14.— Female Death-rate from Phthisis at each Age-period, that of Males
at the same Age-period being stated as 100

1861-70 . . •— • •— •

1891-1900 . . •-«-•---•---•

AGE AND SEX

171

It is necessary to bear in mind that all the preceding figures
deal with deaths. The date at which infection was received
may have been less than a year, or may have been very many
years before death (Chap. X.). Thus the excessive death-rate of
girls aged 10-15 mav De in part due to the strain of the changes
undergone at puberty, — a strain greater than in boys, — calling
latent infection into activity, as well as to recent infection caused
by their indoor habits, as suggested on p. 169.

CHANGES IN THE SEX INCIDENCE OF PHTHISIS. — This subject
deserves further study from the historical standpoint. In
England and Wales the female death-rate from phthisis has been
lower than the male rate from 1866 onwards, in Massachusetts
it was as high as or higher than the male rate until 1896. In
Prussia since 1876, when statistics first became available, the
male has always been higher than the female rate. In Scotland
the female was higher than the male rate until 1885, when the rates
for the two sexes were nearly equal. In more recent years the
position of the two has changed without consistency, but from
1898 onwards the female has always been lower than the male

TAB*LE XXXV

The Relative Male and Female Death-rates from Phthisis^
that of Males being stated as 100 *

England
and Wales.

Massa-
chusetts.

Providence,
U.S.A.

Prussia.

Male.

Female.

Male.

Female.

Male.

Female.

Male.

Female.

1851-55

100

107

IOO

1856-60

100

108

IOO

123

IOO

131

1861-65

100

104

IOO

109

IOO

91?

1866-70

IOO

103

IOO

112

IOO

IOI

1871-75

100

IOO

109

1876-80

IOO

119

IOO

112

IOO

80 in 1876

1881-85

IOO

114

IOO

IOI

IOO

84 ,, 1881

1886-90

IOO

106

IOO

83 „ 1886

1891-95

IOO

104

IOO

85 „ 1891

1896-1900

IOO

83 „ 1896

and 1901

1 The correction of the death-rates for males and females respectively for
differences due to age distribution of population in the two sexes was* not practi-
cable. It is unlikely that such correction would seriously alter the comparisons
in the above table.

172

THE PREVENTION OF TUBERCULOSIS

rate. In Ireland from 1864 to 1873 the male and female rates
were close together ; afterwards the female became increasingly
higher than the male rate. In the last few years the two rates
have approached again ; in 1903 the female death-rate was 2'2
as against 2*1 per 1000 for males. In the table on preceding page
the relative sex incidence of the death-rate in Massachusetts
and England is given for a series of years.

MORTALITY IN THE Two SEXES IN URBAN AND RURAL LIFE.—
The influence of urban or rural conditions of life on the relation
of the male to the female phthisis rate is also of interest. For
Prussia this is seen in the following table : —

TABLE XXXVL— PRUSSIA

The Relative Male and Female Death-rates from Phthisis,
that of Males being stated as 100

Year.

Towns.

Rural Communes.

Male.

Female.

Male.

Female.

1876
1881
1886
1891

100
100
100
100
IOO
100

74
76

73
74

IOO
IOO
IOO
IOO
IOO
IOO

88
89
90
95
93
95

1896
1901

This table fits in with the facts set forth on pp. 220 to 2*24,
which showed that the female death-rates from phthisis are
nearly equal in rural and urban counties of England, while the
male death-rates are much higher in urban than in rural counties.
The comparison in the case of England and Wales can be pursued
into the different age-periods. The result is shown in Figs. 15
to 18, the data from which are taken from p. xcvi of Dr. Tatham's
Letter to the Registrar-General (1905).

(a) Comparison of urban and rural life for males. In Fig. 15
it will be observed that throughout the early part of life up to the
age-period 25-35 the male phthisis rate is higher in rural than
in urban counties. After that age the rural rate ceases to rise
and falls slowly, while the urban rate rises, being highest at the
age-period 45-55. The evil effects of urban conditions of life

AGE AND SEX

173

MALES

Death Rate, per Million

1 1 !

<>>

<X;

ffi

•MM

ise

fc» •

— .

—

1 Iftil r

I6«

/ •

!/ 1^

I 113

'•\Q

/I 7I

//i r

fl3

5i//i

i\i i

/*68f |

4lfi

E9€

Ifi7

» J

UNDER
5
YEARS

5-
-10

w-
-/s

-25

25-35

35-45

45-55

55-65

65
UPWARDS

FIG. 15. — 1905. Death-rate from Phthisis per million of Males living at each
Age-period, in Urban and Rural Counties

174

THE PREVENTION OF TUBERCULOSIS

and work in increasing the male phthisis rate at the higher ages
are well shown.

(b) Comparison of urban and rural life for females. In Fig. 16

FEMALES

Dearth Rate, per Million

l£2J

68
/

J*5

(

136"

<J;

<•

*"V

JI7J

'flS

IIO£

^^*

- .

-„

02 £

I i
t f

87!

79(

£2J

•\

24£

*>4

161

UNDER
5
YEARS

5-
-10

10-
-15

15-
-20

20-
-25

25-35

35-45

45-55

55-65

65
UPWARDS

•1500

1000

500

FIG. 1 6. — 1905. Death-rate from Phthisis per million of Females living at
each Age-period, in Urban and Rural Counties

the contrast to the male experience is very evident. At the
ages 5-15 urban and rural experiences are almost identical.
From 15-20 to 25-35 the rural phthisis rate among females is
much higher than the urban. From that age-period onwards
the rural is lower than the urban rate.

(c) Comparison of males and females in urban districts. The
failure of the female rate to rise to the same extent as the male
rate at ages after 20 is well seen in Fig. 17.

In Fig. 1 8 is given a similar comparison for rural
counties.

AGE AND SEX

URBAN

Death Rate per Million

29'

f2&<

I9(

Ifif

^ /

I6i

13(

(

^l7

1^0

(

?9/

)

«8

*IG

S7B

Iftfi

167

UNDER
5
YEARS

5-
-10

10-
'15

15-
-20

20-
-25

25-35

35-45

45-55

55-65

65
UPWARDS

FIG. 17.— 1905. Death-rate from Phthisis per million of Males and Females
living at each Age-period in Urban Counties

176 THE PREVENTION OF TUBERCULOSIS

RURAL

Death Rate, per Million

2000

1500

1000

500

FIG. 18. — 1905. Death-rate from Phthisis per million of Males and Females
living at each Age-period in Rural Counties

CHAPTER XXIV

PERSONAL CONDITIONS LOWERING RESISTANCE TO

INFECTION

IT has already been stated, that not only differences in age
and sex, but also more or less temporary individual con-
ditions, affect the proclivity to tuberculosis. Of these
fatigue, injuries, and attacks of diseases other than tuberculosis
are important ; and the state of nutrition, with particular refer-
ence to alcoholism, also needs discussion.

FATIGUE. — Over-exertion is well known to predispose to
infection. The common method of origin of an ordinary catarrh
is an illustration of this, and there are numerous instances in
experimental bacteriology. Thus Charrin and Roger showed
that normal rats, which are but slightly susceptible to anthrax,
become highly susceptible when fatigued by working at a tread-
mill.

Clinically a history of bodily or mental over-exertion, of pro-
tracted emotional excitement or anxiety, as after a competitive
examination or the prolonged nursing of a sick relative, is a
frequent prelude of acute phthisis. On this point Dr. BufEon-
Fanning (p. 24), says : —

To my mind there are few causes more powerful to determine the
outbreak of pulmonary tuberculosis than physical over-exertion. In at
least 10 per cent, of my patients the disease seemed directly attributable
to their having overdone themselves. A feat of endurance is apt to
overstrain the constitution, and break down the defences of an apparently
healthy man against the tubercle bacillus. It had already gained, we
assume, a footing in his system, and only waited an opportunity to mani-
fest its activity. I have been struck by the frequency with which con-
sumption attacks men who have distinguished themselves in various
athletic pursuits. This remark particularly applies to such sports as tax
the powers of endurance, such as long-distance bicycle riding or running,
rowing, or, in fact, any exhausting exercise. It is important to recognise
that, although such exercise be taken in the open air, it is conducive to
the development of consumption if it entails exhaustion or fatigue.

i78

THE PREVENTION OF TUBERCULOSIS

INJURY. — The apparent influence of local injury in deter-
mining the site of tuberculous disease of bones and joints is well
recognised. Injury to the chest wall has sometimes appeared
to light up active phthisis. There is no reason to doubt that
injury may, by lowering the local phagocytal influence, enable
latent tubercle bacilli to assume active life.

DISEASES OTHER THAN TUBERCULOSIS. — Tuberculosis is
commonly associated with certain diseases, especially with
chronic insanity. The death-rate from phthisis is very ex-
cessive in the insane in asylums. From the pathological evidence
collected by Dr. Mott it is clear that a very large part of this
tuberculosis was present in a latent condition when the patients
were admitted to the asylums ; and that the tuberculosis must
be regarded as acting in insanity as it does in diabetes by
hastening death, the devitalised condition of these patients
enabling the tubercle bacillus to proceed with its ravages un-
molested. The annual death-rate from tuberculosis in borough
and county asylums is about 16 per 1000 occupants, which is
more than seven times as high as that in the total adult population
of England and Wales.

Certain acute infectious diseases, especially influenza, whoop-
ing-cough, measles, and to a less extent scarlet fever and enteric
fever, undoubtedly favour the occurrence of tuberculosis. Prob-
ably they act in two ways : (a) in all these diseases irritation of
mucous membranes and denudation of their epithelium is
caused, and the way is opened for the entrance of the tubercle
bacillus ; (b) probably these diseases act more commonly by

TABLE XXXVII.— ENGLAND AND WALES
Annual Death-rate per million of Population

1888.

1889.

1890.

I89I.

1892.

1893.

Influenza
Phthisis

1568

2
1573

157
1682

574
1599

.58

Note, — In 1890, although probably doctors had not yet begun to recon
deaths as due to influenza to the full extent which the facts justified,
was already widely prevalent, and the sudden excess of deaths ascril
to phthisis occurred in this year. Probably many phthisical patienl
with an unstable tenure of life died as the result of intercurrent influenza.

LOWERING RESISTANCE TO INFECTION 179

causing swelling and infiltration of lymphatic glands, often
already containing tuberculous foci, the migration from which of
tubercle bacilli to internal organs is thus greatly favoured. The
influence of influenza in increasing the death-rate from phthisis
is shown in our national death returns. As a rule, the annual
death-rate from phthisis shows no epidemic peaks, but declines
smoothly by a small percentage year by year. This course was
interrupted in the years 1890-91 in which influenza after a long
interval again became epidemic, as shown in the table on the
preceding page.

Common catarrhs are credited with an important influence in
causing phthisis, especially when neglected. Possibly they act like
acute specific fevers by denuding epithelium and by causing
glandular enlargements, thus setting free encysted tubercle
bacilli. More often the real connection is one of identity. What
is regarded as a " severe cold," a slight " attack of influenza/'
or a " touch of bronchitis," is in fact an attack of pulmonary
tuberculosis, from which the patient temporarily recovers, with
frequent relapses. Whether there be any connection between
neglected catarrhs and phthisis or not, it is certain, as pointed
out by Clifford Allbutt, that the belief in it has had a lamentable
effect on the treatment of the latter disease. Indoor confinement
and stuffy rooms have been prescribed, when abundant fresh air
was indicated. The common indication for treatment both in
catarrh and in febrile phthisis is absolute rest with as close an
approximation to open-air conditions as possible.

The association between bronchitis and phthisis has been
much discussed. Many cases of senile phthisis are overlooked
on account of the presence of emphysema. It is likely that
many cases starting as true bronchitis have phthisis engrafted
on this disease. This is especially so in many occupational
diseases.

MALNUTRITION. — As shown on p. 230, good nutrition is
considered by some authorities to play a very important part
in the prevention of tuberculosis, although the evidence given
on pp. 230 to 243 does not justify the conclusion that on a
national scale any marked inverse relationship between phthisis
and nutrition holds good. The same remark applies" to ex-
posure to weather, cold, and hardship, which may be regarded
as representing so much excessive loss of benefit derivable

i8o THE PREVENTION OF TUBERCULOSIS

from a given amount of food. Thus Ransome (1890, p. 50)
says : —

The Highlanders, who inhabit well-built houses on the mainland of
Scotland, are subject to the same fate as the other inhabitants, whilst
the ill-fed, ill-clothed fishermen of St. Kilda and the Hebrides, who are
of the same race, hardly ever contract the disease.

In another paragraph on the same page Ransome gives a
second illustration, which may also be quoted :—

The terrible mortality from phthisis that prevailed at one time amongst
the finest soldiers of the British Army was certainly not brought on by
starvation or misery. It occurred for the most part when they were
not on active service, but in a time of peace, when they were well fed and
well cared for so far as their bodily comfort was concerned — far better,
in fact, than the half-starved workpeople and labourers, who only died of
the disease at one-third the rate they did.

The experience of Ireland, given more fully on pp. 217 and
233, tells the same story. Between 1870 and 1903 the wages of its
agricultural labourers have increased 42 per cent., while the cost
of food has greatly diminished and its death-rate from phthisis
has increased.

Dr. Stafford of the Irish Local Government Board has
recently given the death-rates from phthisis in the years
1900-02 in the two Dublin Poor Law Unions and in the county
of Mayo respectively. In Dublin the phthisis death-rate is 3-4
and in Mayo 1*4 per 1000. He adds that

for scantiness of the means of subsistence the general condition of the
inhabitants of County Mayo could scarcely be surpassed. It is clear,
therefore, that poverty alone may be present in an acute form and on a
large scale without producing an excessive mortality from tuberculosis,
and that some other factor or factors as well as poverty exercise a determin-
ing influence in producing the excessive death-rate from tuberculosis.

It is important to bear in mind these illustrations following
from the fact that circumstances other than differences of
nutrition affect the proclivity to tuberculosis. They show that
no general measures of improvement in well-being by themselves
suffice to control the disease. But beyond question malnutrition
favours tuberculosis, and while the evidence in Part II. amply
shows that other factors are more important, no system of
measures for controlling tuberculosis can be regarded as final
which omits to do what is practicable for preventing malnutrition.

LOWERING RESISTANCE TO INFECTION 181

ALCOHOL. — That alcoholic indulgence favours the occurrence
of phthisis is shown by abundant evidence, and is well recognised.
Thus the late Professor Brouardel (1901) of Paris said :—

Alcoholism is in fact the most powerful factor in the propagation of
tuberculosis. The most vigorous man, who becomes alcoholic, is without
resistance before it.

Although some have obtained opposite results, there are
many experiments on record tending to show that infections in
general are more rapid and more grave in alcoholised animals.
Drs. Achard and Gaillard found in experimenting on rabbits
that giving alcohol hastened the progress of experimental tuber-
culosis. For the human being Landouzy has expressed the
influence of alcoholism as follows : " I'alcoolisme fait le lit de
la tuberculose."

Alcohol and phthisis are related as indicated above, through
the diminished resistance to the disease caused by alcohol, and
with that we are chiefly concerned in this chapter. Alcoholic
indulgence, and still more the occupation of selling alcoholic
drinks, commonly expose persons to more frequent infection ;
and this is a prominent factor in causing the excessive death-
rate from phthisis in certain occupations (p. 159).

CHAPTER XXV

HEREDITARY DISPOSITION TO PHTHISIS

SO far we have been chiefly concerned with factors of causa-
tion which are all more or less ascertained and defined.
The influence of heredity differs from these in being still
more or less sub judice.

It is considered as acting in two ways : by direct trans-
mission before birth from parent to infant of the germs of disease ;
or by the transmission from parent to offspring of a special
weakness or openness rendering certain persons more liable to
infection than others.

THE DIRECT TRANSMISSION OF TUBERCULOSIS from parent
to child may occur before birth, either germinally — a very rare
phenomenon — or during intra-uterine life, a more common, but
still rare, event.

The passage of the tubercle bacillus through the placental
tissues to the foetus has been proved by a number of pathologists.
Thus Johne found tubercles in the lungs and bronchial glands of
the eight months' foetus of a tuberculous cow. MacFadyean found
cheesy foci in the liver and portal glands of a five days' old calf.
Similar cases have been described in the human foetus. Frankel
(1906) thinks that the danger of haematogenous infection through
the placenta is commonly understated. He quotes Schmorl,
who found tuberculous nodules in 9 out of 20 or 45 per cent, of
the placentas of tuberculous women examined by him ; and
these were found not only in cases of miliary tuberculosis or
advanced phthisis, but also in a case of incipient phthisis. It
is possible, furthermore, that the instances in which obvious
tuberculous lesions are found in the new-born child do not cover
the entire ground. Other infants may have latent tuberculosis,
which develops into obvious disease later in life.

This view is commonly associated with the name of Baum-

garten, though it was held before his day. He believes that

182

HEREDITARY DISPOSITION TO PHTHISIS 183

either germinal or intra-uterine transmission of infection is the
most common cause of tuberculosis, and that long latency of the
infection is the rule rather than the exception. He goes further,
believing even that a person may have been infected by trans-
mission through two generations from a tuberculous grandparent.

The views of Baumgarten, apart from the last-named point,
are supported by the fact that microscopic examination of
the liver and inoculation experiments with fcetal tissues show-
ing no naked-eye evidence of disease have occasionally shown
the presence of tubercle bacilli. Baumgarten considers long
dormancy of tubercle bacilli in lymphatic glands, the medulla
of bone, etc., as common, the young tissues of growing animals
having special resisting power against the bacilli. His view
involves the unlikely supposition that a very large part of the
human race carry within them tubercle bacilli at birth. At the
same time the analogous case of congenital syphilis, with long
latency of an infection acquired before birth, indicates that
congenital tuberculosis is within the range of possibility. It is
possible, as J. K. Fowler has suggested, that evidence will accumu-
late in favour of the view that sometimes tuberculosis of the
glands, joints, and bones in children may have been transmitted
from the parent and remained dormant for several years. To
prove such cases it would be necessary to show that the mother
was tuberculous, and that there had been no exposure to infection
after birth. In the absence of evidence on the latter point,
either the ordinary view of infection after birth, or the view that
infection was acquired before birth, would be tenable.

The fact that visible tuberculosis is more commonly found

TABLE XXXVIII

|?1

v«

«J

0 c

ro w

o J3

3 SQ

Q£

~ 0

«l

:>i

*-• o>

ijS

tt>

Number of autopsies .
Number with tuber-

184

250

i89

1 60

134

1542

culous changes

...

263

Per cent, of total .

...

6-1

10-5

17-0

277

267

29-6

ITf

2,5

17-0

184 THE PREVENTION OF TUBERCULOSIS

with each additional month after birth, may be explained either
on the supposition that early-life tuberculosis is in the main
acquired after birth ; or by assuming that ante-natal tuber-
culosis remains long latent so far as symptoms are concerned.
The following illustrations on this point will suffice. Cornet
(1904, p. 307) gives the figures on the preceding page relating to
a number of autopsies made on children under 5 years old dying
in children's hospitals in Berlin.

These figures clearly show that whether infection is received
before or after birth, visible changes are not usually shown in
the body until some months later. (The figures in the above
table must not be regarded as giving any indication of the true
frequency of fatal tuberculosis in children. To do this it would
be necessary to compare the deaths from this disease with the
number of children living at the same ages. The figures do,
however, show its rarity in the first few months of life.)

Veterinary results are to a like effect. Thus Cornet (1904,
p. 308) gives the distribution of tuberculosis among cattle in
Saxony, where the inspection of meat is compulsory, as follows : —

Of 120,490 calves up to 6 weeks of age
,, 665 cattle from 6 weeks to I year

6,328 ,, ,, I to 3 years old
i» !3,3°7 „ „ 3 to 6 „ „
,, 11,101 ,, over 6 years old

3, or 0*002 per cent.

i, M 0-15
440, „ 6-9
1,285,,, 97
1,881, „ 16-9

The most probable interpretation of the preceding facts is
that post-natal infection is the usual source of tuberculosis,
though ante-natal infection occasionally occurs, and it may be
somewhat more frequent than is generally recognised.

HEREDITARY PREDISPOSITION. — Phthisis is usually regarded
as a typically hereditary disease, in the causation of which
family predisposition plays a large part. The extent to which
heredity is held to operate has diminished as our knowledge of
the causation of tuberculosis has become more exact. The most
prevalent view is contained in the following statement by Drs.
C. J. B. and C. Theodore Williams (1887, p. 58) :—

Family predisposition has by general consent held a very prominent
place, but the value of its influence in the causation of phthisis has been
modified of late years by the fuller recognition of other causes which had
been to some extent overlooked — such as damp, inflammatory attacks,
etc. These and other direct sources of phthisis must exercise in our calcu-
lations a depreciatory influence on the amount we assign to hereditary

HEREDITARY DISPOSITION TO PHTHISIS 185

transmission, and numerous cases of this disease which have hitherto
been held to originate in a consumptive ancestry, will now be traced to
a nearer and more direct cause. Nevertheless, no small number of cases
owe their origin to hereditary predisposition, though it is not always easy
to demonstrate their hereditary character. Its exact value as a predis-
posing agent, its mode of transmission, the varieties of the disease in
which its influence is most apparent, — all these and other points of interest
are by no means settled questions, but still open to further inquiry.

Similarly Dr. S. West (1902, vol. ii. p. 449) states that
" recent additions to our knowledge of tuberculosis have greatly
modified the views held as to the influence of inheritance in
phthisis " ; but after giving statistics he concludes that " family
predisposition is an essential factor in phthisis, though probably
not exerting so important an influence as has been hitherto
believed."

The evidence on the strength of which it is considered that
hereditary predisposition forms an important factor in the
causation of phthisis consists usually in showing that a large
percentage of the parents and other relatives of the total con-
sumptives had also suffered from the same disease. West (p. 449)
says that about 28 per cent, of the total cases taken at random
yield, on an average of a large number of cases, a history of
phthisis in the parents, and about 25 per cent, more in collateral
relatives. Walshe (1871, p. 461) in a careful investigation of
162 cases found that 26 per cent, of them had one or both parents
similarly diseased. J. E. Squire (quoted by Fowler, p. 312)
gives 12,509 cases of phthisis, showing in 24^8 per cent, of these
cases that one or both parents had been consumptive. When
grandparents and collaterals were included, the percentage of
heredity became 62*3. Williams (1887, p. 63) thinks that " an-
average of 12 per cent, for direct hereditary transmission, and of
48 per cent, for family predisposition, are not unfair estimates
for the upper classes." Wilson Fox found a history of direct
inheritance in 33 per cent, of hospital cases.

Facts like the above, although they are commonly regarded
as good evidence of hereditary influence, are almost valueless
unless further tested. This was realised long ago by Walshe
(p. 461), who observed about his own results : —

Does this result, that about 26 per cent, of my tuberculous patients
came of a father or mother, or of both parents, similarly diseased, prove,

186 THE PREVENTION OF TUBERCULOSIS

even in this limited proportion, the reality of hereditary influence
in the production of the disease ? I think not. It shows that of a
given generation (fc) about 26 per 100 came under ascertainable con-
ditions of a tuberculous parent (generation a). But this ratio of 26 per
loo might be, and probably is, no higher than that of the tuberculised
portion of the population generally.

In another paragraph (p. 54) he says : —

If it be true, as always taught, that one in every three persons dyii
from all diseases indiscriminately in the Paris hospitals has tubercle ii
the lungs, the existence of an almost universal family taint becomes
unavoidable inference.

Phthisis, like scarlet fever, is a common and an infectioi
disease, and the futility of depending on statistics like the
already quoted, as evidence of hereditary predisposition, ma''
be illustrated from the latter disease. For some years pas
I have ascertained in the course of my official experience th<
family experience of households invaded by notifiable infectious
diseases ; and I recently abstracted 100 family histories of
scarlet fever in which the records were sufficiently complete
to be trustworthy. Out of every 100 patients belonging
to different families, both parents of seven patients had
suffered from scarlet fever previously, the fathers only of
sixteen patients and the mothers only of nine patients had
suffered from scarlet fever, while in 68 per cent, neither parent
had suffered from this disease. The resemblance to the
percentages for tuberculous families is striking, and both
sets of figures alike fail to prove any true hereditary predis-
position.

HEREDITARY PREDISPOSITION OR INFECTION. — It is easy
to prove heredity in the case of a disease like haemophilia,
where (a) the disease is rare and presumably not infectious,
and (b) either all or almost all the cases occur among thos
whose ancestors had the same disease. But in phthisis we have
to deal with a disease which in the first place is infectious, an<
would therefore give no such clear evidence of heredity, eve
if heredity were potent ; and which, in the second place, is vei
common, causing in the general community about one out oi
every twelve male and one out of every seventeen female
deaths from all causes. Since it is infectious, one cannc

HEREDITARY DISPOSITION TO PHTHISIS 187

expect all the cases to be limited to families with hereditary
taint, however strong this influence may be, and in actual
fact it is not so limited. Finally, even if it be shown that the
number of adult deaths from phthisis amongst those with a
tuberculous family history is in that class much greater than
the number among a corresponding number of the general
population similarly situated as to age and sex, it does not
necessarily follow that this is due to hereditary predisposition.
It may result from greater exposure to infection. There cannot
be said to exist satisfactory data enabling this doubt to be
cleared up. The nearest approach to such data is embodied
in a " first study " of the statistics of phthisis by Professor
Pearson, in which the family history of a hypothetical random
sample of the general community is compared with that of
consumptives. Even these, however, fail to distinguish between \y
family infection and the inheritance of family predisposition.
An examination of the mathematical method used by Professor
Pearson would be outside the scope of the present discussion ;
but it is important to note as a matter involving no criticism
of method, that his results depend in part upon hypotheses
which may not be accepted generally as justified, and upon
ascertained data which may be regarded as too few to warrant
conclusive inferences. Indeed he himself states : " This investi-
gation does not profess to be more than preliminary, and its
results need confirmation when much more numerous data are
available." He proceeds, however, to state that : " I feel fairly
confident that for the artisan class the inheritance factor is far
more important than the infection factor." This statement
goes beyond Professor Pearson's data, and his assumption that
in towns the artisan classes can scarcely escape infection,
except by the absence of the tuberculous diathesis is unproven.
By infection he doubtless means efficient infection, and no point
is clearer in the pathology of tuberculosis than that efficient
infection depends largely on the dosage of infective material.
The considerations in Chapter XIII. indicate that infection is
much more limited and localised than is usually supposed.
It is to be hoped that Professor Pearson's most interesting
researches may be continued, and that he may receive* in the
future more ample and more complete data from physicians
than he has hitherto had placed at his disposal. It would

i88 THE PREVENTION OF TUBERCULOSIS

be a great advantage if, in such a research on a larger scale,
consumptive families could be classified into groups according
to the length of interval between the termination of one case
and the earlier symptoms of successive cases in the same
family.

The question asked by Burton-Fanning (1904, p. 22) cannot be
regarded as a serious contribution towards the solution of the
problem, without further detailed evidence than is given. He
asks : —

If it is entirely a matter of infection and not of heredity, why are the
members of the family picked out, and other occupants of the house, such
as the servants, avoided ?

In the context this writer gives no evidence to show that
the servants actually escape. Instances are on record in which
they are known to have fallen victims after prolonged un-
skilled attendance on consumptives, though the frequent
migrations of servants render it difficult to obtain such evidence.
Before importance can be attached to this question, there must
be evidence on a considerable scale that with fairly equal degrees
of exposure to infection (both as to duration and intimacy)
servants escape when relatives suffer. The remarks in Chapter X.
on long latency have also to be borne in mind in interpreting
results.

On the whole, we shall probably not err greatly if we agree
with Koch's statement (1901, p. 26) that

great importance used to be attached to the hereditary transmission of
tuberculosis. Now, however, it has been demonstrated by thorough
investigation that, though hereditary tuberculosis is not absolutely
non-existent, it is nevertheless extremely rare, and we are at liberty, in
considering our practical measures, to leave this form of origination
entirely out of account.

THE PRACTICAL ASPECTS OF HEREDITY IN TUBERCULOSIS.
— The statement last quoted from Koch must command par-
ticular approval, when considered in relation to administrative
measures. From the standpoint of practical public health
administration, if it were ultimately to be established that
heredity exercises a greater effect on the transmission of tuber-
culosis than has hitherto been attributed to it, the measures oi

HEREDITARY DISPOSITION TO PHTHISIS 189

precaution indicated by this result might be increased in number,
but none of those of which the adoption is recommended on
other grounds would become more safely negligible than they
are now considered to be. The inheritance of a disposition
to tuberculosis if demonstrated as a general phenomenon would
show the presence in the community of a larger number of
susceptible persons than could be inferred from other con-
siderations. The existence of this larger number of susceptible
people would call not for the neglect but for the more careful
enforcement of the precautions by means of which susceptibility
is prevented from developing into actual infection. The logical
alternative is to kill off the susceptible stock or, as has been sug-
gested, to allow them to infect their susceptible brethren and
together with them perish of their disease. Such proposals _J
have only to be stated in their crude terms in order to be
apprehended and reprehended as an unsocial negation of
civilisation.

MARRIAGE OF AND BETWEEN CONSUMPTIVES. — As the matter
is not separately dealt with in Part III. of this book, it is con-
venient to add here a note as to the practical bearing of the
preceding facts and considerations on the marriage of, and
particularly on the marriage between, consumptives. Assuming *
that advice based on physiological and medical considerations
will be allowed to carry weight in a matter in which the affec-
tions alone as a rule are allowed control, it is evident that in
many instances the marriage of those of consumptive stock is to
be deprecated, especially when both parties come of such stock.
On the other hand, when it is remembered that in at least 30 per
cent, of the adult population there is a history of consumption
in the antecedents, a sweeping condemnation of such marriages
can be justified only if it is shown that this percentage is made
up by a much higher percentage in a relatively small portion
of the total population. The measure of the actual danger^,
in any given instance would be made on the strength of a number
of facts : —

(1) At what age did phthisis show itself in the preceding
generation ? Has the man or woman now concerned passed
that age ?

(2) What is the interval since the man or woman now con-
cerned was last exposed to infection from the consumptive

igo THE PREVENTION OF TUBERCULOSIS

relative ; and prior to that what was the duration and extent
of exposure ?

(3) Are the circumstances of the person now being advised
such as are likely to call into activity any latent infec-
tion ?

CHAPTER XXVI

CONDITIONS OF ENVIRONMENT LOWERING RESISTANCE
TO INFECTION; SOCIAL MISERY; AND INSANITARY
CIRCUMSTANCES

TUBERCULOSIS is most prevalent and most fatal under
conditions of social misery, and when the surroundings
of the patient are insanitary. It is not surprising, there-
fore, that it is frequently regarded as due to social misery, and
that for its prevention many reformers are satisfied with an
appeal for general social reform, without attempting to analyse
the constituents of social misery which in particular favour
tuberculosis. Without attempting any complete analysis of
social misery and of the insanitary circumstances so closely
associated with it, it may be said that in it are united in a vicious
circle, ignorance, privation, and suffering, and that efforts against
any of these will undoubtedly help to reduce the amount of
tuberculosis. These factors are in themselves complex. Thus
privation involves the operation of several influences, to each of
which it is difficult to apportion its true weight. Underfeeding
and defective nutrition (pp. 179 and 230) undoubtedly play a
part in producing the excess of tuberculosis found in the poor,
though only a relatively small part. Neglect of the ordinary
rules and precautions of a hygienic life, as to cleanliness, wearing
of suitable apparel, precautions after exposure to rain and
weather, and so on, doubtless also favour tuberculosis, though
no preponderant weight in the balance can be ascribed to them.
Unfavourable sanitary circumstances of the poor, especially
housing, play their part ; this is gauged in relation to other
factors — so far as the data permit — on pp. 224 to 229. Domestic
overcrowding has already been fully considered on pp. 146 to 149.
It undoubtedly plays a very large share in the production of
tuberculosis ; and to this factor more than to any other attention

I92 THE PREVENTION OF TUBERCULOSIS

is required, if the decline in the death-rate from tuberculosis is
to be made more rapid than at present.

As no special chapter in Part III. is devoted to ordinary
sanitary measures in relation to the prevention of tuberculosis,
it is convenient to consider here the measures practicable against
it. There are two ways in which overcrowding can be abated :
one is the slow measure of official inspections, followed by official
notices in the instances in which overcrowding is detected.
Those who have official experience know the limitations of this
method, valuable though it is. Before the limit of legal over-
crowding (about 350 cubic feet for each person) is reached, there
may be social overcrowding of a most objectionable character,
over which official inspection can exercise no control. Even
when there is suspicion of legal overcrowding, it is very difficult
to obtain conclusive evidence of its existence, except in lodging-
houses in which night inspections are possible. Under these
circumstances official remedies against overcrowding are bound to
operate slowly, although much improvement has already been
accomplished.

The alternative remedy is the removal from the congested
dwelling of those liable to convey infection. This has been
done for typhoid and typhus fevers and for small-pox, and has
led to an immense reduction in their prevalence. In scarlet fever
and diphtheria similar measures have not been successful to
an equal extent, because of the failure to track slight cases of
these diseases, which remain at home or in school spreading
infection. In phthisis, as shown in Part II., the evidence
indicates that similar removal of advanced cases from the
poorest homes has been a predominant cause of the great decline
in the death-rate from that disease already secured.

Overcrowding is nearly always associated with other
house conditions — such as defective light and air and absence oi
thorough ventilation — which undoubtedly protract the extra-
corporeal life and retard the destruction of the tubercle bacilli.
Do they do more than this ? Some experimental results appe;
to indicate that they may. Thus Trudeau inoculated a numl
of rabbits with equal doses of tubercle bacilli ; half of these wen
allowed to run free in the open air, and the remainder were place
in a damp hole to which sunlight had no access. Both sets oi
rabbits were killed at the same time, and it was found that th(

LOWERING RESISTANCE TO INFECTION 193

first had recovered or only had slight lesions, while the second
had extensive tuberculosis.

Ransome's experiments (1895, p. 15) point in the same
direction. In 1889-90, experimenting with Dreschfield, he showed
that

the air of a poor cottage in Ancoats, with poor ventilation and undrained
basement, in which several cases of phthisis had occurred, was able to
preserve unchanged the virulence of tuberculous sputum for two or
three months at least, but that the same sputum exposed freely to air
and light in a hospital for phthisical patients and also in a well-lighted,
well-drained, and well-ventilated house entirely lost the power of com-
municating the disease to guinea-pigs by inoculation. A further research
carried on in 1894 in conjunction with Professor Delepine proved that
less than two days' exposure to air and light with only one hour of sunshine
was sufficient to destroy the virulent power of tuberculous sputum when
it was exposed in a clean, well-drained, well-lighted house. Evidently
in the air of the Ancoats cottage there must have been some form of
organic impurity favourable to the life of the bacillus.

Whatever be the interpretation put upon these experiments,
there can be no difference of opinion as to the ill-effects of over-
crowding, defective light and air, absence of thorough ventilation,
and still more of domestic uncleanliness in favouring the occur-
rence and spread of tuberculosis. Probably these factors operate
chiefly by facilitating the spread of infection ; but it is possible
that they also tend to devitalise the occupants of such houses
and render them more ready victims of infection. Whatever
opinion be held on this point, the indication clearly is to adopt
the most strenuous efforts to remove these evil conditions,
wherever found.

CHAPTER XXVII
CLIMATE AND SOIL'

. — The anxious inquirer after indications as to the
climate associated with the lowest death-rates from tuber-
culosis would not obtain any satisfactory hints from the
statistics scattered throughout this book, or found elsewhere. It
maybe said in brief that there is scarcely a climate which has not
been looked upon at one time as predisposing to this disease, and,
at another as curing it. There is no certain evidence that it is
less prevalent at high than at low altitudes, except in so far as
the former are usually more isolated and less densely populated
than the latter. Hirsch (vol. iii. pp. 197-8) has said :—

The disease occurs c&tevis paribus in all geographical zones with
uniform frequency; equatorial and subtropical regions are visited with
consumption not less than countries with a temperate or an arctic
climate. . . .

The only statements that can be made in this connection
with absolute certainty are that

1. Anything favouring an open-air life diminishes tuberculosis.

2. Tuberculosis is less prevalent in the less densely populated
and more isolated communities.

SoiL.1 — In regard to soil, there is almost equal uncertainty.
Thorne is quoted by Roberts (1902) as saying that in the pre-
vention of pulmonary tuberculosis " nothing would do good
unless people refused to live on a damp subsoil." A damp
subsoil is stated in all text-books of hygiene to be a most im-
portant cause of phthisis.

The proved infectivity of the disease makes it somewhat
difficult to adjudge what importance should still be attached to
soil in relation to its causation. It is therefore desirable to

1 The greater part of the rest of this chapter appeared as an article on " The
Influence of Soil on the Prevalence of Pulmonary Phthisis "(Practitioner, February
1901).

CLIMATE AND SOIL

195

summarise the evidence and to discuss it in the light of modern
pathology.

In order of time, the first observations on the subject were
made by Dr. H. I. Bowditch (1862). He laid down the " Law of
Soil Moisture " in the following two propositions : —

(1) A residence on or near a damp soil, whether that dampness be
inherent in the soil itself, or caused by percolation from adjacent ponds,
rivers, meadows, marshes, or springy soils, is one of the primal causes of
consumption in Massachusetts — probably in New England, and possibly
in other portions of the globe.

(2) Consumption can be checked in its career, and possibly — nay,
probably — prevented in some instances by attention to this law.

Dr. Gavin Milroy in the Seventh Annual Report of the Registrar-
General for Scotland (pp. xlvii-xlviii) quoted Dr. Bowditch's
conclusions drawn " from a very thorough inquiry into one of
the causes of consumption in Massachusetts." He then pro-
ceeded to investigate the law. Such an explanation he found
would agree with the very different proportion of deaths from
consumption occurring in the eight principal towns of Scotland.
Taking a five-yearly average (1857-61) the death-rate from
consumption per 100,000 of population was found to be 206 in
Leith, 298 in Edinburgh, 310 in Perth, 332 in Aberdeen, 340 in
Dundee, 383 in Paisley, 399 in Glasgow, and 400 in Greenock.
Attention was then drawn to the fact that if each town had been
arranged in the order of comparative dryness of its site,

they would almost have arranged themselves in the above position —
Leith and Edinburgh the most free from consumption, and also having
the driest sites ; Glasgow and Greenock the most ravaged by that disease,
and beyond all comparison situated on the dampest sites.

Dr. G. Buchanan's investigation of the same subject was
embodied in two reports, which were written before he had seen
the remarks summarised above from the Seventh Report of the
Registrar-General for Scotland, or Dr. Bowditch's essay on the
subject. He adds : —

I should not insist on this point, except for the purpose of giving to
the conclusions which Dr. Bowditch and myself have obtained, the
additional weight that they deserve from having been arrived at by a
second inquirer, wholly ignorant of and therefore unbiassed by the work
of the first. ,

Dr. Buchanan's first report is contained in the Ninth Report

196 THE PREVENTION OF TUBERCULOSIS

of the Medical Officer of the Privy Council (1866). This report
summarises the improvements carried out in 25 towns visited
in the course of 1865-66, in which the authorities had carried
out works designed for the improvement of the public health.
The towns were selected as being places where structural sanitary
works had been most thoroughly done, and were not chosen for
any previously ascertained improvement in their health. The
general result of the inquiry, so far as phthisis is concerned, was
that when the sanitary improvements carried out had been
associated with drying of the subsoil, the phthisis mortality had
declined, sometimes to one-third or even one-half of its previous
amount. Great difficulty was experienced in ascertaining the
degree of drying of the soil, as sewerage works were not executed
with this direct object in view. It became necessary therefore
to indicate the degree of drying " in as accurate general terms as
may be." In the table on the following page I have set forth the
main results of Dr. Buchanan's research, arranging the towns
according to the stated influence of sewerage works on the subsoil.
Thus in the 6 towns in which " much drying " of the subsoil
followed the carrying out of works of sewerage, the mortality
from phthisis declined to degrees varying from 49 to 17 per cent. ;
in 4 of the 5 towns in which " some drying " occurred a decline
of from 43 to 19 per cent, occurred, but at Ashby-de-la-Zouch an
increase of 19 per cent, occurred ; in 5 of the 7 towns in which
" minor degrees of drying " occurred, the reduction was from
i to 32 per cent., in Chelmsford the death-rate from phthisis
remained stationary, and in Carlisle it increased by 10 per cent. ;
while in 3 of the 5 towns in which " no change in the subsoil "
occurred, it was reduced from 5 to 8 per cent., and increased
at Brynmawr to the extent of 6 per cent., and at Alnwick to the
extent of 20 per cent. Dr. Buchanan notes in his report that at
Leicester a greater reduction of mortality from phthisis occurred
during the carrying out of the sewerage works than was subse-
quently maintained ; and that at Stratford " a large reduction
of phthisis was for the time observable," although the subse-
quent decline was only i per cent. It is noted also that towns
which like Salisbury made special arrangements for drying
their subsoil improved conspicuously, as did also those towns
with large sewers and those with deep storm culverts. Failure
to reduce phthisis is also stated to be most observable where,

CLIMATE AND SOIL

197

as at Penzance and Brynmawr, the soil already contained little
water, or where the storm water was not properly treated, or
where the deep drainage consisted of impervious pipes laid down
in compact channels, as at Penrith and Alnwick.

Four exceptional cases are pointed out by Dr. Buchanan :
Chelmsford and Carlisle, which had more lowering of subsoil

TABLE XXXIX

Much Drying of Subsoil.

Some Drying of Subsoil.

Phthisis

Degree of

Phthisis

Degree of

Death-rate
(all Ages)

Change in
Phthisis

Death-rate
(all Ages)

Change in
Phthisis.

per 10,000

Death-rate.

per 10,000

Death-rate.

Living.

Salisbury
Ely .
Banbury

26|

-49 per cent.
-47 „

Rugby .
Worthing
Cheltenham

284

- 43 per cent.

Macclesfield

— 3^ »

Bristol .

-22

Croydon

(59* l

(-I7)1,,

Warwick

-19

Cardiff .

34l

-17 „

Ashby .

254

+ 19

Various Minor
Degrees of Dry-
ing of Subsoil.

No Change in
Subsoil.

"w" .

*Eo

|j>

|<|

3 o

Jto *

9 •ei H

S||

-i|

'§£ **

If!

Leicester

434

-32p.c.

Doubtful amount

Penzance

-5p.c.

Newport

-32

of drying.
Local drying.

Brynmawr
Morpeth

28£

3oi

+ 6 „

-8 „

Dover .

264=

-20

Do.

Penrith

~ 5 >»

Merthyr

- II

Some recent

Alnwick

+ 20 ,,

drying.

Stratford

26!

- I

Some local

drying.

Chelmsford .
Carlisle.

324
32

nil
+ 10 „

Slight drying.
Drying with

local defects.

1 Phthisis and lung diseases together.

i98

THE PREVENTION OF TUBERCULOSIS

water than some towns which stood well as regards reduction of
phthisis ; and Worthing and Rugby, which, on the other hand,
experienced a greater reduction of phthisis than other towns in
which there occurred a more complete drying of the subsoil.
The following remark by Dr. Buchanan on this point deserves
quotation : —

Perhaps it had better be confessed that there are exceptions to the
rule of subsidence of phthisis after drying of subsoil ; or the suggestion
may be allowed that the nature of the change in climatic conditions,
produced by drying the subsoil of a locality, is not everywhere the same
(the environs of Chelmsford, for example, still get flooded through the
action of a mill-dam), and that different degrees of effect may hence be
produced on consumption.

Before discussing the facts above summarised, it is desirable
to summarise Dr. Buchanan's second report made in the follow-
ing year, in which he proceeded to examine the apparent relation
between wetness of soil and prevalence of consumption, " with
direct reference to geological considerations." The necessity
of taking into account surface peculiarities quite as much as
the great divisions of the geologist is pointed out. The statistics
of 58 registration districts in the counties of Surrey, Sussex, and
Kent, embracing a population of 1,118,372, living on 3812 square
miles, were taken, the registered phthisis mortality at ages 15-55
being calculated for each district. On tabulating these it soon
appeared that " the districts arranged in the order of the pre-
valence of consumption in them are also to a very large extent
arranged in the order of the dryness or wetness of their soils."
Although this was so, the difficulties in classifying districts
properly were very great, owing to the .t that one section of
the population of a district might be living on pervious and

TABLE XL

Groups of Districts.

Percentage Proportion of Population.

On Pervious Soils.

On Retentive Soils.

A. With least phthisis
B. With next least phthisis
C. Middle as to phthisis .
D. With still more phthisis
E. With most phthisis .

90-9
877
79*5
79'2
64-2

9-1
12-3

20'5

20-8
35'8

CLIMATE AND SOIL

199

another on impervious strata. In such a district the number
living on each kind of soil was estimated, and from the results
thus obtained and the mortality statistics the groups on the
previous page were derived.

The preceding classification is, as explained by Dr. Buchanan,
open to objection, because, for instance, in group D. low plains
of gravel - covered chalk are reckoned under pervious soils,
" which might, so far as their water-holding faculty goes, as fitly
find a place among the retentive formations."

The alternative plan of classifying districts according to their
geological conditions brought out more certain conclusions :
(a) On examining the prevalence of phthisis upon pervious soils
from which water can drain away, as compared with its pre-
valence upon retentive soils, it was found that " the descending
series of the percentage numbers on sands and the ascending
series of those on clays was wonderfully nearly regular for the
districts arranged in the order of their consumption ; so much
is this the case, indeed, that they could not be expected to be
more regular unless one should go the length of contending that
phthisis was a disease influenced by no other circumstance than
the one condition of soil."

(b) Within the limits of " pervious soils " may be included
great ranges of wet and dry soils, according to the elevation of
the ground and the dip of subjacent impervious beds. Thus,
Chichester, situated on low-lying gravel over London clay, had
a very unfavourable position for pulmonary tuberculosis, while
districts on the same gravel, with a sloping clay under it, as at
Croydon, Epsom, Richmond, occupied a more favourable position.
In chalk areas again, for similar reasons, there was least phthisis
on the more elevated portions. On the other hand, low-lying
districts on gravel and chalk near the sea, e.g. Dover, had a
favourable phthisis mortality.

(c) When comparing impervious districts differences were seen.
London clay had commonly a much less degree of wetness than
the Weald clay, and there appeared to be a corresponding differ-
ence in the phthisis mortality. The general results from this in-
quiry are so important that they deserve complete reproduction: —

(i) Within the counties of Surrey, Kent, and Sussex, there is, broadly
speaking, less phthisis among populations living on pervious than among
populations living on impervious soils.

200 THE PREVENTION OF TUBERCULOSIS

(2) Within the same counties, there is less phthisis among populations
living on high-lying pervious soils than among populations living on low-
lying pervious soils.

(3) Within the same counties, there is less phthisis among populations
living on sloping impervious soils than among populations living on
flat impervious soils.

(4) The connection between soil and phthisis has been established in
this inquiry —

(a) by the existence of general agreement in phthisis mortality between
districts that have common geological and topographical features, of a
nature to affect the water-holding quality of the soil ;

(6) by the existence of general disagreement between districts that
are differently circumstanced in regard of such features ; and

(c) by the discovery of pretty general concomitancy in the fluctua-
tion of the two conditions, from much phthisis with much wetness of soil
to little phthisis with little wetness of soil.

But the connection between wet soil and phthisis came out last
year in another way, which must here be recalled,

(cT) by the observation that phthisis had been greatly reduced in
towns where the water of the soil had been artificially removed, and that
it had not been reduced in other towns where the soil had not been dried.

(5) The whole of the foregoing conclusions combine into one-— which
may now be affirmed generally, and not only of particular districts — that

WETNESS OF SOIL IS A CAUSE OF PHTHISIS TO THE POPULATION LIVING
UPON IT.

(6) No other circumstance can be detected, after careful consideration
of the materials accumulated during this year, that coincides on any large
scale with the greater or less prevalence of phthisis, except the one con-
dition of soil.

(7) In this year's inquiry, and in last year's too, single apparent ex-
ceptions to the general law have been detected. They are probably not
altogether errors of fact or observation, but are indications of some other
law in the background that we are not yet able to announce.

The independent generalisations of Bowditch and Buchanan
have been generally accepted, and have formed the basis of
advice which has determined changes of residence for thousands
of phthisical patients. There have been, however, attempts
made to minimise or rebut their conclusions. Thus it was
pointed out by Pearse that in several rainy districts of Devon-
shire phthisis was but seldom a cause of death ; and that the
mortality from phthisis was less at Wisbeach, in the fen district,
than at Axminster on the red sandstone (Lancet, 1876, December,
p. 833). In Holland, again, there is less phthisis than in
France, and " the more elevated provinces with diluvial soil
suffer more than the deep depressions with an alluvial soil,

CLIMATE AND SOIL

2OI

such as Zealand, which has the smallest phthisical death-rate "
(Hirsch, p. 203).

In this country, the late Dr. C. Kelly, Medical Officer of
Health of the combined district of West Sussex, a portion of the
special area investigated by Buchanan, published statistics
which are not confirmatory of Buchanan's results. In his report
for 1879 he showed that the phthisis death-rate had been dis-
tinctly lowered in that district in recent years, " while very little,
if any, change has taken place during the same period in the
drainage of the soil." Sir R. T. Thorne (1888, p. 51) commenting
on this statement, said that the large amount of agricultural
drainage which had then already been effected throughout
the kingdom would be expected to have produced a result in rural
districts very similar to that brought about -by sanitary drain-
age in towns. On this point further evidence appears desirable.
Dr. Kelly gave the following statistics for West Sussex. This
is a district which covers an area of 335,492 square acres, or
about 524 square miles, with a population in 1887 of 105,520.
The different soils found in this district are (i) pervious soils,
which include the upper and lower greensands, the chalk and
the lower Tunbridge Wells sands ; (2) the retentive soils, which
include the Weald clay, the clayey beds of the lower greensand
and the gault ; and (3) moderately pervious soils, sloping from
the sea to the South Downs, where the chalk is covered for a
depth of 15 to 50 feet with loam and brick-earth.

TABLE XLI

Death-rate per 1,000,000 Living at all

Ages from

Nature of Soil.

Population.

Phthisis.

Lung
Diseases.

All Causes.

Pervious ....
Moderately pervious .

33.820
29,640

i5'4
1467

2131
1892

14,852
14.463

Retentive ....

23.530

1542

2583

14,942

It will be observed that the amount of phthisis is not appre-
ciably greater among populations living on a retentive than
among populations living on pervious soils, although other
respiratory diseases are in excess on the former soil.

202 THE PREVENTION OF TUBERCULOSIS

In view of the discrepant results indicated in the preceding
statistics we may ask whether there is an essential relationship
between wetness of soil and phthisis mortality among the popu-
lation living on such a soil, or whether the commonly experienced
excess of phthisis on wet soils is not due rather to the fact that
those who are found dwelling on a wet soil are likely to be of
a lower class of the community, worse housed, and more exposed
to the infection of phthisis. Buchanan himself agrees that
there are exceptions to the law, and suggests that " they in-
dicate the presence of other influences in the subsoil, which
have hitherto escaped detection." Hirsch suggests, as a more
probable explanation, that other etiological factors besides
the influence of soil come into force under the given circum-
stances, and serve to neutralise the benefits even of the most
favourable conditions of soil ; and with this suggestion I
agree. It appears probable that much of the benefit ascribed
to drying the soil has been due really to other factors of
improvement which commenced to operate about the same
time as the former.

It is difficult to fit in our present knowledge, that the essential
cause of tuberculosis is the tubercle bacillus, with the wet soil
theory. It cannot be maintained that such a soil favours the
growth of the tubercle bacillus, an organism the extra-corporeal
cultivation of which is beset with difficulties. We can only
conclude that the wet soil operates merely as a predisposing
cause. It implies greater loss of heat by evaporation, more
easy provocation of catarrhs, especially when, as would com-
monly happen, it is associated with cold and wet houses. Against
these factors a house even on a wet soil can in a large measure
be protected.

The wet soil must be placed, like overcrowding and insuffi-
cient nutrition, among predisposing causes, infection being the
chief and essential cause. It must be placed furthermore in
a lower place than either overcrowding or underfeeding.

Consumption is essentially a disease of crowded populations,
of indoor occupations, transmitted by infection, favoured by the
rebreathing of respired air, and by organic filth of all kinds.
Crowding, especially crowding of the sick, has greatly declined,
and was already in the process of declining, while the sewerage
works referred to in Table XXXIX. were being effected.

PART II

THE INCIDENCE OF TUBERCULOSIS UPON
COMMUNITIES

203

CHAPTER XXVIII
INTRODUCTORY

ACTUAL experience on a large scale is the final test of
hypothesis and the surest basis for action. This maxim
is particularly applicable to public health administration.
The study of communal experience is therefore of the utmost
help to the public health service ; but for trustworthy results
this study must be conducted with a clear recognition of the
complexity of the material to be examined. With no statistics
of disease is this caution more necessary than with those relating
to tuberculosis.

In the foregoing chapters tuberculosis has been seen to be
an infectious disease having a variable period of incubation,
and a course which may extend intermittently or continuously
over many years. Its prevalence and the death-rate due to
it may be favoured or hindered by a great variety of personal,
economic, and sanitary conditions affecting the populations at
risk. Many of these conditions are themselves composite and
of great complexity ; and during a considerable part of their
infective sickness most patients are able wholly or partially
to keep at work and to migrate from one district to another.
Without detailing the difficulties which these characters of
tuberculosis introduce into statistics measuring the prevalence
of the disease in different communities, or the errors which
may arise from applying to such statistics the methods appro-
priate to acute disabling infectious diseases, it suffices for the
present purpose to recognise that the causation of tuberculosis
in communities has all the complexity of its causation in the
individual, with the added complexity due to variations in
economic and sanitary environment and to the migration of
infected persons.

To obtain the best practical results we must simplify this
complexity. As already seen, a considerable number of in-

205

206 THE PREVENTION OF TUBERCULOSIS

fluences either promote or hinder the spread of tuberculosis ;
but the preceding chapters could afford little information as to
their relative importance. Were it possible to adopt all known
measures of precaution and all the methods of treatment, this
absence of quantitative information would have merely academic
interest. Practical administration, however, can afford no such
wholesale reproduction of laboratory conditions. The amount
of money and energy available for the public health service,
though it may fluctuate from generation to generation, is always
limited ; and of the measures that would aid in the prevention
or cure of disease only a portion can be put into simultaneous
operation. Thus any such measure yielding less than the
utmost value for the resources expended represents an amount
of avoidable and permitted disease proportionate to the relative
inefficiency of the measure. It will be seen, therefore, that
the rational as opposed to the capricious or random selection
of measures is supremely important to the public health service ;
and where it can be had, actual experience is the safest and
final guide. The chief purpose for which the incidence of
tuberculosis upon communities must now be studied is to learn,
if possible, from actual experience the relative extent to which
any or all of the elements of economic and sanitary environment
have promoted or hindered the spread of the disease.

Such study is of course beset with the ordinary dangers
of statistical reasoning, which are much the same as those of
any edged tools in unskilled hands. In order to learn the
causes of variations in the incidence of a disease upon com-
munities, any sets of figures intended to measure this incidence
must in particular be free from the fallacies due to migration
of patients, whereby an infection may be acquired in one district
and be chronicled as disease or death in the statistics of another.
For this reason among others local statistics have to be handled
with caution even when they concern acute infectious diseases
of only a few weeks' duration. Tuberculosis is not only an
infectious but also a chronic disease, which on the average
probably extends over years and often escapes recognition during
a large part of the time. Fallacy is almost inevitable in such
a case if inferences as to causation are sought from individual
groups of local statistics.

If, for example, sanatoria for consumption were established

INTRODUCTORY 207

in certain towns or counties of a country otherwise poorly
provided with them, merely elementary statistical reasoning
would prevent a comparison between the death-rates of such
towns or counties, which would attract consumptives beyond
from outside their bounds, and those of towns or counties without
sanatoria, with any idea that the comparison could give informa-
tion as to the effect of sanatorium provision upon the general
prevalence of phthisis. Similarly the figures of a small rural
county with a population less than that of many single towns,
could only be used for inference as to the causes of variations
in its tuberculosis death-rates if correction were made for the
migration of healthy persons to towns and of sick persons to
their country homes, where they can live at a smaller cost and
nearer their own people.1

Nor is it merely its long activity nor its still longer latency
which demands a wide basis of observation before conclusions
can be drawn as to the causation of tuberculosis. Its endemic
prevalence is affected, as we have seen, by factors of sanitary,

1 The difficulty of forming non-fallacious conclusions from "parochial"
statistics concerning an infective disease of protracted latency and protracted
duration may be further illustrated by the phthisis death-rates in tenement
houses and in the different districts of a large town. It is well known that the
phthisis death-rate is higher in populations inhabiting one room than in those
inhabiting dwellings with two or more rooms ; and is greatest in overcrowded
dwellings of any given size. The association between the phthisis death-rate and
size of dwelling and overcrowding is complex, and before drawing inferences as to
the effect on phthisis of the increased infection and lowered resistance accompany-
ing overcrowding, we should ascertain among other things to what extent the
inhabitants of these overcrowded tenements drifted into them after and perhaps
because they had become consumptive. Similarly, in comparing different
districts of a large town or even small towns with each other, allowance has
to be made for the influx of consumptives into poorer districts as they go down
in the social scale. If this can be done, — and it implies a complete knowledge
of each patient's history and of the duration of the latent period of his disease, —
it has further to be noted that inasmuch as the opportunities for infection by
phthisis vary enormously in different districts, the effect of measures against
infection must correspondingly vary. We must therefore either compare
the influence of such measures on large masses of population in whom this source
of error is likely to be equalised, or on small aggregations having a like com-
position. It is evident, for instance, that efforts against infection may have had
a greater effect on the death-rate from phthisis in a district whose death-rate
from this disease is still 2 per 1000 than similar efforts in another district of a
different social stratum whose death-rate from phthisis is only i per 1000. For
the above and other reasons, local statistics of phthisis cannot be .used for
comparison with those of other districts without fallacy, unless corrections are
made which only the most intimate investigation will render practicable.

208 THE PREVENTION OF TUBERCULOSIS

including social and economic, environment, which themselves
are of high complexity and largely interdependent. Such
phenomena may be unrecognisable in experience on a small
scale.

To eliminate or minimise the effects of migration and com-
plexity we must study communities in which the balance between
immigrant and emigrant cases is small relatively to the total
volume of disease, and which are so large as to allow the operation
of complex phenomena to become evident. The use of figures
relating to large communities is further commended for the
study of tuberculosis because their size reduces the chance
of the results being determined by some local or accidental
feature among the complex relevant conditions of environment.
The experience of smaller communities can only be taken either
as hints which may possibly be confirmed by other information,
or as illustrations of the manner of action of influences of which
the existence has been demonstrated independently.

In the investigation which is summarised in the following
pages it has been found possible to obtain significant results
as to the causes of the variation" in death-rates from tuber-
culosis by grouping these rates for given communities and
periods with the figures which represent for the same com-
munities and periods the variations of sanitary and economic
environment, thus disclosing what the figures can tell of the
relationship between the two sets of phenomena. The following
chapters include the results of the comparison of such of these
data as are available. It will be found that improvement
in general communal health and in the individual factors
affecting it has not always corresponded with the reduction of
tuberculosis, although the statistical evidence shows a probable
connection between most of these factors and the disease. If
no constant correspondence had appeared between the course of
tuberculosis and any element of environment, no conclusion
could have been obtained from the statistical study of communal
experience, and we should have been left to draw the most
probable inferences we could from the facts stated in Part I.
Such a result would not have been surprising. Communal
experience has to be studied not in the orderly sequence of
individual influences provided in laboratory experiment, but in
the simultaneous and highly complex combinations of influences

INTRODUCTORY 209

found in communal life. In these combinations nothing is
more common than to find that the number of unknown
quantities is too great and the facts too few to permit of an
approximate estimate of the respective values of the unknowns.
It will be found, however, that the course of tuberculosis has
followed that of one element of sanitary environment, namely,
the institutional segregation of tuberculous patients. From an
administrative standpoint, this result has considerable con-
sequences. It is desirable therefore to examine in detail the
evidence as to each of the elements of sanitary environment
concerned.

In most cases the figures relating to phthisis have been
taken as representing tuberculosis, as they are recorded more
fully, and are based on diagnosis which is more accurate than
that of total tuberculosis. In almost all cases the incidence
of the disease has had to be measured by its death-rate.

CHAPTER XXIX

TUBERCULOSIS AND GENERAL HEALTH IN VARIOUS
COMMUNITIES: VIRULENCE, NATURAL SELECTION,
AND DECADENCE

f I ^HE first teaching of communal experience on this subject,
the evidence for which will be outlined in the present
chapter, is that the control of tuberculosis is not merely a
question of the improvement of general health and of sanitary con-
ditions. No result could be more important or more encouraging
for practical purposes. Those concerned in the service of public
health know how much remains to be done before it can be said
to have done its best. If general sanitary conditions are under-
stood— as they are in this connection — to include all those con-
ditions which affect general health, the task that remains to be
done is indefinitely great. The improvement of conditions of
housing, abolition of overcrowding, the enforcement of a higher
standard of specific and general cleanliness, the removal of
injurious conditions of work, whether in mine, factory,
workshop, shop, office or home, the promotion of reason-
able recreation in our towns, the removal of hindrances to
temperance and thrift, all of which come within the range of
the task, illustrate the vastness of the physical, economical,
and even moral problems involved, and of their importance
to national life, happiness, and efficiency. The cultivation of a
popular sanitary conscience is therefore an object of supreme
importance to the well-being of any community, and the con-
nection between tuberculosis and bad general sanitary conditions
can be utilised to the full extent in stimulating this conscience.

But though this connection is far-reaching and intimate, it
must not be allowed to obscure other influences which have
had more direct effect on tuberculosis. There are few sanitary
improvements that do not in some measure tend to hinder the
spread of tuberculosis. This fact is evidenced so strongly and

TUBERCULOSIS AND GENERAL HEALTH 211

in so many ways, that the doctrine that the control of tuber-
culosis must be sought not by measures specially directed against
the disease, but by improvement in general sanitary environment,
has been adopted by many as the final formula on which the
control of tuberculosis must be based. The correctness of such
a doctrine does not follow necessarily from the many facts illus-
trating the connection between tuberculosis and sanitary environ-
ment ; and an examination of the actual experience of large
communities shows that it is contradicted by the facts. To
those who hope for the extirpation of the disease, this result is a
matter of congratulation. The demonstration of the formula
which says that tuberculosis is to be conquered mainly through
improvement in general sanitary conditions, and not through
special measures acting in conjunction with them, would have
been full of profound discouragement and the sickness of hope
deferred. If the control of tuberculosis must await the general
perfection of sanitary conditions, including the economic and
moral circumstances which form an essential part of them, no
reasonable limit could be put to the time which must elapse
before tuberculosis disappears.

The belief that no practicable special measures exist by which
the disease can be controlled more rapidly and directly than
by measures of general sanitary reform, is not supported by past
experience in regard to other infectious diseases which have been
extirpated wholly or in part. Cholera, typhus and enteric fever
in England, and small-pox in Germany have been stamped out
or greatly diminished by adding to the necessarily partial
measures of general sanitary reform a complete application of such
special measures as actual experience has shown to be efficient.
Tuberculosis can be extirpated similarly, if similarly the slow
effect of only gradually improving sanitary circumstances be
supplemented by special measures having a more rapid and
specific effect on the disease. If such measures are contained
in the general body of sanitary improvement, they require to
be dissected out and identified before they can be applied with
rapidity and completeness.

There has been no constant relation between improved
general sanitary circumstances and reduction in tuberculosis.
The most definite expression of the course of general sanitary
(including social) improvement in the gross and of tuberculosis

212

THE PREVENTION OF TUBERCULOSIS

is to be found in the course of the death-rate from all causes
other than tuberculosis and the death-rate from tuberculosis.
For the reasons explained previously, the death-rate from tuber-
culosis will be taken to be measured by that of phthisis.

In Table XLII. the death-rates from pulmonary tuberculosis
and from all other causes in various countries and capital cities
are given for 1881-85 and for 1901-03 or 1901-02. These
relatively recent periods are taken for comparison, because in
some instances earlier figures are unobtainable.

TABLE XLII

Death-rate
from all Causes
except Phthisis.

Death-rate from
Phthisis.

Percentage
Change in

1881-85.

1901-03.

1881-85.

1901-03.

England and Wales

17-97

14-94

1-83

1-23

-17-0

-327

Scotland .

17*45

1576

2-II

1-47

- 9 '9

-30-3

Ireland .

15-90

J5*45

2-08

2-15

- 2-8

+ 3'4

Norway .

1575

12-58

1-39

1-92

-20-4

+ 38-1

Prussia .

22-29

17-90

3-n1

I '931

-197

-37*9

Massachusetts

16-68

1477

3-14

1-67

-"'5

-46-8

Paris

19-99

H-iS

4-41

3-65

-29-3

-I7'2(?)

Berlin .

21-38

1376

3*32

2-04

-337

-38-5

Copenhagen

19-38

14-81

2-892

1-38

-237

-52-2

London .

1878

I5-38

2-20

I>65

-19-2

-25-0

Manchester

1876

16-34

2-42

2-01

-13-1

- 16-9

Edinburgh

16-34

I4-74

I-89

*•'$«

- 9'5

-20'I

Glasgow .
Dublin .

22-34
24-25

17-83 3
23-02

3T4
3*55

i-683
3-28

-20-0
~ 5'2

-46-5
- 7'6

Belfast .

20-32

18-32

378

3-o8

- 10-2

-18-5

1 Tuberculosis.

2 1880-84.

3 1901-04.

It will be noted that in all cases the general death-rate apart
from phthisis has declined ; as has also the phthisis death-rate
in all except Ireland and Norway.

The increase in Ireland is really greater than it seems. Emi-
gration, as will be seen later (p. 217), has altered the age and
sex distribution of the population by removing a large part of
the young and middle-aged, among whom most deaths from
phthisis occur ; and when the figures are corrected for age and
sex distribution, the true increase of phthisis on the assumption
of constant age and sex distribution is seen to be really larger

TUBERCULOSIS AND GENERAL HEALTH 213

than the figures show. Thus when the crude phthisis death-
rate in Ireland for 1891, which was 19-3, is corrected for age
distribution of population so as to make it comparable with that
for 1901 (21*5), it becomes 177 per 10,000 ; and the crude increase
of 12 per cent, becomes a corrected increase of about 22 per cent.
A very high decrease of general death-rate apart from phthisis
is shown by Norway, which shows increase of its phthisis rate.1
It will be noted also that in every country and city in which
a decrease of phthisis has been shown this decrease is greater
than that of the death-rate from all other causes. This
disparity is of very variable extent, but except in Dublin and
Manchester the disparity between the two diseases is always
great. Table XLIIL, calculated from Dr. Tatham's data for
England and Wales, makes a similar comparison analysed in
detail into sexes and ages.

TABLE XLIIL — ENGLAND AND WALES

Percentage Decline or Increase of Death-rate when the experience oj
1861-70 is compared with that of 1896-1900

Males.

Females.

At Ages

General Death-

rate minus

Phthisis.

rate minus

Phthisis.

-14

-60

-16

-65

-49

-67

-45

- 5^

10-

-46

-68

-42

-61

15-

-32

-59

-40

-63

20-

-34

-52

-36

-62

25-

-28

-43

-29

-58

35-

-15

-25

-14

-46

45-

- i

-17

-44

55-

+ 6

-19

+ 2

-40

65-75

+ 3

-24

-36

All Ages ....

-13

-38

-14

-54

xThe official figures relating to Norway, by reason of the increased complete-
ness of certification, show a higher increase than is likely to have occurred in
fact ; but no reasonable correction in this respect would show decline of
phthisis during the period in question ; and the argument developed in the
text — which would remain the same if even a stationary death-rate from
phthisis were substituted for the increase shown by the official figures — is
unaffected. So far as England and Ireland are concerned the figures may be
accepted within narrow limits of error.

214 THE PREVENTION OF TUBERCULOSIS

It is clear from this table that in England, as in the instances
in Table XLII. to which reference has been made, the reduction
of the phthisis death-rate is enormously greater than that of the
general death-rate from all other causes ; and the discrepancy
is especially great at the working years of life in which phthisis
causes its heaviest death-rate. If phthisis had shared only to
an equal extent in the general reduction of mortality, a pre-
sumption would have arisen that the improvements in general
sanitary conditions which have been operating to reduce the
general death-rate, such as higher wages, cheaper food and
clothing, improved sanitation, and other allied influences, are
in themselves a sufficient explanation of the reduction of phthisis.
The above figures show that, however much these influences
have contributed to the reduction, they do not explain it suffi-
ciently, unless it be assumed that phthisis is far more susceptible
to the operation of these influences than other diseases. For this
view there is no evidence, and I am not aware that it has been
put forward. On the English figures, therefore, the variation
in the phthisis rate must accordingly be taken to have involved
co-variations in some phenomenon or group of phenomena
which have had no material effect on the general death-rate.

The same conclusion results from the figures of other countries.

Where phthisis has been reduced, the reduction has been not at

the rate of the reduction of the general mortality but at a much

faster rate. The extra rapidity of the decline of phthisis is not a

fixed part of the reduction of the general mortality, but a part

which varies widely from country to country ; in two countries

an improvement in general mortality has been accompanied by

an actual increase in mortality from phthisis, and in one of them

both the improvement in general sanitary conditions and the

increase in the death-rate from phthisis have been exceptionally

large. Thus in the experience of a considerable number of

countries, the conditions improving general health have not

had any constant effect on the prevalence of tuberculosis, anc

in Norway, in which an exceptional improvement in genei

health has occurred, it has been accompanied by increc

in mortality from tuberculosis. It follows therefore that,

whatever may have been diminishing tuberculosis, improvement

in general sanitary and social circumstances has not been ttu

principal cause, and that an influence or influences of moi

TUBERCULOSIS AND GENERAL HEALTH 215

powerful and rapid operation must have been at work in the
communities examined.

So far as this comparison carries us, variations in the death-
rate from tuberculosis might be wholly independent of any
sanitary conditions. From what has been seen in Part I., this
alternative is clearly incorrect, seeing that many conditions
affecting general health are known independently to have a
powerful and direct effect on tuberculosis. Simultaneously,
however, with the operation of general sanitary improvement
other influences may have been at work independent of sanitary
conditions or not dependent on them directly ; these influences
may have done more to modify the prevalence of tuberculosis
than any influences of sanitary environment, and it is con-
ceivable that the control of tuberculosis is not to be expected
primarily through measures of further sanitary reform, whether
general or special.

The three influences not necessarily associated with general
sanitary environment which have been suggested as having
possibly operated in different communities to produce the
recorded variations in the death-rate from tuberculosis are : an
attenuation of the virulence of the infecting organism ; a process
of weeding-out of the more susceptible population; and an
exactly contrary process of survival of the unfit and consequent
decadence of the average population.

Variations of virulence in the specific micro-organisms are
known to have occurred with some infectious diseases. They
have been demonstrated by variations in the type as well as
the severity of the clinical symptoms, and hitherto only when
such variations have been demonstrable has a variation in the
virulence of the disease been suggested. There is no evidence
that such a variation has occurred in the case of tuberculosis ;
and the suggestion is made in the teeth of a considerable volume
of evidence to a contrary effect. The clinical types of the
disease, as recorded in the contemporary descriptions of Graves,
Watson, Walshe, Flint, and others at the beginning of the period,
show the same varieties of type and duration as are now seen.
No well-marked distinction has been established between the
types of tuberculosis in different countries. Though consump-
tives probably live longer now than they did formerly, it must
be remembered that the rational treatment of the disease has

216 THE PREVENTION OF TUBERCULOSIS

only become general in recent years. The assumed attenuation
of virulence which is held to be displayed in one country because
its tuberculosis bill has decreased, can scarcely be assumed
to have existed simultaneously in neighbouring and inter-
communicating countries in which the disease has increased,
notwithstanding the fact that the clinical types of the disease,
so far as can be ascertained, have remained unchanged in both
countries during the whole period under examination. All the
evidence available tends therefore to show that outside bacterio-
logical laboratories no change of virulence has occurred in the
bacillus of tuberculosis, and the only evidence from which it
has been sought to infer such a change is the decrease of the
prevalence of tuberculosis in certain countries, the actual
phenomenon to explain which this otherwise unsupported
assumption has been made.

The hypothesis that the reduction of the disease may be
due to elimination of susceptible strains of human beings depends
similarly on the mere fact that it is consistent with the decrease
which has occurred. The evidence of the transmission of sus-
ceptibility has not been sufficient to show that this trans-
mission occurs so frequently as to be a predominant factor
in the transmission of the disease. On the other hand, there
is abundant evidence to show the existence of susceptibility,
not inherited and permanent, but temporary and acquired
through circumstances of environment. It is equally clear that
the liability to infection is affected by extent of dose, and that
a considerable proportion of the population in contact with
tuberculous patients is exposed to extreme and prolonged
infection. Persons placed in these circumstances would acquire
infection with greater certainty than others, and when they
were children of tuberculous parents this occurrence would be
practically indistinguishable from inherited susceptibility, and
has doubtless often been regarded as such. Even if the inherit-
ance of susceptibility had been demonstrated as a common
occurrence, it could only explain the decreases that have
occurred in most countries on the assumption that the sus-
ceptible victims had a special infertility. The mere death of
susceptible patients at the end of a chronic infectious disease
of long duration and extending most often into middle life can
have had little or no effect on the susceptibility of the children

TUBERCULOSIS AND GENERAL HEALTH 217

of these patients, unless these children are on the average
much less numerous than the children of entirely healthy
stocks. Although there appears to be a difference between the
two stocks in this respect, it does not suffice to explain results
already obtained.

In considering the suggestion that decadence has been
responsible for the increase of phthisis, where this has occurred,
we may turn again from the discussion of interesting but quite
unverified hypotheses to the more sober study of actual experience.
The country in regard to which this has been oftenest urged is
Ireland. The undoubted general poverty of the country makes
the suggestion primA facie plausible ; and unhappily plausible
hypotheses whose face is their fortune are often accepted because
no one is concerned to ask for more solid credentials. If the
instructive experience of Ireland in regard to phthisis is to be
explained by an ill-defined influence of which the control is
hard and uncertain, the prospect of mastering the endemic
prevalence of phthisis in Ireland would be postponed to an
extent that would discourage administrative reform directed
against more definite causes. In itself, therefore, the alleged
decadence of the Irish people in Ireland deserves careful con-
sideration ; and the study is not the less desirable because,
as we shall find, the existence of a general average decadence
of population in Ireland is, so far as phthisis is concerned, a
wanton speculation contradicted directly by the facts.

The suggestion is that the long stream of emigration from
Ireland has left behind it a physically inferior population of
excessive susceptibility to phthisis. This emigration reached
its height in 1851, when over 34 per 1000 of the entire popula-
tion left their country ; but it has continued up to the present
time, still averaging 9 per thousand per annum during the
present century. That the effect of this emigration has been
to leave a decadent residual population is merely an assumption,
which at the outset is discredited to some extent by the fact
that the birth-rate in Ireland (corrected for the number of
women at child-bearing ages and for the number of married
women) has increased from 35*2 in 1881 to 36*1 per thousand in
1901, against a decrease in England from 347 to 28-4. It is
discredited further by the fact that the majority of those driven
from Ireland were among the poorest, and these through their

218 THE PREVENTION OF TUBERCULOSIS

poverty must have been the least fit. The cottiers and farm
labourers on the smallest holdings emigrated in the largest
numbers ; those who remain are children of the families who
could resist the stress of famine and evictions, and who in recent
years have been living in progressively better conditions than
their predecessors. Even a comparative examination of the
present population does not show an appreciable difference in
the communal susceptibility to phthisis between rich and poor
towns. Belfast is the part of Ireland which probably has
suffered least from emigration, and is commercially the most
prosperous. Yet its death-rate from phthisis was 307 per
100,000 in the five years 1901-06, as compared with 315 in the
much poorer and more crowded city of Dublin.

These considerations, though much more weighty than the
general speculation by which decadence in the Irish population
is alleged, are still to some extent inferential. Fortunately it
is possible to settle the question definitely by actually following
the emigrated population and comparing their susceptibility
with that of the residual Irish.

The chief emigration from Ireland has been to the United
States. If the cause of the increased death-rate from phthisis
in Ireland is the physical inferiority of its residual population,
the death-rate from phthisis of the Irish population in the
United States ought to be lower than that in Ireland. It is
practically certain that no disturbing influence in such a com-
parison is exercised by greater well-being or better sanitation
or housing in Ireland than in the United States. The American
Census Report for 1900 gives the death-rates from phthisis in the
registration area and its subdivisions among whites in the
census year, classified according to the birthplaces of the mothers
of the deceased. For all inhabitants of these States the phthisis
death-rate in 1900 was 113, for English (defined as above) 135,
for Scotch 173, for Germans 167, for Irish 340. The difference
is seen both in cities and in rural districts, the phthisis death-
rate of the Irish in rural districts being 239, as compared with
a general rate of 108. In Ireland in the same year the phthisis
death-rate was 226 and in Dublin 346. These are death-rates
uncorrected for age distribution. For such correction we turn
to the vital statistics for the city of Providence, Rhode Island,
which are well known to be among the most trustworthy in

TUBERCULOSIS AND GENERAL HEALTH 219

the United States. Dr. Chapin, the city registrar and medical
officer of health, has published statistics corrected for age dis-
tribution which enable a corrected comparison to be made.
He applied the death-rate from phthisis in Ireland in 1901 for
sex and age periods to the population of Providence in 1900
born of Irish mothers. " It was found that the theoretical
mortality from phthisis of this element of the population (of
Providence) according to these (the Irish) data was 258 per
100,000 living. The actual rate for the period 1896-1905 was,
however, 339. The mortality from phthisis of the Irish in
Providence is therefore 81 per 100,000, or 31^4 per cent, more
than the mortality of the Irish in Ireland."

It is clear therefore that, so far from emigration having
increased the communal susceptibility of the residual Irish
population to tuberculosis, the Irish in Ireland have a substan-
tially less susceptibility than their emigrated brethren, and that
this difference is not due to any inferiority in the environment of
the emigrated population. The inability of extreme poverty
to produce a high death-rate from phthisis in a rural popula-
tion is strikingly shown in the County of Mayo (p. 180).

CHAPTER XXX

TUBERCULOSIS IN URBAN AND IN RURAL
COMMUNITIES

IN the present and the succeeding chapters we have to
consider the experiences of large communities over long
periods of time, and to compare the variations in the
figures measuring the incidence of tuberculosis and those, where
they can be obtained, which measure the variations in the element
of experience under consideration.

To avoid misapprehension, a preliminary remark is necessary
as to the years which should be compared. The effect of altera-
tion in environment does not begin to appear till after a certain
interval. If the element in question operates solely by diminish-
ing infection, the interval must be that which represents the
minimum period of incubation and latency. This interval
cannot be stated with any exactness, and it is still less possible
to state the interval which would have to elapse before an altera-
tion which modified resistance of the community to infection
would produce an evident effect. Strictly speaking, the figures
which represent alteration in environment should be compared
with those which represent incidence of tuberculosis at a period
later by this interval. It is fortunate that the run of these figures
in the present inquiry, as might be expected with a disease of
long incubation and latency such as tuberculosis, is such that
changes from one quinquennium to another are not abrupt ;
and in a sufficiently long series of pairs the results of identical
quinquennia can therefore be grouped with substantially the
same result as if the element of environment were represented
by the figures of the next quinquennium or the next but one.

Communities may be grouped most broadly according as they
are urban or rural, and the experience now to be examined shows
the remarkable result that while urban conditions have pro-
moted the prevalence of tuberculosis, they have rarely sufficed

URBAN AND RURAL COMMUNITIES

221

to prevent extraordinary decreases in the disease, nor in all
cases have rural conditions sufficed to prevent increases. Town
life on the whole is less healthy than rural life. Some evidence
of the unquestionable correctness of this belief may be gathered
from an inspection of Table XLIL, p. 212 ; and this difference to
the disadvantage of the towns is seen in tuberculosis as well as
in other diseases. This result may be checked with the help of
two valuable tables by Dr. Tatham, published in the Registrar-
General's Report for 1904, from which the following table is
extracted and calculated. This table deals with an estimated
urban population of 18,262,173, including the chief industrial
centres, and a rural population of 4,327,835, including only a
few unimportant towns and villages. The death-rates have
been corrected for variations in the age and sex distribution of
the respective populations.

TABLE XLIV.— ENGLAND AND WALES
Selected Urban and Rural Counties of the Registrar-General^ 1898-1903

Corrected Death-rates per 1000 of Population.

Males.

Females.

All Causes
except Phthisis.

Phthisis.

All Causes
except Phthisis.

Phthisis.

Urban Counties
Rural ,,

18-4
I3'5

1-66
1-27

IT'S
13*2

I'll

1-07

Urban Counties
Rural

Pro
137

100

portional Figure.

131
100

; (Rural rates = I
133

100

00)

104

IOO

These collective results show no less strongly than those
of individual countries and towns that town life is unhealthy as a
whole, and is favourable to the prevalence of phthisis. If they
could be corrected for the fact that the towns attract the robust
and strong, while the weakly tend to remain in and return to
rural districts, the extent of this mischief would be exhibited
more strikingly and even more accurately. In the absence of
powerful countervailing influences, those countries would there-
fore be expected to have suffered most from phthisis and to have

222

THE PREVENTION OF TUBERCULOSIS

shown most marked increase in the disease in which the excess
of urban over rural population has been the largest and the most
progressive.

An examination of the facts shows, however, that the exact
contrary has occurred.

Table XLV. exhibits for certain countries the distribution of the
population between town and country at or near the beginning
and end of the period under review. The definition of " urban "
varies somewhat in different countries, but in each country
remains the same throughout the period under examination, so
that the results are comparable. The corresponding phthisis
rates are included in the table, and the changes in the death-
rates are expressed as percentages of the earlier figures.

TABLE XLV

<+-! g C

C 52

oS j.;

c w §

.23

c\ _

Percentage

§g|:

o-S

3 X

"II

Oj rcJ

of Total
Population
who were
Urban in

Phthisis
Death-rate.

§J

lilt

wi2o
5 s

"§w
Q o

o fi

rt "^ <U C

C i-

c§'~

§ ^2

S o

u^C

'S rt

'•§ -

1866-

1901-

1861.

1901.

1870.

1903.

England and Wales .
Scotland .

77
70

2'45
2'59

1-23

1-48

100

-50

-45

100
12O

Ireland

1-82

2-15

+ 18

175

1864.

1895.

Prussia

3-201

I-941

-391

1865.

1891.

France

4-57 2

3-65^

Pnone

297

1840.

1890.

Mass.

United States .

3*65

1-67

262

-5°

136

1865.

1891.

Norway .

I '32s

1-92

+ 46

156

1 Between 1877-80 and 1901-03. * Paris. 3 In 1876-80.

More recent data as to urbanisation are contained in Table
XLVI. from Dr. Shadwell's work (1905, vol. ii.).

URBAN AND RURAL COMMUNITIES

223

TABLE XLVI

Percentage of the Population of Great Towns having over 100,000
Inhabitants to the Entire Population of each Country

England.

Germany.

United States.

1881.

1901.

1880.

1900.

1880.

1900.

31-6

35'o

7-2

16-2

14-6

18-8

In context with these results reference may be made again to
Table XLIV. It will be noticed that the excess of the general
death-rate in urban counties over that in rural counties is
approximately equal for males and females (37 arid 33 per cent.),
while the excess of phthisis in urban counties is 37 per cent,
among males and only 4 per cent, among females. In Birmingham
and Sheffield the female death-rate from phthisis, as shown in
Fig. 13, p. 1 66, is actually lower at most ages than that in England
and Wales as a whole. When it is remembered that women
spend much more time at home than men, and that their experi-
ence must reflect more than that of men the influence of home
environment, it becomes clear that the influence of urban life
on phthisis is specifically different from its effect on other causes
of mortality in the aggregate.

The experience summarised thus shows that enormous changes
have occurred both in the extent of urbanisation and in the pre-
valence of phthisis in each of the countries examined, and that
in every country town life has been associated with a greater
prevalence of tuberculosis than has country life. There has
been everywhere a heavy increase of urbanisation, which in
spite of the larger amount of phthisis in towns has been accom-
panied in most countries by a large reduction in the prevalence
of phthisis both in town and in country ; indeed, the countries
with the most town life have suffered actually the least from
phthisis. It follows therefore that, powerful as has been the
influence of town life in assisting the prevalence of tuberculosis,
some other more powerful influences have been in operation in
most countries to restrain the disease.

CHAPTER XXXI

TUBERCULOSIS IN OVERCROWDED COMMUNITIES

THE next fact to be extracted from communal experience
is that even overcrowding has been unable to exert a
predominating influence on the course of tuberculosis.
Overcrowding is the most mischievous factor of town life. Its
operation even in country districts must be detrimental ; and
in towns the privation of light and air which it usually entails
must add greatly to its depressing effect. So much is certain
from general considerations, and it is equally certain that tuber-
culosis as well as other diseases must be susceptible to the influ-
ence of overcrowding. In the last chapter we found as a fact in
international experience that town life, though tending power-
fully to increase the prevalence of tuberculosis, has not sufficed
to cause an increase in the face of other countervailing circum-
stances to be considered subsequently. It is unnecessary or
impracticable to examine separately certain of the factors of
town life. We have seen in Part I. that subsoil drainage is
not likely to have been a factor of primary importance for this
purpose. The substitution in town life of industrial for agri-
cultural conditions is so essential a part of urbanisation that a
separate investigation of its changes could give no different
results from those obtained in the last chapter. The ameliora-
tion of industrial conditions in regard to dust, ventilation,
etc., is not expressed directly in any recorded figures; to some
extent an indirect expression may be found in the evidence which
will be considered as to sanitary education. Neither can a
direct expression be obtained for the changes in provision of
light and air ; but indirectly they are covered by the changes in
overcrowding, which fortunately are recorded sufficiently for
the present purpose. It is in overcrowding that the most
vicious results of town life must be sought ; and they deserve
very careful consideration.

OVERCROWDED COMMUNITIES

225

The difference in total housing accommodation between
urban and rural communities in England and Wales may be
seen broadly in Table XLVII.

TABLE XLVII

1901. — Of the Total Population in Urban and Rural Districts respectively,
the Percentage living in each Class of House was as follows : —

Tenements containing

One
Room.

Two
Rooms.

Three
Rooms.

Four
Rooms.

Five or
more
Rooms.

Total.

Urban Districts .

2'0

7-4

10-3

21 '2

59'i

lOO'O

Rural , ,

0'2

3 '9

8-1

24-0

63-8

lOO'O

Thus, compared with rural districts, ten times as large a pro-
portion of the total population lived in one-roomed tenements
in urban districts, and nearly twice as large a proportion lived
in two-roomed tenements.

The difference in overcrowding between urban and rural
communities in England and Wales is shown in Table XLVIII.
A tenement is reckoned as overcrowded in which on an average
each room, whether bedroom or living room, is occupied by
more than two persons.

TABLE XLVIII

1901. — Of the Total Population in Urban and Rural Districts respectively,
the Percentage Overcrowded in Tenements of four Rooms and under was
as follows : —

Tenements containing

One
Room.

Two
Rooms.

Three
Rooms.

Four
Rooms.

Urban Districts ....
Rural „ ....

o-95
0*09

3'07
i'54

2-63
1-98

2-25
2-23

This table shows that ten times as many one-roomed
15

226

THE PREVENTION OF TUBERCULOSIS

tenements, and twice as many two-roomed tenements were
overcrowded in urban as in rural districts.

Nothing could be more conclusive than these results as to
the difference both in housing and in overcrowding between
urban and rural districts. Compared with rural districts,
towns in 1901 had ten times as large a proportion of the total
population housed in one-roomed tenements ; and of the popu-
lation so housed in one-roomed tenements, ten times as many
were overcrowded in towns as in country. Nearly double
the proportion of town population inhabited two-roomed
tenements as of country population, and of these twice as many
were overcrowded in town as were in the country. Compared
with 1891 marked improvement had occurred in overcrowding
in towns, but very much more in the country districts. By
the side of these improvements have gone, as we have seen,
marked decreases in the prevalence of phthisis, and by the side
of the disparity in housing between town and country there
is the disparity already shown in the urban and rural phthisis
death-rate for males. The female death-rate, which would be
most strongly affected by home conditions, is substantially the
same for towns as for country, in spite of the enormous difference
in housing and overcrowding.

In the case of Ireland, the relations between overcrowding
and tuberculosis are masked completely.

It has already been seen (Fig. 31 and Table XLII.) that the
death-rate from phthisis in Ireland has increased. This higher
death-rate has been associated with a progressive improvement
in conditions of housing. The facts on which this statement is

TABLE XLIX
Percentage of Different Classes of Houses in Ireland

1841.

1861.

1881.

1891.

1901.

ist class
2nd ,,
3rd „ . . .

4th „

3-0
19-9
40-1
37-0

8'3
37-6
457
8-4

97
46-9
39-2
4 '2

10-5
53-6
33-8

2'I

1 1 '2

59'3
28-4
I 'I

lOO'O

100 '0

lOO'O

OVERCROWDED COMMUNITIES

227

I based (Table XLIX.) are taken from a paper by Dr. (now Sir T.)
I Matheson, Registrar-General for Ireland (1903).

The fourth class of houses comprises chiefly houses of mud
I or other perishable materials, having only one room and window ;
I the third class, a rather better class
I of house, having two to four rooms
li and as many windows ; the second
I class is equivalent to what would be
I considered a good farmhouse having
1 five to nine rooms and windows ;
I and the first class comprises all
I better houses. The changes in the
I proportion of these different classes
I of houses are set forth more clearly
I in Fig. 19.

Sir T. Matheson's conclusion is
I that "the material improvement in
|;j the housing of the people of Ireland
' ; since 1841 is very satisfactory, but
I that there is still much to be accom-
Iplished."

Comparing Ireland with England
I and Scotland, Sir T. Matheson finds
I that in 1901 in England 3-6 per
1 cent., in Ireland 87 per cent., and

• in Scotland 17*5 per cent., of the

• total dwellings consisted of only
I one room ; further, that the per-
I centage of the total population living
I in these one-roomed tenements and
I having five or more persons in each
I tenement was 0*15 in England, 178
I in Ireland, and 3-27 in Scotland.

I Thus Scotland has more than double
j i the proportion of one-roomed tene-
l nients that Ireland has, and in nearly twice as many of these
I the number of occupants exceeds five.

Contrasting these facts with the corresponding phthisis
{'death-rates, we see that some counterbalancing influence or
i influences have prevented Ireland from obtaining any lowering

FIG. 19. — Showing steady im-
provement in Housing Condi-
tions in Ireland.

228 THE PREVENTION OF TUBERCULOSIS

of its phthisis death-rate along with its improvement of housing,
and have enabled Scotland with a larger proportion of single-
roomed tenements and more overcrowding than Ireland to secure
a lower death-rate than the latter country.

In Paris the conditions of housing are extremely bad, and
the phthisis death-rate is high and probably almost stationary.
Over one-fourth of the total families were housed in single
rooms, and nearly one-third in tenements of two rooms, and
more than three-fourths in three rooms or less.

Official figures are available for Berlin for every five years
from 1861 to 1895. From these we learn that the number of
one-roomed tenements out of every 100 tenements of all sizes has
been about 50 throughout these forty-five years, while thejnumber
of two-roomed tenements in the same interval has only varied
from 24 to 27 per cent., of three-roomed tenements from 10 to
12 per cent., and of larger tenements from n to 12 per cent,
of the total number. A very large proportion of the population
of Berlin live in block-dwellings, and the average size of these
block-dwellings has increased. Doubtless the standard of these
dwellings as to cleanliness, as elsewhere, has improved; but it
is a remarkable fact that although half the families in Berlin
live in single rooms, the death-rate from phthisis in that city
has declined 45 per cent, between 1876-80 and 1901-03.

In Norway the census returns for the towns show that in
1891 the proportion of dwellings comprising one room was 42*4,
and comprising two rooms was 27^6 per cent, of the total
dwellings, while in 1900, the proportion of one-roomed dwellings
had decreased to 28*1 per cent., and of two-roomed dwellings
had increased to 34*5 per cent, of the total dwellings.

In New York a similar story has to be told. Dr. Hermann
Biggs (1903-04, p. 191) says :—

There has been a more rapid fall in the tuberculosis death-rate in
New York City than in any great city in the world, and this notwith-
standing the fact that the conditions in many respects are much more
unfavourable, because of the very dense population in the great tenement-
house districts of the city, and the large element of foreign born popula-
tion. It should be remembered that in no city of the world is there
such a density of population as exists in many of the wards of the borough
of Manhattan.

As illustrating Dr. Biggs' observation it may be stated that

OVERCROWDED COMMUNITIES 229

the phthisis death-rate was 4-27 in 1881 and 2*40 in 1903, a
I fall of 44 per cent. ; the corresponding rates in London being
2-18 and r6o and its fall 26 per cent.

Further figures comparing the conditions of housing in
different countries are summarised by Dr. Shad well (1905,
vol. ii. p. 198) in the following sentence : "In England the
industrial classes live in separate houses or cottages, in Germany
they live in barracks, and in America in larger houses which
are shared by more than one family." He adds: "We have
nothing to compare in England to the house famine which
prevails in Germany."

The outcome of the available figures is to show improve-
ment of housing associated with

(a) decrease of phthisis (England, Scotland),

(b) stationary or increasing phthisis (Ireland and Norway) ;
and heavily and increasingly congested housing associated with

(a) high and almost stationary phthisis death-rate (Paris),

(b) great decrease of phthisis death-rate, which is still high

(Germany, Berlin, New York).

It is highly probable that neither the association between
improved housing and reduced phthisis in Great Britain, nor
that between very congested housing and high phthisis rates
in the foreign countries quoted is accidental. In view of the
known pathology of the disease, no circumstance could be
more calculated to exercise a uniformly adverse influence on
this disease than overcrowding. Clearly, however, abnormally
high congestion of housing has been unable in most of the above
countries to prevent immense decrease in the phthisis rate ;
and marked improvement in housing in Ireland, which has
brought it well above the level of Scotland as to average number
of rooms per dwelling for the very poor, has not sufficed to
prevent the rise of the phthisis rate. Overcrowding must
therefore be classed with urbanisation as a factor which, though
of proved effect on the phthisis rate, has usually been unable
to overcome counteracting influences by which the phthisis
rate has been diminished.

CHAPTER XXXII

TUBERCULOSIS IN COMMUNITIES OF VARYING
WELL-BEING

THE influence of well-being on the phthisis death-rate has
never been questioned, and in the judgment of many
authorities it is the most important factor. Thus Sir
Hugh Beevor (1901, p. 158) says :—

As the wages rise, phthisis rate falls ; this fall affects especially the
young ; it is due to food supply.

In another place (1899) he says :—

The British public eat more and more. Agricultural returns declare
that in the last twenty years, the yearly ration per head of the public
had increased 10 per cent, in both bread and meat. . . . Nowadays,
patients at Nordrach rightly hold that their extra feeding is a great means
of cure ; nutrition is equally a means of prevention.

Sir Douglas Powell (1904) gives expression to the same view
in the following statement : —

The prevention of consumption involves a much wider issue than
the circumvention of the bacillus. . . . The abolition of the Corn Duties
and other Free Trade legislation, and improved rates of wages, have
done more than any notification law against the disease would have been
likely to have effected.

m
It may be assumed that, in the above extract, the action

which in a well-regulated district would follow on notification
is indicated.

Well-being is, of course, a very complex condition, which
cannot be measured completely by any single element. No
factor, however, more deserves careful attention, and in the
following pages its course is measured independently by the
price of wheat, the cost of total food, the total cost of living,

COMMUNITIES OF VARYING WELL-BEING 231

wages, the amounts of food consumed, and the amount of
pauperism. In considering those elements which relate to food
it must be remembered that we are dealing not with the thera-
peutic effect of these elements on tuberculous patients on whom
they are applied under exceptional conditions and in some excess,
but with their prophylactic influence taken in normal quantities
and in the circumstances of ordinary life. Much clinical experi-
ence appears to indicate that high feeding, especially with
proteids, has a marked beneficial effect in the treatment of
tuberculosis ; and although, so far as I know, there is no record
of its value apart from open-air treatment, and the latter may
therefore possibly be partially responsible for the beneficial
results ascribed to the former, it is likely that the high diet has
been at least an important factor in the therapeutical effect. It
is, of course, quite possible that food in no more than ordinary
amounts, and especially proteid food, may exert in health a
prophylactic influence against tuberculosis similar to the thera-
peutic effect on the consumptive exerted by abnormally high
amounts under open-air conditions. On existing evidence, how-
ever, it is equally possible that a certain minimum excess is
necessary for producing the predominant therapeutic effect
which has been remarked ; and a similar excess may conceivably
be necessary to the production of the fullest prophylaxis that
can be obtained by diet. There is, so far as I know, no evidence
to enable one to decide between these possibilities.

In using the figures which express the extent to which the
countries under comparison have enjoyed the several elements
of well-being, no correction is made for the varying benefit which
different persons and possibly different nations will have derived
from equal amounts of commodities. The absence of such
correction in the present inquiry is without serious importance.
The nation in whom thrift or superior efficiency in utilising their
means might have been supposed to have produced the decrease
in phthisis is Germany ; and if it were in fact shown that Germans
had such superiority over the other nations in question, then
the bare comparison of their means with those of less thrifty
nations would be inconclusive. In the present discussion, how-
ever, the inclusion of France and Norway, whose figures for
phthisis are very different from those of Germany arid whose
reputation for thrift is equally high, avoids the difficulty.

232

THE PREVENTION OF TUBERCULOSIS

PRICE OF WHEAT

In Table L. the proportional prices of wheat and the death-
rates from phthisis in several countries are given relatively to
the corresponding prices or rates in 1901-02, which are stated
as loo.

TABLE L
Relative Figures for Wheat and Phthisis

Wheat.

Phthisis.

ctf .

d) O

c/5

-o ft

i/5

a 2

^ ef

T3
§

|UH

1841-50 .

197

TT6

14.3

1851-60 .

201

140

147

186

229

209

206
169

162

...

246
233

...

(1851-55
\ 1856-60

1861-70 .

1 88

137

132

228

208

200

170

I7S

172

176

124

1 68

219
201

/ 1861-65

\i866-7o

1871-75

20 1

151

206

181

IS2

169

IIO

207

1876-80

175

139

130

164

1 66

146

186

i6s

1881-85

148

120

140

149

128

144

121

189

163

1886-90

116

114

107

112

134

114

128

121

164

1891-95

103

108

103

119

120

112

140

121

1896-1900

105

104

IOO

105

1 08

109

114

I04

119

104

1901-02 .

IOO

326

440

426

328

123

165

147

215

365

167

193

Absolute price in Death-rates per 100,000 from
pence per imperial Phthisis or Tuberculosis in 1901-02

gallon in the years or 1901-03.

taken as standard.

In Figs. 20 to 23 the facts of Table L. are shown diagram-
matically. By the use of proportional figures the curves of prices
and phthisis rates are reduced to the same scale, and can be
exactly compared.

Fig. 20 shows the phthisis and wheat curves for the United
Kingdom. As previously shown by Sir Hugh Beevor, there is a
fairly close relationship in Great Britain between the phthisis
and wheat curves. There is one important exception to this

COMMUNITIES OF VARYING WELL-BEING 233

statement. Prior to 1875 a great reduction of phthisis had
occurred, without cheapening of wheat.

In, Ireland, which has shared the benefits of cheaper bread,
there is obviously no relation between the price of wheat and

rF~T

—220

-210

7200^ £*-•-'> ^_

-190

-180

-170

—160

~l
~ O

-ISOo

I o»

—140

- v.

- o

—130 *.

I V

-.20^

i.
V

—no a.

-too
~90

- 80

FIG. 20. — Proportional Death-rates from Phthisis in England and Wales,
Scotland, and Ireland, and Price of Wheat in the United Kingdom, 1841-50
to 1901-03
Note. — The curves in Figs. 20 to 26 do not show actual prices and death-rates,

but only the proportional changes in them.

the death-rate from phthisis. It may be stated further that the
price of potatoes per cwt. in the ten years 1864-73 averaged
53d. ; in the ten years 1894-1903, it averaged 4od. These are
the means of the extreme values given in the Annual Reports of
the Registrar-General for Ireland.

234 THE PREVENTION OF TUBERCULOSIS

<" 2

V 8

FIG. 21. — Proportional Death-rates from Phthisis in Paris, 1861-69 to 1901-02,
•[and Price of Wheat, 1841-50 to 1901-02

150 5

-130 v

-110

v x

"* •.
"% 0
c *%
^ *•

^ -•-.

FIG. 22. — Proportional Death-rates from Tuberculosis in Prussia, 1877-80 to
1901-02, and Price of Wheat, 1841-50 to 1901-02

COMMUNITIES OF VARYING WELL-BEING

235

Fig. 21 shows the course of the phthisis curve for Paris and the
wheat curve for France. As already stated, it is probable that in
Paris the phthisis rate has declined little, if at all. Even if we
accept the official figures of declining phthisis, no correspondence
is visible between the official figures of variation of phthisis rate

ft1 'S1 '?' ' I ' '850
-aio s v $ iiliiii

FIG. 23.— Proportional Death-rates from Phthisis in Massachusetts, 1851-55
to 1901-02, and Price of Wheat, 1841-50 to 1901-02

and price of wheat. The proportional phthisis rate increased
from in in 1876-80 to 121 in 1881-85, while the proportional
price of wheat fell from 173 to 146. Between 1891-95 and
1901-02 the price of wheat has been almost stationary, and the
recorded death-rate has fallen from 112 to 100.

As will be seen in Fig. 22, the form in which figures are avail-

236

THE PREVENTION OF TUBERCULOSIS

able compels comparison between Germany and Prussia, and
also the substitution of tuberculosis for phthisis. Between
1876-80 and 1886-90, tuberculosis declined only from 164 in
1876-80 and 162 in 1881-85 to 151, while wheat declined from
130 to 107 ; while between 1886-90 and the present time, the
decline of wheat has only been from 107 to 100, that of tubercu-
losis from 151 in 1886-90 and 128 in 1891-95 to 100.

In the United States, where the margin of wages is great, and
where the price of wheat cannot be of such vital importance,
the two curves are fairly correspondent up to 1890, but then
diverge widely : a rise of wheat from 91 to 100 since 1891-95
having been associated with a fall in phthisis from 139 to 100
in 1891-95, and 119 in 1895-1900.

The data given above for the course of phthisis and of wheat
prices are connected by the following coefficients of correlation i1—

Price of Wheat and Phthisis Death-rates

Period of Observation.

Coefficient of Correlation.

England and Wales
Scotland
Ireland
Prussia
Paris

1866-1902
1868-1902
1866-1902
1877-1901
1866-1902

+ •90
+ •87
-•80
+ '55
+ •31

Expressed in words these figures summarise the preceding
tables and curves by showing a close co-variation between
phthisis rates and wheat prices in England and Scotland ;
moderate and poor co-variation in Prussia and France respec-
tively ; and considerable inverse co- variation in Ireland.

TOTAL COST OF FOOD

The data for a review of total cost of food in certain countries
from 1877 are furnished in Government Blue Books (1903,
pp. 215 and 224). " Index numbers " are employed in the
following table based on the retail prices collected by the Labour
Department of the Board of Trade, of bread, flour, potatoes,
beef, mutton, bacon, butter, tea and sugar ; value being attached

1 The sense in which this term is used is stated in the Note on p. 295.

COMMUNITIES OF VARYING WELL-BEING 237

to each of these articles in accordance with the annual amounts
spent by households in the purchase of the various articles.

TABLE LI
Relative Figures for Total Cost of Food and Phthisis

Total Cost of Food.

Phthisis in

Tuberculosis

United
Kingdom.

Germany.

England
and Wales.

Scotland.

Ireland.

in Prussia.

1877-80 .

135

112

1 66

157

165

1881-85 .

126

105

149

144

163

1886-90 .

102

134

128

1891-95 •

103

119

1 20

121

1896-1900.

108

114

104

1901.

100

IOO

(The cost of food and the phthisis death-rates respectively in 1901 are stated as 100
the other figures being given in proportion to the values for 1901)

The same values are also shown in Figs. 24 and 25.

Jv f I '

.•i\ i I I

FIG. 24. — Proportional Death-rates FIG. 25. — Proportional Death rates

from Phthisis in England and Wales, from Tuberculosis in Prussia, and

Scotland, and Ireland, and Cost of Cost of Food in Germany, 1877-80

Food in the United Kingdom, to 1901
1877-80 to 1901

238

THE PREVENTION OF TUBERCULOSIS

It will be noted that in and since 1886-90, the price of food
has remained almost stationary ; during the same period the
phthisis death-rate in England has fallen in the proportion of
134 to 100, and of Scotland in the proportion of 128 to 100. In
Ireland a rise of phthisis has been accompanied by a marked
decrease in the cost of food, though Ireland has experienced
the same cheapening of food as Great Britain.

In Germany (Fig. 25) between 1877 and 1886 the death-rate
from tuberculosis in Prussia was stationary, while the total cost
of food fell from 115 to 95, or from 112 to 105 in the consecutive
periods 1877-80 and 1881-85. On the other hand, in the period
1886-90, in which the cost of food was as low as in 1901, the
death-rate from tuberculosis was 50 per cent, higher.

The correlation coefficients which connect these data are
as follows : —

Total Cost of Food and Phthisis Death-rates

Period of Observation.

Coefficient of Correlation.

England and Wales
Scotland ....
Ireland ....
Germany ....

1877-1901
1877-1901
1877-1901
1877-1901

+ •90

+ •88

-'49

+ •42

These figures show close co-variation between the phthisis
rate and the total cost of food in England and Wales and in
Scotland, poor co-variation in Germany, and some inverse
co-variation in Ireland.

TOTAL COST OF LIVING

The figures enabling the relationship between total cost of
living and the phthisis death-rate to be stated, are derived from the
second Fiscal Blue Book (Memoranda, etc., Second Series). They
refer to workmen's expenditure in London and large towns in Great
Britain, the relative price in 1900 being in each case stated as 100.
The proportional costs in 1881-85 and m I9°° respectively were :
for food 133 and 100, for rent 89 and 100, for clothing 105 and
100, for fuel and clothing together 75 and 100 ; and for all the

COMMUNITIES OF VARYING WELL-BEING 239

above four chief items of workmen's expenditure 116 and loo.1
The cost of living in the United Kingdom has therefore declined
considerably, as compared with what it was in 1881.

Fig. 26 shows the course of the phthisis death-rate, and the
total cost of living in England and Wales.

The total cost of living in England has been fairly uniform
during the last fifteen years; during approximately the same
period the phthisis death-rate has declined in the proportion
of 134 to 100.

00
0)0)

130

o
o

120 0>

\1l»

0 5 \

— V.

100 (Jj

FIG. 26. — Proportional Death-rates from Phthisis in England,
and Total Cost of Living, 1881-85 to 1901-03

There are independent reasons for believing that in Ireland
the prices of total food, clothing, fuel, and rent have varied in
the same directions and approximately to the same extent as
in Great Britain ; and on this assumption the coefficient of
correlation has been calculated for Ireland as well as for England
and Scotland.

1 The proportional weights adopted in giving the data in Fig. 26 have been :
food, 7 ; rent, 2 ; clothing, 2 ; fuel and light, i, of a total expenditure on these
items of 12.

240

THE PREVENTION OF TUBERCULOSIS

Thus when to cost of food is added that of clothing and fuel
and rent, which in importance are second only to the cost of
food, the direct co-variation with the phthisis rate becomes less
marked in Great Britain and some inverse co-variation con-
tinues to be shown in the experience of Ireland.

Total Cost of Living and Phthisis Death-rates

Period of Observation.

Coefficient of Correlation.

England and Wales
Scotland ....
Ireland ....

1880-1903
1880-1902
1880-1903

+ 76

+ 76
-•24

WAGES

It may be suggested that the lack of correspondence between
cost of living and death-rate from phthisis may be due to the
disturbing effect of changes in wages. Unfortunately, exact
comparison of wages can only be made from official data for
workmen engaged in skilled trades and for agricultural labourers.
It is probable, however, that these wages give some clue to the
corresponding wages of other workmen.

Table LI I. compares the recent experience of different coun-
tries.

TABLE LII

Comparison of Rates of Wages in Skilled Trades

•gl

c
•

"c w

'c ^3

Dc/3

Number of quotations of wages on which the following

47o

248

184

141

s. d.

j. d.

Mean weekly wages for/I. Capital cities .
15 skilled trades \2. Other cities and towns .

42 o
36 o

36 o

22 10

24 o

22 6

75 o
69 4

Percentage comparison fi. Capital cities .
(United Kingdom — ioo)\2. Other cities and towns .

100
100

86
63

179
193

British money wages are the highest in Europe, and the
margin over the cost of living is probably the greatest in Europe.

COMMUNITIES OF VARYING WELL-BEING 241

The Board of Trade's Report gives the following comparison
of average family incomes : —

United Kingdom.
100

France.
83

Germany.
69

United States.
123

The preceding official data are confirmed by facts inde-
pendently collected by Dr. Shadwell (1905, vol. ii. pp. 81 and
91). He gives the following ratios for wages of unskilled
labourers in the three countries : —

England.
100

Germany.
79

United States.
143

and he believes that these figures more nearly represent the
actual state of matters than those in Table LIL, which give the
ratios for skilled workmen as 100, 57, and 179 in the capitals,
and 100, 63, and 193 in other towns.

The only comparison of wages practicable between different
parts of the United Kingdom is for agricultural labourers. The
data for this comparison are derived from an important report
by Mr. Wilson Fox, C.B. (Cd. 2376, p. 5). He gives the following
table :-—

TABLE LIII

Average Earnings per Week (including the Value of all Allowances in Kind)
of Able-bodied Male Adult Ordinary Agricultural Labourers

Percentage Increase

1902.

1898.

between

1898 and 1902.

s. d.

England

17 5

16 9

4-0

Wales ....

17 7

16 6

6-6

Scotland

!9 5

18 2

6-9

Ireland.

10 9

10 2

On p. 5 of the same report Mr. Fox remarks : " There is no
doubt that the position of a farm labourer in Ireland is not so
good as in other parts of the United Kingdom, but it may be
added that he gets his house and fuel cheaper, and frequently
has the opportunity of renting land on which he grows potatoes
and keeps pigs, goats, and poultry."

This report enables a comparison to be made for agricultural
labourers over a long series of years in the three parts of the
16

242

THE PREVENTION OF TUBERCULOSIS

United Kingdom. The following table illustrates the course of
wages on certain sample farms between 1850 and 1903. The
rates of$;wages are expressed in percentages, the year 1900
being taken to represent 100 : —

TABLE LIV

1850.

1860.

1870.

1880.

1890.

1900.

1903.

England and Wales (69 farms)

100

101

Scotland (6 farms)

§5

100

103

Ireland (10 farms)

IOO

101

Mr. Wilson Fox, in answer to an inquiry, kindly writes
me the following statement (May 16, 1906) : " As stated on
p. 220 of the Report " (On the Wages, etc., of Agricultural Labourers
in the United Kingdom), " the employers who furnished these
records were asked if the allowances in kind, given in addition
to cash wages, had varied during the period of years for which
wages were quoted, and you will see from the notes appended
to the various records that on the whole there was very little
variation, the tendency being to increase the extras as well as
the rates of wages. It seems safe to assume, therefore, that
there has been no diminution in the social well-being of farm
labourers in Ireland, and that the steady rise in wages shown
on p. 137 is not overstated/'

TABLE LV

Ratio of Average Rates of Wages in Different Countries (exclusive in all
Cases of Agriculture) (Cd. i76i,/. 275). Wages in 1900 = 100

United
Kingdom.

France.

Germany.

United
States.

Years.

Principal
Groups
of Trades.

Mean of
Skilled Trades.

Groups of
Principal Trades
under Imperial
Insurance Scheme.

Average
of all
Trades.

1881-85

83-4

86-9

90-5

1886-90

84-6

80-9

93'3

1891-95

89-4

...

84-9

95'8

1896-99

917

96-0(1896)

927

96-0

1900

lOO'O

COMMUNITIES OF VARYING WELL-BEING 243

Comparing the past with the present, there has been great
increase of wages all round (Tables LIII. and LV.).

The greatest increase has been in Germany, the least in the
United States. The above ratios indicate the course of wages
in each country, not the absolute amounts. Germany, which
shows the greatest increase of wages, still pays its workmen a
lower average wage than that in other countries. Unfortunately,
the comparison for Germany does not extend back beyond 1886.
Between 1886-90 and 1891-95 the death-rate from tuberculosis
fell 15 per cent., while wages rose 5 per cent.

In Norway between 1885 and 1900, wages have increased
in different industries from 24 to 53 per cent. Its phthisis
death-rate meanwhile has not decreased.

Thus in Germany and in Ireland wages lower than the
British are associated with a higher phthisis rate, while in the
United States much higher wages are associated with a much
higher phthisis rate. In Great Britain and the United States
rise of wages has accompanied decrease of phthisis ; in France
no such correspondence has appeared; and in Ireland and
Norway considerable increase of wages has been associated
with some increase of, or with a stationary death-rate from,
phthisis.

AMOUNTS OF FOOD CONSUMED

Without entering into the figures which are given in detail
elsewhere (1906, p. 343), it may be said that no uniform cor-
respondence is to be found between the figures of food con-
sumption per head of population and those of phthisis. England
with the lowest phthisis rate has by far the highest consumption
of meat, though not of other foods ; and Belgium, with sub-
stantially the same phthisis rate and the same decrease as
England in the period under examination, consumes less meat
than any country except Ireland, and less than half the amount
consumed in England. France, with a large and steadily
increasing consumption of meat and of other foods, has, judging
by Paris, the largest phthisis death-rate, with no certain evidence
of improvement.

PAUPERISM
Hitherto we have dealt with the experience of various

244 THE PREVENTION OF TUBERCULOSIS

countries in regard to the positive elements of well-being. It
remains to see to what extent these results can be checked by
figures expressing the absence of well-being. Owing to the
different methods of relieving poverty, we can only examine
the figures relating to poverty in the countries of the United
Kingdom, using for this purpose the poor-law returns. Before
doing so, it is desirable to realise what figures of pauperism
really indicate. Pauperism is officially-relieved poverty ; and
poverty itself, while most often due to absence of means, may
also arise from the unskilful, careless, or mischievous use of
means, from thriftlessness, sloth, or intemperance. The con-
ditions which accompany poverty, such as protracted exposure
to infection, insufficient nutrition, and ignorance, work in a
vicious circle with the conditions that cause it, till it is difficult
or impossible to distinguish those elements of poverty repre-
senting destitution, and relievable by the provision of ampler
means, from those which are of an origin independent of
material supplies, and which would persist even in a community
free from economic deficiencies. Poverty therefore is itself
a most complex phenomenon, not to be remedied by any single
set of measures ; and figures of actual poverty, even if they
could be had, would not in themselves suffice to estimate the
causes from which the poverty arose nor the steps which would
be necessary to remove them. In fact, however, we have not
figures of poverty, but only of pauperism, i.e. of State-relieved
poverty. The amount of pauperism depends obviously, not
alone on the extent of poverty, but also on the test or standard
by which the scale of relief is determined ; and a given amount
of poverty will beyond doubt yield very different figures of
pauperism at various epochs and in various districts according
to the scale of relief which happens to be applied. These con-
siderations need to be remembered when an attempt is made
to bring the complex phenomena of pauperism into relation
with experience as to phthisis. It will be seen, shortly, that
,jn the United Kingdom during the period under observation
there has been a correspondence between the variations of
phthisis and those of pauperism so marked as to justify the use
of the figures of total pauperism as approximate indexes of the
total amounts of phthisis, when the actual phthisis figures
cannot be had. This does not mean that the variations

COMMUNITIES OF VARYING WELL-BEING 245

in pauperism explain the variations in the death-rate from
phthisis. Within the bundle of phenomena which constitute
pauperism such an explanation may be found ; but until we
ascertain which individual element or elements of the bundle
contain the explanation, to explain the figures of phthisis by

ENGLAND & WALES

-y -

ert

FIG. 27. — England and Wales. Showing the relative Changes in the Number
of Indoor and of Total Paupers and in the Deaths from Phthisis per 100,000
of Population from 1857-60 to 1901-03

those of pauperism is for any practical purpose to explain a
complex ignotum by a yet more complex ignotius.

In considering the experience of Great Britain it must be

246

THE PREVENTION OF TUBERCULOSIS

remembered that about 1870 there was a vigorous and largely
successful movement for insisting on the " house-test " for relief ;
and the sudden drop of total pauperism about this date and
during the subsequent decade arose largely from this cause.

LONDON

Pa up

srr

::!ai

«o

FIG. 28. — London. Showing the relative Changes in the Number of Indoor
and of Total Paupers and in the Deaths from Phthisis per 100,000 of Popula-
tion from 1857-60 to 1901-03

COMMUNITIES OF VARYING WELL-BEING 247

Simultaneously there was great improvement in the workhouse
accommodation, particularly in its infirmary department. The
experience of Ireland has been even more striking, because in
the opposite direction to that of England and Scotland. In

SCOTLAND

10?

,>-

•*,

y /

—

•^

\—

•

•^

79"

^"*"

^•^

"**>[

*s»^

{il

— ,

— ~

\A.n

I 0 /•)

Jf*

JF*

~~£~~

is:

§
ft

FIG. 29.— Scotland. Showing the relative Changes in the Number of Indoor
and of Total Paupers and in the Deaths from Phthisis per 100,000 of Popula-
tion from 1857-60 to 1901-03

Ireland, as shown in Fig. 30, a rigid system in which indoor, i.e.
institutional, relief was almost alone given, has been superseded
by a largely outdoor, i.e. domestic, system. As in the England

RELAND

200

FIG. 30.— Ireland. Showing the relative Changes in the Number of Indoor
and of Total Paupers and in the Deaths from Phthisis per 100,000 of Popula-
tion from 1857-60 to 1901-03

COMMUNITIES OF VARYING WELL-BEING 249

of former times, this has been associated with a great increase of
official pauperism ; and apart from the facts which independently
make it improbable that this increase of official pauperism was
due to increase of privation in this very poor country, such a
sweeping change in administration must have produced an
increased number of paupers for a given amount of destitution.

Unfortunately there are no figures of pauperism for foreign
countries suitable for comparison with our own ; and it is there-
fore desirable to examine those of the United Kingdom with
some minuteness. The course of pauperism in each country of
the United Kingdom and in London is shown in Figs. 27 to 30.
In order to compare the curve of total pauperism in each instance
with the corresponding curve of the phthisis death-rate, the
curves of total pauperism and of phthisis have been reduced to
the same scale by stating the experience for the earliest period
in each instance as 100, and the subsequent rates in their pro-
portion to this.

It will be seen that if allowance be made for the reduction in
the relief figures introduced about 1870 by the more rigid insist-
ence on the " house-test," there is a correspondence between the
curves of phthisis and of total pauperism. The following table
shows the corresponding percentage declines of each for the
whole period and for its constituent quinquennia : —

TABLE LVI. — ENGLAND AND WALES
Percentage Declines of Rates of Phthisis and of Pauperism

0 ^
*> r^

o o JL
~

Phthisis Death-rate.
Total Pauperism -rate

52*3

9'4
177

8-1

22*1

IO'2

10-3
4-2

icrg
9'5

5 '4

6-8
6-0

The total decreases for the entire period — 49*8 per cent, for
phthisis and 52*3 per cent, for pauperism — are surprisingly
close. Individual quinquennia show some discrepancies ; but
as phthisis has a long course and may have a still longer period
of latency, and as any administrative influence is likely to operate

250 THE PREVENTION OF TUBERCULOSIS

slowly, a close quantitative relation between the figures for short
periods cannot be expected.

The correspondence in London and Scotland when allowance
has been made for changes in administration, though not so close
as in England and Wales, is nevertheless close.

In Ireland, if we make the necessary allowance for the great
increase of outdoor relief due to administrative causes shown
in Fig. 30, and compare the subsequent curve of pauperism with
that of phthisis, a close correspondence is seen. It would be
unsafe to assume on historical grounds alone that the lack of exact
parallelism between the earlier parts of the curves of phthisis
and pauperism is due merely or mainly to administrative change.
There is, however, independent evidence of the fact. It has
already been shown (p. 241) that the economic condition of Ireland
has not become worse, and that so far as can be measured by the
tests already given it has improved. Agricultural labourers in
1881 formed 46*0 and in 1901 44*3 per cent, of the total male
population of Ireland over 10 years of age ; and between 1870
and 1900 the wages of these labourers had increased 42 per cent.
Food has become cheaper, rents are low, overcrowding has
declined, and is less marked than in Scotland (p. 227). It is
clear that poverty has been growing less in Ireland during the
period of observation, and that the increase of pauperism has
therefore been due to altered administration and not to increase
of destitution.

The figures of pauperism and of phthisis for the entire period
are connected by the following correlation coefficients : —

Correlation between Total Pauperism and Phthisis

Period.

Coefficient of Correlation.

England and Wales
Scotland ....
Ireland ....

1866-1903
1868-1902
1866-1902

+ •89
+ •90
+ '83

These figures summarise a close co-variation in each of these
countries between phthisis death-rate and total pauperism..
This result is what would be expected from the pathology of
the disease. However minutely pauperism is analysed, each

COMMUNITIES OF VARYING WELL-BEING 251

element which is disclosed is such as would favour an increased
phthisis rate. In each of these countries, therefore, the figures
of pauperism confirm the a priori expectation that pauperism \
contains enough phthisiogenetic influences to make its figures
vary closely with the figures of phthisis.

CHAPTER XXXIII

TUBERCULOSIS IN COMMUNITIES OF VARYING SANITARY
EDUCATION AND SANATORIUM PROVISION

KOCH teaches on a priori grounds that direct infection has
a preponderating influence on the prevalence of phthisis ;
and the facts here reviewed will be found to lead by another
road to the same conclusion. In a passage quoted by Dr.
Bulstrode (1903, ii. p. 208), Koch says : " The fact that tubercu-
losis has considerably diminished in almost all civilised States
of late is attributable to the circumstances that knowledge of the
contagious character of tuberculosis has been more and more
widely disseminated, and that caution in intercourse with con-
sumptives has increased more and more in consequence."

This statement, so far as I am aware, has not been supported
by evidence, and it is by no means a consequence of Koch's
discovery that tuberculosis is infectious. Before such a state-
ment can be accepted, it must be shown, not only that caution
in intercourse with consumptives has increased, but also that
the increase of this caution occurred at a period and to an extent
warranting the inference. Prior to 1884 when Koch's discovery
of the tubercle bacillus was first fully set out, suspicion of in-
fectivity had no notable influence on medical or public action.
Had Koch's contention on this point been correct, the chief

TABLE LVII
Percentage Decline in Phthisis Death-rate

(

<»

oo
oo

§,

""*

England and Wales .
Scotland ....

9 '4
3'5

8-1
7-2

I0'2

8-0

I0'3

io'9

10-9

4'5

6-8
13-1 (1901-02)

VARYING SANITARY PROVISIONS

253

reduction of phthisis should have occurred since 1884. In
Germany this has been so : in Great Britain it is otherwise.

Table LVII. gives the quinquennial percentage decline of the
phthisis rate before and since 1885 in England and Wales and in
Scotland (the last period is two years).

The rate of decline was substantially as great before as
since the infectivity of phthisis became generally known to the
medical profession. In recent years the rate of decline has
diminished. In Scotland the rate of decline has been more
irregular.

The figures of other countries are interesting in the same

connection.

TABLE LVIII

Percentage Decline of the Death-rate from Phthisis or Tuberculosis

between

1881-85
and
1886-90.

1886-90
and
1891-95.

1891-95
and
1896-1900.

1896-1900
and
1901-02 or
1 903 or 1 904.

Switzerland

Prussia ....

Paris

In several of these countries a slackening of the rate of decline
of the phthisis death-rate is noticeable in recent years. It will
not be contended by the anti-contagionist that education and
consequent precautions have caused this diminution in the rate
of decline. Neither, on the other hand, is it possible to show
that the extremely limited action taken on directly preventive
lines has so far impressed itself on national statistics. As the
matter stands, there is no evidence of a causal connection suffici-
ently large to be traceable between the decline of the phthisis
death-rates and the progress of education in hygienic matters.

Similarly, no practical result can have followed from the
amount of voluntary or compulsory notification of phthisis
which has occurred in England. This is by no means because
notification has no useful part in the prevention of tuberculosis,
but because it is useless without the administrative mechanism
which is necessary for turning it to account for the welfare both
of the community and of the patient. No valid conclusions as

254

THE PREVENTION OF TUBERCULOSIS

to the utility of notification could be drawn from the experience
of towns which are not so equipped, or which have been so only
for a short term of years ; and in view of the important part
which notification should play in a properly arranged mechanism
for the control of tuberculosis, the error of attempting to draw
such conclusions is more than an academical fault, and is much to
be deprecated.

Nor conversely can it be imagined that similar educative
influences have been entirely absent from Ireland and Norway,
in which an increase, or from France in which probably no decline,
of phthisis has occurred. The action taken in consequence of
knowledge of the infectiousness of phthisis has doubtless varied
greatly in different countries and in different parts of the same
country. In Germany alone can treatment in special sanatoria
have any claim to the decline which has occurred, as the use of
these elsewhere has until a few years ago been on a very small
scale compared with the total amount of disease. Sanatorium
treatment, furthermore, has, with the same exception, been
employed chiefly for well-to-do patients who from the public

TABLE LIX. — SANATORIA IN GERMANY

Public.

Private.

Prussia.

Year Opened.

Tuberculosis

Number of Beds.

Death-rate

per 1000.

1854

300

1873

...

120

...

1875

1876

114

...

1881

100

307

1885

1887

...

100

290

1889

205

279

1892

248

1893

103

...

248

1894

275

237

i«95

196

231

1896

195

...

217

1897

504

214

1898

958

135

197

1899

590

119

202

1900

8i7

205

1901

794

196

1902

811

VARYING SANITARY PROVISIONS 255

health standpoint need it least. Even in Germany the sanatorium
treatment of phthisis was, as will be seen in Table LIX.,
on a very small scale until after 1892, when the first popular
sanatoria were opened (Santoliquido, 1903) ; and these institu-
tions cannot have played more than an insignificant part in
the great decline of the death-rate from tuberculosis which took
place between 1886-89 and 1890-93. Of the great value of
sanatoria in the treatment of phthisis there can be no doubt,
nor of their even greater educational value ; but their main
utility lies in the future.

CHAPTER XXXIV

THE GENERAL RELATIONSHIP OF INSTITUTIONAL
SEGREGATION TO TUBERCULOSIS AND CERTAIN
OTHER INFECTIOUS DISEASES

WE have seen that both general improvement in communal
health and each individual measure which tends to
produce it must work powerfully towards the reduction
of tuberculosis, but that nevertheless the disease has varied in
communal experience in a quite irregular relation to each and
all of these important influences. In the words of Sir William
Broadbent (1905, p. 118) : —

Supposing that the best possible sanitation, the best possible food, and
the best possible conditions of life, were an adequate protection against
phthisis, we ought to have no such thing amongst the better classes.
But it does get there somehow.

In Norway, Ireland, France, and Austria, the same influences
of improved general health, well-being, and sanitary education
have operated as in Great Britain, Germany, Belgium, and the
United States, side by side with widely different variations in
the respective death-rates in these countries from tuberculosis.
Similar discrepancies have been seen when other elements of
sanitary environment have been compared with the variations
of the disease.

It will next be seen that the only constant correspondent
between the variations in the prevalence of tuberculosis and ii
any element of sanitary environment consists in the relation t<
tuberculosis of the institutional segregation of patients.

Whether for good or harm, the segregation of infectiv<
patients is likely to influence the spread of tuberculosis. Th<
operation of this measure on tuberculosis follows obviousb
from the infectious character of the disease ; and it will

convenient here to recall what has been described on this subjecl

256

RELATIONS TO INSTITUTIONAL SEGREGATION 257

in Part I. The vast majority of pathologists and hygienists
are agreed that the chief source of infection in human tuber-
culosis is the tuberculous human patient. Whether he is more
infectious at early or at later stages has not been ascertained
definitely ; but in cases of pulmonary tuberculosis it may be
assumed safely that the infectivity varies with the amount of
the sputum. There is no evidence that with advancing disease
the patient becomes less able to disseminate infection ; on the
contrary, in advanced cases the patient is less able to control
its hygienic disposal. The period of latency of the disease
appears to be very variable. Small doses of infection lead to
immediate limitation of the disease, which may be followed after
a long interval by invasion of other parts of the body from the
localised tuberculous lesion. Pending such an explosion the lesion
may be utterly unrecognisable by clinical symptoms. Experi-
mentally, statistically, and clinically, it has been shown that
" the disease as a rule advances not by a continuous progress,
but by a series of successive invasions separated by variable
intervals. After each invasion, or, as it has been termed,
eruption of tuberculosis, there is a temporary self-limitation of
the disease." The earlier invasions may date years back.
During the patient's life they may be wholly unsuspected or
evidenced only by the recollections of an earlier attack of pleurisy
or haemoptysis, often many years prior to the diagnosed tuber-
culosis ; and this earlier attack may itself be a secondary result
of a still earlier disease in the bronchial or mesenteric glands.

The infection of tuberculosis, in short, is often acquired with-
out at the time causing any recognisable illness in the infected
person. Most acute infections, as for instance that of scarlet
fever, are either followed by a recognised attack of the disease
within a few days, or the person escapes entirely. The infection
of tuberculosis, while it appears to require a much larger dose
or more protracted exposure before evident disease is produced,
may, on the contrary, be saved up within the infected person
for years, and be discovered only after lapse of time and change
of circumstances have destroyed the chance of tracing its origin.
The infection which may be spread by an individual patient,
or even by a whole group of patients within the practice of a
single physician, may thus be wholly or partially concealed,
and give rise to a mistaken estimate of the infectivity of the
17

258 THE PREVENTION OF TUBERCULOSIS

disease. No better evidence of this fact can be needed than
the historical circumstance that for many centuries the existence
of any infectivity at all escaped recognition, and indeed did not
become accepted doctrine until it had been demonstrated by
actual experiment on animals. But though commonly unknown
by the patient and his family, and commonly unrecognisable
even to the physician in charge of the infecting case, the com-
municated infection remains within the body of the community
as a standing danger. In the proportion in which such latent
infections come ultimately to fruition as disease they are bound
to appear in the actual experience of the community ; and it
is necessary to turn to that experience for sure and unspeculative
guidance in seeking to master the disease.

It is evident that institutional segregation is different
qualitatively from domestic segregation. The average home,
both in its bedrooms and its living rooms, has far less special
accommodation per head, and a far lower standard of pre-
cautions against infection, than the average institution. Two
persons and often three may occupy the same bed in the
home; never more than one in the hospital. In institutions,
and by reason of the abundance of gratuitous labour, notably in
workhouse infirmaries in this country, the average standard of
cleanliness is far higher than in most homes. Spittoons and
spit-cups are provided and cleaned, washing of body and bed-
linen is not spared, and the floors, etc., of each room are kept
scrubbed and kept free from dust. In private houses, the
crowding of furniture, the presence of mats and carpets, and
the exigencies of life in the families of the poor, do not encourage
and sometimes do not even permit of such frequent and per-
sistent cleanings. It follows that the inmates of the home,
including children of the most susceptible age, must be far
more exposed to infection when the patient remains at home
than are the inmates of an institution to which he is transferred.
It remains to see how far the institutional segregation of
infective patients which is secured in institutions in general
has in actual fact served to control the spread of the disease.
Before turning to the facts of communal experience by which
these theoretical anticipations are confirmed, a hypothetical case
suggested by Sir Hugh Beevor (1905) may serve to illustrate the
order of magnitude of the influence under consideration.

RELATIONS TO INSTITUTIONAL SEGREGATION 259

Let it be supposed that no influence was operating to
control the prevalence of consumption except that of institutional
segregation. In Brighton 20 per cent, of the total consumptives
are segregated in its workhouse infirmary, and for the purpose
of this calculation this proportion may be supposed to hold
good for England and Wales. The examples given on p. 274
suggest that one-third of a year may be taken as the average stay
of each patient, and Sir Hugh Beevor in common with others
apparently would put the total period of infectivity at three years.
If these figures hold good for England and Wales, it follows that
just over 2 per cent, of the total infection of phthisis is prevented
from spreading outside institutions. On this supposition, and
if personal infection were the sole means of communicating the
disease, the death-rate from phthisis ought to have declined
in each year to the extent of the segregation, namely, 2 per cent.
A reference to Table LVII. shows that from 1871 to the present
time the decline year by year in the death-rate from phthisis
has been usually under 2 per cent. The calculation, although
interesting and suggestive, does not, of course, give any accurate
measure of the institutional segregation of phthisis, nor even of
its practical effect. Other influences besides segregation have
been operating, some to restrain and some to promote the
spread of the disease ; the extent of segregation may have been
more or less than has been assumed ; its quality must un-
doubtedly have varied from place to place ; and when figures
such as those of a single town are considered, the order of
magnitude of which is vastly less than those of the whole country,
the result is influenced by migration as previously indicated.
The calculation shows, however, that the influence of segregation
in institutions, as practised in England, has an order of magnitude
fully sufficing to explain by itself the decrease of phthisis which
has been secured, and it illustrates aptly the far-reaching result
which may be hoped for from the withdrawal of infection from the
community even to an extent which on careless inspection may
appear to be too slight to have exercised an appreciable effect.

A brief statement of the history of typhus fever in Ireland
and of leprosy in Norway throws some side-light on the influence
of segregation in two other infectious diseases, one very acute
and the other very chronic in its course. These diseases, like
tuberculosis, have in the past been associated very closely with

FIG. 31. — Comparison of the Changes in the Death-rates from Typhus and from

percentage deviations from the average

RELATIONS TO INSTITUTIONAL SEGREGATION 261

PHTHISIS.

Phthisis in Ireland and in England and Wales, as shown in each country by
death-rate for the entire period.

262 THE PREVENTION OF TUBERCULOSIS

unwholesome conditions of life, and the history of their decline
is instructive in its bearing on the problem of tuberculosis.

TYPHUS IN IRELAND. — The history of typhus in Ireland is
closely wrapped up with that of want and famine. Famine
has caused rapid spread of typhus, in the main because it has
increased enormously the wanderings of vagrants from one part
of Ireland to another, and to other countries. The disease
began to abate when fever hospitals were generally provided,
and when the families of infectious patients became relatively
immobilised by the provision of poor-law relief. Fig. 31
displays the course of the death-rate from typhus and from
phthisis in Ireland and in England since 1868. It will be seen
that typhus has declined greatly in both countries ; in England
it has approached extinction, and in Ireland it is following,
though more slowly, in the same direction.

Phthisis, on the other hand, though it has declined greatly
in England, in Ireland has not only not declined, but has even
shown some increase. In the light of these national experiences,
it can scarcely be maintained that diminution of domestic over-
crowding and improvement in housing, — which have been
regarded as the predominant factors in the decline of both diseases,
— can have produced for typhus a diminution in both countries,
and for phthisis a diminution in one country and none in the
other. The detailed facts given in Chapters XXX.-XXXII.
show that in both countries there has been marked diminution
of overcrowding, improvement in housing, and cheapening of the
means of living along with increase of wages. These facts
justify the inference that some differentia between the two
countries exists for phthisis, which does not exist for typhus
fever ; and the history of the two diseases in Ireland and in
England fits in with this inference. In Ireland the chief mass of
sickness, especially of phthisis, is treated domestically (see
pp. 280 and 282 for details, and especially p. 284 for the facts
bearing on the quality of institutional treatment in Ireland).
This is not the case in regard to typhus fever. By means of
fever hospitals and by preventing the wanderings of the poor,
the dissemination of typhus has been greatly diminished; an<
Ireland has secured a decrease of typhus, as has also Englan<
by similar means. In both countries, doubtless, diminishe<
domestic overcrowding and clearing of crowded courts an<

RELATIONS TO INSTITUTIONAL SEGREGATION 263

other dwellings has helped in producing the result ; but the
detailed experience of Ireland : clearly indicates that the im-
mobilisation of infection has been the chief operative factor.

LEPROSY IN NORWAY. — The history of leprosy forms an
interesting chapter in the history of disease, more particularly
so in its bearing on the history of tuberculosis. Both diseases
are caused by bacilli producing granulomatous tissue changes ;
in both there may be a long period of latency before the signs of
disease appear ; and in both the disease is commonly protracted
and intermittent in its progress. Both likewise are diseases to
which the designation " sub-infectious " has been applied,
though the name is misleading, and is no more applicable to
them than to syphilis, in which similar phenomena of long latency
of symptoms, and of protracted and intermittent course are
seen, and in which, furthermore, hereditary predisposition is not
known to occur. The further interest attaches to leprosy, that
acute differences of opinion exist as to the cause of its partial or
complete disappearance from England and some other countries,
which recall the similar differences of opinion as to the cause of the
great decline of tuberculosis in certain countries during the last
forty years.

The history of the disappearance of leprosy has been associated
with the existence on a very considerable scale of leper asylums
in the countries from which the disease has disappeared. In
mediaeval England such lazar houses were numerous, and
although complete segregation of all patients was never secured,
there doubtless was segregation of a large percentage of the total
cases during a considerable part of their illness. Here again the
resemblance to what has been happening in the case of tuber-
culosis, as will be shown shortly, is striking. There is no intrinsic
difficulty in accepting it as fact that in leprosy, — in which, as in
tuberculosis, infection occurs chiefly after protracted contact
of an intimate character, — the isolation of lepers must, if
carried out to a sufficient extent, have served to bring about a
steady decline and eventual disappearance of this disease. This
conclusion is confirmed by the experience of Norway, which
amounts almost to a check experiment. In this country until

1 Further details of the history of typhus in Ireland are given in an address by
the author on " Poverty and Disease as illustrated by the Course of Typhus Fever
and Phthisis in Ireland " (Journal of the Royal Society of Medicine, Dec. 1908).

264

THE PREVENTION OF TUBERCULOSIS

far on in the nineteenth century there were few leper asylums. As
Dr. Vandyke Carter put it, there never prevailed in Norway " the
same systematic and rigorous opposition to the leprous pest as
was aroused in Europe generally." During the first half of the
nineteenth century leprosy was increasing in Norway. Thus the
yearly average number of fresh cases of leprosy ascertained and

NORWAY - LEPROSY

200

FIG. 32. — Norway. Number of Total Lepers and of Lepers in Asylums per
100,000 of population, 1856-60 to 1901-05

registered in 1840-45 was 43, in 1846-50 it was 124, in 1851-55
it was 219, in 1855-60 it was 233, and in 1861-65 it was 225.
Even allowing for the possibility of increasing accuracy of
registration, it is clear that there was no decline in this disease.
In 1856 notification of cases by medical men became com-
pulsory, and for all years onwards the official statistics state
the total number of known cases of the disease and the number

RELATIONS TO INSTITUTIONAL SEGREGATION 265

segregated in asylums. The diagram on preceding page shows
these facts for quinquennial periods. It will be observed that
the steady pursuit of an intelligent policy of segregation of leprous
patients, — almost entirely without compulsion,1 — has been asso-
ciated with a steady and continuous decline of the prevalence of
leprosy. At no time has there been total segregation of all known
cases. Of the total cases about 16 per cent, were segregated in
1856-60, 27 per cent, in the next period, 30 per cent, in 1871-75,
32 per cent, in 1876-80, then 36 and 46 per cent, in the two next
periods, the proportion of segregation in the three most recent
quinquennial periods being about 52 per cent, of the total known
cases. In the light of our knowledge that leprosy is a com-
municable disease, of its history in other countries, and of the
close correlation between the phenomena of segregation and
diminution of disease (which is expressed by a coefficient of
correlation of '95 for the entire period), it is reasonable to give
the chief place to segregation as the means by which the diminu-
tion of disease has been secured.

1 Some indirect compulsion has been exercised by refusing non -institutional
relief.

CHAPTER XXXV

TUBERCULOSIS IN COMMUNITIES WITH VARYING
AMOUNTS OF INSTITUTIONAL SEGREGATION

THE exact measure of institutional segregation of phthisis is
the ratio stating how many of the total days of sickness
(number of patients and number of days of sickness) is
passed in institutions. This ratio and the equivalents for it
which have to be used in practice may for convenience be called
the segregation ratio. The need for equivalents for the ratio as
stated above arises from the fact that we are dealing with actual
recorded experience, and the statistical material has to be taken
from the records as they happen to exist. These records appear
in very various forms in different communities. In existing
circumstances of notification they can never state directly the
number of days of tuberculous sickness, and only exceptionally for
comparatively small communities can they state the number of
such days passed in institutions. It becomes necessary therefore
to select other figures which vary approximately with the total
days of tuberculous sickness and the total days of tuberculous
sickness passed in institutions. Such figures may represent
them respectively on quite different scales ; but so long as com-
parison is made only between segregation ratios, in which the
substituted figures represent similar phenomena, the particular
scale on which they represent the phenomenon of institutional
segregation is of no consequence. From the records in various
countries we can learn either how many of the total deaths
from all causes and from tuberculosis or from phthisis occur
in institutions, or how many of the total paupers are indoor
paupers, or how many cases of tuberculosis or phthisis are
treated in institutions, and how many deaths from these diseases
occur in the whole community for each case treated in an
institution.

From what has been said, it will be seen that these figures

266

AMOUNTS OF INSTITUTIONAL SEGREGATION 267

measure with approximate accuracy the ratio which states how
many of total days of tuberculous sickness are passed in institu-
tions. Thus, for instance, in the absence of change of type of
disease and of material change in efficiency of treatment, the
number of deaths from tuberculosis is an approximate measure of
the number of cases, and so is the number of deaths from all
causes for short periods during which the relation of the death-
rate for phthisis to that for all causes does not vary markedly.

The fraction of total deaths in the population occurring in
institutions is by far the most direct measure of the amount of
sickness, and Table LX., calculated from the census returns,
shows for England and Wales how preponderantly public
institutions are occupied by the sick and not the healthy.
Deaths are taken at the average for 1891-95 and 1901-03 respec-
tively, the difference between these and the deaths for 1891 and
1901 being immaterial for the present purpose.

TABLE LX

Per 100,000 of Total Population and per 100 Deaths in Total
Population there were in

Workhouses

including Work-
house Infirmaries
and Schools.

Hospitals.

Lunatic
Asylums.

Total
Institutions.

c o

- 0

C 0

fi O

g||

§31

w 2 ta

z 8^

8§l

S-23

ill

^^ s,

•5^0

111

all §,

Iffl

J2 rt a
•5 i> o

c »-.

v O P-i

s ^>

S O PH

c ^ ^

Q °

c t>

rt Q P-i

** 0,3

083

i> r3
cu^3

083

^ 0*3

o 83

Q 8"^

1-1 O

- 0

M o

1891 .

630

7'i

3*5

276

I'l

1001

117

1901 .

641

8-1

120

5'5

280

i-5

1041

The fraction of deaths in the total population occurring in
public institutions was accordingly fifteen times as large as the
fraction of the total population which was housed in these
institutions.

Apart from figures of mortality, the nearest approach to a
satisfactory index of tuberculosis is probably to be found in the

268 THE PREVENTION OF TUBERCULOSIS

number of the pauper population. It is the last part of the
population to be reached by ameliorating influences tending to
control tuberculosis, and would therefore be expected to have a
higher sickness rate than the general population, and to yield
figures of which the variations will correspond with some accuracy
to the variations in prevalence of tuberculosis. We have seen
that this theoretical expectation has been verified, at least for
the United Kingdom, in the close co- variation of the numbers
of paupers and of deaths from tuberculosis respectively over
a long period ; and the numbers of paupers relieved during the
periods here in question do therefore actually represent on some
scale those of total cases of tuberculosis during the corresponding
periods.

In using these indirect measures of institutional treatment
of tuberculosis and of its prevalence, it must be remembered
that they are indirect and approximate. Thus, for instance,
figures for institutional treatment usually give the number of
cases and not days of treatment, and while they tell how many
people were segregated in institutions, do not show the average
duration, still less the quality of the treatment. Any of
these indirect forms of segregation ratio has therefore to be
verified wherever possible by the application to the same com-
munity and period of one or more other forms of the ratio, and
checked where practicable by a special examination of sample
constituent communities whose figures are included in the total.
This has been done so far as the information obtainable has
allowed. It will be seen that the results obtained by applying
different ratios to the experience of the same country and period
are usually, though not invariably, in good agreement ; and
where this is not the case, fortunately other data have been
available to explain the discrepancy and enable a more correct
segregation ratio to be formed.

Where, again, the segregation ratio — the proportion of sick
days spent by consumptives in institutions — is expressed as the
proportion of total paupers who receive indoor relief, it is assumed
that the number of days of sickness is the same in each class.
This assumption is probably incorrect ; but to such extent as
consumptives admitted to indoor relief are, in fact, treated
longer than the average of other paupers, the error would be to
exhibit the extent of segregation as being less than it really is,

AMOUNTS OF INSTITUTIONAL SEGREGATION 269

and for the present purpose the figures may therefore be used
with safety.

As has been pointed out previously, the phthisis rates with
which these ratios should be compared are not those for the same
but for a somewhat later period, the interval representing the
time taken for the effect of segregation to show itself. For the
present purpose this comparison can in any sufficiently long
series of years be made with the phthisis figures of the same year,
not because the phthisis is affected immediately by simultaneous
changes in other phenomena, but because the numerical differ-
ence between closely consecutive phthisis figures in the present
material happens to be small.

The countries in which the fullest records of experience have
been obtained in regard to institutional segregation are England
and Wales, Scotland, and Ireland. It is not always realised how
large a proportion of the total population is at any one time in
public institutions ; and, without quoting the actual figures,
Table LXI. shows to the nearest whole number the number of
total population at the censuses of 1891 and 1901 to every one
inmate of a public institution.

TABLE LXI

For every Inmate of a Public Institution the Total Population of the

Country was

England and Wales.

Scotland.

Ireland.

1891 . . .

164

1901 ....

137

The figures available for England and Wales and for London
permit a statement of the fraction of total deaths in the popula-
tion occurring in institutions, which, as we have seen, is one of
the measures of the amount of institutional segregation. Tables
LXII. and LXIII. give these figures, together with those of the
death-rate from phthisis for a considerable period. They show that
the decrease in phthisis was accompanied by a large and steady
increase in institutional segregation measured by the fraction of
total deaths occurring in institutions ; and the rate at which
these changes occurred is shown more conveniently in Figs.

270

THE PREVENTION OF TUBERCULOSIS

33 and 34, in which the rate of change of the phthisis death-rate
is shown by the side of the rate of change of the segregation
ratio, the curve for the segregation ratio being inverted as shown
on the left-hand scale.

TABLE LXII. — ENGLAND AND WALES
Percentage of Total Deaths in Public Institutions

Workhouses

Death-rate

Years.

and
Workhouse
Infirmaries.

Hospitals.

Lunatic
Asylums.

Total
Institutions.

per looo of
Population
from
Phthisis.

1869-70

1-9

8-3

2'45

(1866-70)

1871-75

8-8

2-22

1876-80

6-3

2-4

0-9

9-6

2-04

1881-85

6-6

2-9

I'O

10-5

I-83

1886-90

3 '4

i'i

1 1 '2

I-64

1891-95

7-2

3'9

1*1

I2'2

I'46

1896-1900

4-6

I '4

137

I-32

1901-03

8'5

5'9

1-8

16-2

1-23

TABLE LXII I. — LONDON
Percentage of Total Deaths in Public Institutions

Years.

Workhouses
and
Workhouse
Infirmaries.

Public,
Lunatic,
and
Imbecile
Asylums.

M. A. B.
Hospitals.

Other
Hospitals.

Total
Institu-
tions.

Death-rate
per 1000 of
Population
from
Phthisis.

1852-55

9'6

167

1856-60

9-0

0-6

16-3

1861-65

9-0

0-4

16-2

2'8o

1866-70

9-1

o'5

16-3

2-86

I87I-75

9'8

o'5

173

2-51

1876-80

ii'3

0-4

18-6

2-40

1881-85

12-3

0-4

20-5

2-u

1886-90

1 1 -8

i'9

8V7

23-1

1-88

1891-95

13*3

2'O

9'4

267

1-87

1896-1900

14-8

2'I

IO'2

29-2

i -80

1901-03 .

177

2-8

2 '2

I2'2

347

1-65

(1901-04)

Thus in England and Wales, in the period 1866-1903, segrega-
tion measured by the fraction of total deaths occurring in in-

AMOUNTS OF INSTITUTIONAL SEGREGATION 271

stitutions has approximately doubled, and the death-rate from
phthisis has approximately halved ; in London segregation has
not quite doubled and the phthisis death-rate is rather more
than half. The closeness of numerical correspondence may be
and probably is accidental, for, as pointed out above, close
numerical concordance is not to be expected in the courses of
complex associated phenomena operating among other complex
influences. The data show, however, not only a very close

FIG. 33. — England and Wales. Logarithmic Curves showing Rates of Change in
the Phthisis Death-rate and in the Proportion of Institutional to Total
Deaths from all Causes

correspondence between the increase of total institutional
segregation measured by the ratio in question and the decrease
of phthisis, but an even more striking similarity i|(. the rates at
which these changes have occurred. The experience is summar-
ised in the high correlation coefficients of -91 for England and
Wales (1878-1903) and "90 for London (1866-1904).

27:

THE PREVENTION OF TUBERCULOSIS

The experience so far as it is available of the chief individual
classes of institutions exhibits the manner in which this result
has been obtained.

Workhouse infirmaries have been the most important agency

FIG. 34. — London. Logarithmic Curves showing Rates of Change in the Phthisis
j, 4 Death-rate and in the Proportion of Institutional to Total Deaths from all
Causes

in segregation. These institutions are used to a much greater
extent for tuberculosis than in the earlier history of poor-law
administrations. Figs. 27 and 28, expressing the data of Tables

AMOUNTS OF INSTITUTIONAL SEGREGATION 273

LXV. and LXVL, have shown the general reduction which has
occurred in total pauperism side by side with the steady mainten-
ance of indoor relief at a stationary level in England and Wales,
and an actual increase of indoor relief in London. In 1848-49
over 60 out of every 1000 inhabitants of England and Wales
were paupers as against 20 in 1902-03. The whole of the re-
duction was in persons receiving outdoor relief, and the number
of indoor paupers remained stationary at from 7 to 8 per 1000
of population. Thus of the total pauper population, who, as we
have seen, are the most subject to disease of all kinds and notably
to tuberculosis, the segregation in workhouses in 1848-49 amounted
to about one-eighth, and was increased by 1902-03 to over one-third.
The fact expressed in these figures is explained by Mr. Fleming,
who speaks of the " great change in the character of workhouse
inmates during recent years. . . . The able-bodied inmates are
gone and the sick inmates have come " (1902-03, p. 84). When
the frequency of tuberculosis is remembered, these figures and
this fact become equivalent to the statement that, as has been
seen already for the total institutions for England and Wales,
there has been during a period of vast reduction in tuberculosis
also a vast increase in the extent of segregation of tuberculous
patients in workhouse infirmaries.

As a matter of practical importance, individual inquiry has
been made among 27 Boards of Guardians in London and 85
of the chief provincial towns, to ascertain the extent to which
workhouse infirmaries treat consumptives in separate wards.
In 12 of the metropolitan infirmaries out of the 27, consumptives
were treated wholly in the same wards as other patients, and : in
only 9 were they treated entirely in separate wards. Out of the
85 provincial infirmaries only 23 treated consumptives wholly
and 13 partially in separate wards. It appears therefore that,
although separate treatment is not rare, the more common
practice is to treat consumptives in general wards. Incidentally
it may be observed that taken in context with the general reduc-
tion in the prevalence of phthisis, this fact is very striking evidence
of the superiority of segregation in infirmaries over what is
practicable at home, and agrees well with the general considera-
tions to which attention was drawn on p. 258. It must be
remarked further that, although these results show great good
to have arisen without the use of separate wards, it is obviously
18

274 THE PREVENTION OF TUBERCULOSIS

desirable to have consumptive patients treated separately when
it can be arranged.

Figures are not available in most cases to express the duration
of stay of consumptives in workhouse infirmaries . For all diseases
the average number of days' stay for each patient in certain
provincial infirmaries in 1 897 was: Salford,97; Leeds, 95 ; Croydon,
86; Birmingham, 74; West Derby, 60 ; Kensington, 48. From the
nature of the disease the stay of consumptives was probably longer
on the average ; thus in Kensington in 1897 all patients had
an average stay of 48 days, consumptives of 144 days in 1898
and 95 in 1902. In Sheffield in 1904 the average stay of each
phthisical patient was 311 days, and in Brighton 221 days.
While therefore the segregation of each patient must have
extended over a large portion of the period of his illness, there is
considerable variation in the period of segregation in different
towns. The existence of this variation indicates that while
increased segregation in institutions has been followed by de-
crease in phthisis in various towns and countries, the decrease
caused by institutional segregation must have varied at least
according to the differences in average duration of treatment
and according to any other variations in the efficiency of the
segregation.

After workhouse infirmaries, the most important institutions
for segregation of tuberculosis are lunatic asylums. The per-
centage of lunatics treated with relatives and others was 18*4
in 1859, and fell to 5-5 in 1902. The death-rate from tubercu-
losis in borough and county asylums in 1901 was 15*8 per cent,
of the inmates, or over ten times as great as in the general popula-
tion. Of these tuberculous lunatics the majority were tuber-
culous on admission, according to the results of Dr. Mott
(1905). Subject therefore to such allowance as maybe required
by the fact that lunatics seldom expectorate,1 the segregation
of each tuberculous lunatic has been equivalent to the with-
drawal from the community of ten ordinary tuberculous persons.
The proportion of lunatics in asylums to the total population in
1902 was over 0*3 per cent., and their segregation must therefore
be taken to have been equivalent to the withdrawal of say 3 per
cent, of normal population or the same amount of average

1 They are often dirty in their habits, and large numbers of tubercle bacilli
must be passed in the faeces.

AMOUNTS OF INSTITUTIONAL SEGREGATION 275

infection from the community. The average stay of each
patient is about five years, or far longer than in any other great
class of institutions. When the considerable increase in the
extent to which lunatics are now lodged in asylums is considered,
it is evident therefore that during the period of decline of tuber-
culosis a large, sustained and increasing segregation of tuberculous
patients has taken place in these institutions.

The disproportion between accommodation and need in the
case of special hospitals is too great for them to have had a large
effect on the total amount of tuberculosis. In the past con-
siderable numbers of consumptives were treated in general
hospitals, but the returns of most of them show an increasing
unwillingness to admit such patients. Thus in the Royal
Infirmary (general hospital) of Glasgow the proportion of
total deaths due to phthisis has fallen from 16*9 per cent, to
4 per cent. With this decrease of treatment of phthisis in
general hospitals has been associated the great increase of its
treatment in workhouse infirmaries.

The experience of large towns has been similar to those of
the whole country. For the reasons described on p. 207, the
experience of small towns into and out of which there is much
migration is, like the experience of separate quarters of large
towns, of very doubtful value. In certain towns the segregation

TABLE LXIV

Brighton.

Sheffield.

Salford.

Proportion

Phthisis
Death-
rate.

Per Cent, of
Total Deaths
from
Phthisis in

Phthisis
Death-
rate.

Per Cent, of
Total Deaths
from
Phthisis in

Phthisis
Death-
rate.

Per Cent, of
Total Deaths
from
Phthisis in

Institutions.

1866-70

2-95

9-6

...

1876-80) '

2'47

1.1*7

2-23

6-3

1881-85!
1886-90! '

i-93

14-3

1-90
170

7-9

10-3

2-36

14*4 l

1891-95 \
1896-1900) '

1-63

15-8

i '35

14*3

20'0

1-94
178

19-2
23*5

1901-04

1-40

20*2

I-252

26-1 2

1-82

-27-6

2 1901-05.

276 THE PREVENTION OF TUBERCULOSIS

ratio has been obtained in the more direct form of the part of
the total deaths from phthisis which occurred in institutions.
Of the total deaths in London from phthisis, 31*4 per cent, in
1889 and 33*5 per cent, in 1904 occurred in workhouses, work-
house infirmaries, and sick asylums ; in Sheffield the proportion in
workhouse infirmaries and sick asylums was in 1876-80 only 63
per cent., and it rose in 1901-05 to 26*1 per cent. ; in Salford in
1884-90 it was 14*4 per cent., rising to 27*6 per cent, in 1901-04 ;
in Brighton it was 9-6 in 1866-70, rising to 20*2 per cent, in
1901-04. The course of these figures is set out in Table LXIV.
by the side of the phthisis death-rate for the towns in question,
and, as was seen in the country as a whole, and for institutions
as a whole, there is shown constant increase of segregation in
workhouse infirmaries accompanying constant decrease of
phthisis.

Coefficients of correlation summarising this correspondence
for a long series of single years work out at '67 for Salford
from 1884 to 1904, and *8o for Sheffield from 1876 to
1905.

Summarising all this experience, it will be seen that in England
and Wales a large and continuously increasing amount of insti-
tutional segregation of phthisis, measured by the fraction of
the total mortality occurring in institutions, has been accom-
panied for nearly forty years by a large and continuous decrease
of the disease, and that throughout the entire period each of
these changes has gone on at much the same rate as the other.
The same association appears when segregation is measured
in the more direct form of the fraction of deaths from phthisis
in the whole community occurring in institutions as seen in
the experience of certain large towns.

These results may now be compared with those obtained by
regarding segregation as measured by either the fraction of
total pauperism which is treated in institutions, or the ratio
in which the number of paupers treated in workhouses and
workhouse infirmaries stands to the total number of deaths
from phthisis in the community. The results obtained in either
of these ways confirm the conclusion obtained by the use of the
other measures of segregation.

Table LXV. is a summary in quinquennial periods of the
data for this comparison for the individual years from 1866 to

AMOUNTS OF INSTITUTIONAL SEGREGATION 277

TABLE LXV. — ENGLAND AND WALES

Number per 100,000 of
Population of

Segregation Ratio.

For every 100 Indoor
Paupers there were the
following Number of

Deaths from
Phthisis.

Indoor
Paupers.

Total
Paupers.

Deaths from
Phthisis.

Total
Paupers.

1866-70

1871-75
1876-80
1881-85
1886-90
1891-95

1896-1900
1901-03 .

245

222
204
183
I64
146
132
123

726
662
668

730
709
687

4652
3828

2870
2749
2489
2356
2218

34
3i
31
25
23

21
19

641

578

446

393
388
362
340
322

1903. A clearer view of the total result is given in Table LXVL,
which shows for England and Wales, and also for London, the
respective percentages which the phthisis death-rate and the
segregation ratio in question of 1901-03 are of the corresponding
figures of 1866-70.

TABLE LXVI

. In

Phthisis
Death-rate for
1901-03 as
Per Cent, of
Phthisis
Death-rate for
1866-70.

Patio Ind°°r

Indoor Pauperism

° Total
Pauperism
for 1901-03 as
Per Cent, of
same Ratio for
1866-70.

10 Total Phthisis Deaths
for 1901-03 as Per Cent,
of same Ratio for
1866-70.

England and Wales .
London .

50
58

50
38

53
44

This experience for the entire series of individual years is
expressed by a coefficient of correlation of — '94 between segre-
gation measured by the fraction of pauper population treated
in institutions and the phthisis death-rate.

The rate at which segregation, measured by comparison
of indoor and total pauperism, has varied is shown in context

278

THE PREVENTION OF TUBERCULOSIS

with the rates of variation of the death-rate from phthisis in

Fig. 35-

Each of these results is closely similar to that obtained
by the previous measures of segregation. In the whole country
segregation, measured in any of the ways, has approximately
doubled, while the death-rate from phthisis has been halved.

FIG. 35. — England and Wales. Logarithmic Curves of Phthisis Death-rates
and of Ratio of Indoor to Total Paupers, 1861-65 to 1901-03

In London exactly the same has happened ; measured by the
fraction of the pauper population treated in institutions, the
amount of segregation has more than doubled.

No figures are available for Scotland or Ireland by which
segregation can be expressed in terms of institutional deaths.
Measured by the other ratios, the data for Scotland are given in
Tables LXVII. and LXVIII., and in Fig. 36.

AMOUNTS OF INSTITUTIONAL SEGREGATION 279
TABLE LXVIL— SCOTLAND

Segregation Ratio.

Number per 100,000 of

Population of

For every 100 Indoor

Paupers there were the

following Number of

Deaths
from
Phthisis.

Indoor
Paupers.

Total
Paupers.

Deaths
from
Phthisis.

Total
Paupers.

1866-70.

259

253

3896

1 02

1540

1871-75 .

248

224

3210

H33

1876-80.

230

235

2597

1105

1881-85 .

211

236

2742

1162

1886-90.

1 88

224

2168

968

1891-95 .

176

212

1978

933

1896-1900

168

227

2085

919

1901-03 .

147

242

1922

794

FIG. 36.— Scotland. Logarithmic Curves of Phthisis Death-rates and of
Ratio of Indoor to Total Paupers, 1861-65 to 1901-0^

280

THE PREVENTION OF TUBERCULOSIS

In Scotland as in England the facts for the two terminal
periods as given in the following table bring out more clearly
the relationship between the different factors.

TABLE LXVIIL— SCOTLAND

Phthisis Death-rate
for 1901-03
as Per Cent, of
Phthisis Death-rate
for 1866-70.

. Indoor

. Indoor Pauperism

Ratio Total laupcnsm

for 1901-03
as Per Cent, of
same Ratio
for 1866-70.

Ratio Total phtHisis Deaths

for 1901-03
as Per Cent, of
same Ratio
for 1866-70.

As in the experience of London, the proportionate extent
of segregation appears to have been somewhat larger when
measured by the ratio of indoor to total paupers than when
measured by the more direct ratio of indoor paupers to total
deaths from phthisis in the whole community. On both
measures, however, these data show very close correspondence
between increased segregation and decrease of the death-rate
from phthisis; and in the more direct segregation ratio, given
in the 3rd column approximately, the same numerical closeness
appears between the increase of segregation and the decrease
of the phthisis death-rate as was seen in the experience of
England and Wales and of London ; a decrease of the phthisis
death-rate of about 56 per cent, having been associated in
Scotland with an increase of about 60 per cent, in institu-
tional segregation. As with England and Wales, the rates
at which segregation has increased throughout the entire period
have been much the same as the rates at which the death-rate
from phthisis have declined. The experience is summarised
by a coefficient of correlation of — -91 between segregation, ex-
pressed as the fraction of total pauperism treated in institutions,
and the phthisis death-rate.

The data for Ireland are given in Tables LXIX. and LXX.

In Ireland a decrease in the amount of institutional segrega-
tion has been accompanied by an increase in the death-rate
from phthisis ; and measured by the more direct segregation
ratio, there is again numerical identity between the extent

AMOUNTS OF INSTITUTIONAL SEGREGATION 281

TABLE LXIX. — IRELAND

Segregation Ratio.

"Number per 100,000 of

Population of

For every 100 Indoor

Paupers there were the

following Number of

Deaths
from
Phthisis.

Indoor
Paupers.

Total
Paupers.

Deaths
from
Phthisis.

Total
Paupers.

1866-70 .

182

963

1233

128

1871-75 .

190

882

1389

I58

1876-80 .
1881-85 .

200
208

903
1019

1569
2198

22
20

174
215

1886-90.

2I3

954

2332

244

1891-95 .

2I4

906

2204

243

1896-1900

213

944

2244

237

1901-03 .

215

947

2272

240

TABLE LXX. — IRELAND

Phthisis Death-rate
for 1901-03
as Per Cent, of
Phthisis Death-rate
for 1866-70.

. Indoor

. Indoor Pauperism

for 1901-03
as Per Cent, of
same Ratio
for 1866-70.

Katio Total phthisis Deaths

for 1901-03
as Per Cent, of
same Ratio
for 1866-70.

118

1 86

121

to which the death-rate from phthisis has increased and the
extent to which institutional segregation has decreased. In
two respects, however, the experience of Ireland appears to
differ from that of England and Wales and of Scotland. The
absolute amount of segregation, although steadily decreasing,
has nevertheless, so far as gross figures are concerned, been
greater than in England and far greater than in Scotland, while
the phthisis death-rate has not only increased but has from
1881-85 onwards been higher than in England and from 1886-90
onwards than in Scotland. Moreover, the rates at which the
apparent extent of segregation has changed in Ireland during
the period in question show much less numerical concordance
with the corresponding changes in the phthisis death-rate than

282 THE PREVENTION OF TUBERCULOSIS

has been seen in the experience of England and Wales and of
Scotland. Each of these discrepancies is merely one of quantity
and not of kind, and leaves segregation and the death-rate
from phthisis varying universally as in England and in Scotland.
Their explanation throws a light on the practical working of
institutional segregation.

Theoretically the discordance might be due to one or more
of three causes. The concordance in England and in Scotland
might have been mere coincidence. This explanation, as will
be seen shortly, is inadmissible because the comparison of in-
stitutional segregation with phthisis in a considerable number
of other countries shows similar concordance. Presuming
therefore that institutional segregation tends to reduce phthisis,
it might be that in Ireland the influence of factors tending to
increase phthisis has been greater than in either of the other
countries. To some extent this has probably been the case ;
but although it might assist in explaining the greater prevalence
of phthisis at the present time in Ireland than in England or
Scotland, it has no bearing on the increase in Ireland itself,
unless Ireland at the present time is in a worse economic and
sanitary condition than in the past, which, as already seen, is
not the case. An examination of the demographical and adminis-
trative conditions of the country gives, however, independent
and direct explanation of the lower specific result produced by
institutional segregation in Ireland. It has been seen already
that the population of Ireland contains a smaller proportion
than either England or Scotland of persons at the ages specially
liable to die from phthisis, and a higher proportion of persons
at the ages when pauperism mostly occurs. Apart, therefore,
from any question of specific efficiency, the specific result of
pauper segregation must have been lower in Ireland than in
England or in Scotland. This apparent reduction of specific
result of segregation in workhouses is the greater because, as is
shown in the Reports of the Irish Local Government Board,
many artisans and labourers when sick, in the absence of other
medical institutions, resort to the workhouse infirmary for all
classes of diseases ; and their cases, which would include a
much lower proportion of tuberculosis than occurs among
paupers, swell the figures of apparent segregation. So much
is clear as to the specific result of what appears as segregation

AMOUNTS OF INSTITUTIONAL SEGREGATION 283

in Irish experience, apart from any question of its specific
efficiency. There is, however, unanimous and conclusive
evidence that the quality of the segregation is notably inferior
in Ireland to that given in England or in Scotland. The extent
of institutional segregation is greater in Dublin than in the
rest of Ireland, the indoor paupers in the Unions of North and
South Dublin numbering 94 per 1000 in 1903, as compared
with 80 per 1000 in the rest of Ireland. The average stay of
each pauper in workhouses in North and South Dublin is 70 days,
in the rest of Ireland 39 days. Clearly therefore the institutional
segregation of phthisis may be taken to be more extensive in
Dublin than in the rest of Ireland. Yet Sir Charles Cameron
(Ann. Rep. 1904, p. 31) says concerning Dublin : —

" The hospitals rarely keep consumptives whose cases are
hopeless, to the termination of their disease by death. If such
cases were retained in hospital, it would prevent the circulation
of much tuberculous infective matter."

This statement is confirmed by the data contained in a return,
kindly supplied by Mr. J. E. Devlin of the Irish Local Govern-
ment Board, which has enabled me t- calculate the average
duration of residence of phthisical patiei ts in the Dublin work-
houses. It is shown in the following tal le, in comparison with
similar returns for English workhouses.

TABLE LXXI

Average Residence (in Days) of all Phthisical Patients in Workhouses, to
Time of Discharge or Death (not including Patients still in the
Institution]

Based on Experience of the

Undermentioned Number of

Patients who have

Institution.

Days.

Left the

Died in the

Institution.

North and South Dublin Workhouses,

1904-05

272 156

Brighton Infirmary, 1897-1905 .
Kensington Infirmary, 1888 " >.

175
144

165
107

181
68

>, „ 1902

I51

• 112

Sheffield (Firvale) Infirmary, 1904

3ii

284 THE PREVENTION OF TUBERCULOSIS

The above return relates to North and South Dublin, which in
1903 had a population of 379,666.

It will be noted that, unlike the experience of Kensington
Infirmary (see p. 274), the institutional residence of consumptive
patients in the Dublin workhouse is less than that of all patients
in the aggregate.

In addition to the necessarily low specific effect of segrega-
tion in Ireland due to the constitution of the population, to
the much shorter duration of average residence in workhouse
infirmaries in Ireland than in England or in Scotland, and to
the very imperfect conditions of Irish workhouses which diminish
the efficiency of segregation, the great increase in outdoor relief
must have exerted a powerful influence in promoting the pre-
valence of tuberculosis, owing to its inevitable effect in increasing
domestic at the expense of institutional treatment, and to its
effect in continuing an enormous number of domestic foci of
tuberculous infection such as are invariably implicated in the
average home treatment of phthisis among the poor.

On these grounds the lower specific value of institutional
segregation in Ireland need not be taken into further con-
sideration.1

The experience of the United Kingdom will now be com-
pared with that of foreign countries, and it will be seen that
the inquiry is carried into a larger number than was used in
examining the other factors of phthisis. This course is desirable
in regard to segregation and was unnecessary for the other
factors, because each of the factors discussed earlier in this paper
showed failure to maintain co-variation between the factor and
the phthisis death-rate in one or more of the countries examined.
This failure does not appear when segregation is tested over the

1 Comparisons have been freely made in this inquiry between the condition
of different countries at a given period as regards food, housing, etc. ; but the
necessity of caution in making a similar comparison between different countries
as regards segregation has been emphasised. The reason for this is obvious.
Such factors as a given amount of food, of house accommodation, wages, etc.,
mean much the same in any country, and can with approximate accuracy be
compared with the corresponding phthisis death-rates in each country. It is
otherwise with segregation until we can obtain more accurate measures of its
duration and its character as well as of the number of segregated persons.
Administrative variations like those shown in the experience of Ireland are
enormous ; and country can only be compared with country so far as the general
trend of observation goes. Each country needs separate study as to the
contents of any institutional segregation which its statistics show.

AMOUNTS OF INSTITUTIONAL SEGREGATION 285

1891-1900.
189

1901-02.

192 per 100,000 of Population1

same countries, and it is therefore necessary to extend the inquiry
over a wider area in order to make sure that the continued con-
cordance was not fortuitous.

The death-rate from phthisis in Norway (1904, p. 30) was —

1881-90.
141

In 1902, of the total deaths in Norway 5*9 per cent,
occurred in hospitals and lunatic asylums. The average
duration of treatment of all the patients treated in hospitals in
1902 was 35 days. It is evident, therefore, that there is com-
paratively little institutional treatment of sickness in Norway
as a whole, together with increasing phthisis. Separate hospital
statistics could not be obtained for Christiania, but facilities
for hospital treatment are doubtless more extensive than in
the rest of Norway, and there has been considerable fall in its
phthisis rate.

No Swedish statistics for the entire country are obtainable.

TABLE LXXII
Death-rate per 100,000 of Population from Phthisis

1861-70.

1871-80.

1881-90.

1891-1900.

All Swedish towns together .
Stockholm .....
Gottenburg .....
All other towns ....

306

433
279

195

324
406
326
299

300
346
322

277

270
292

303
256

Stockholm is the only town of Sweden showing any marked
decline in its phthisis rate. The detailed statistics show, both
in small and large towns, either insignificant declines, or a
stationary phthisis rate. There are few hospitals in Sweden,
as shown by the following extract from the report to the Paris
Congress on Tuberculosis (1905, p. 205) : —

" Notwithstanding the excellent general organisation of
Swedish hospitals, only a small number of consumptives can
be treated in them, owing to the fact that the great majority
of the hospitals were organised only for the case of acute diseases.
The official figures for 1890-1900 show that only about 1500
tuberculous patients have been treated each year in all the
provincial hospitals of the kingdom, while the number of

1 See footnote on p. 2 1 3.

286

THE PREVENTION OF TUBERCULOSIS

patients suffering from tuberculosis is about 60,000 (1905,

P- 4)-"

Stockholm is better furnished with hospitals than the other

towns, and it alone shows any decline of phthisis, though its
death-rate is still very high.1

As regards Denmark, statistics are obtainable only for Copen-
hagen. These have been kindly furnished by Dr. E. M. Hoff.

TABLE LXXIII. — COPENHAGEN
Phthisis and Hospital Treatment

Percentage of the

Cases of Phthisis

Phthisis Death-rate

Total Deaths from

treated in Hospitals

Years.

per 100,000 of

Phthisis which

Per Cent, of Total

Population.

occurred in

Deaths from Phthisis

Hospitals.

in the Population.

1860-64

307

1865-69

297

...

1870-74

342

1875-79

.. .

1880-84

289

1885-89

251

1890-94
1895-99

205
183

1900-04

149

147

Evidently there is, as Dr. Hoff states, a large amount of
institutional treatment of phthisis in Copenhagen ; and he adds
that the average number of days' treatment for each patient
has in recent years increased much more rapidly than the number
of patients. More recently, further particulars have been
published (1905, p. 7). It is stated that —

" Notwithstanding the enormous increase of accommoda-
tion required, owing to the growth of the town and new ideas
concerning phthisis, up to the present all requests for admission
have been satisfied ; and no consumptive desiring to be admitted
has hitherto been refused owing to lack of room."

In 1895, on an average 40 beds in the municipal hospitals
were always occupied by consumptives (deaths from phthisis

1 R. Koch quotes Carlsson's statement that 410 cases of pulmonary phthisis
are being cared for in the hospitals of Stockholm, " no small number for a city
of 300,000 inhabitants " (Lancet, 26, v. 1906, p. 1450. Nobel Lecture on " How
the Fight against Tuberculosis now stands ").

AMOUNTS OF INSTITUTIONAL SEGREGATION 287

in that year in Copenhagen, 661) ; in 1904, the number of beds
thus always occupied was 270, not including the Sanatorium
of Boserup (deaths from phthisis in Copenhagen in 1904 were
632). The mean duration of treatment of three successive series
of cases of phthisis, in years 1890-1904, was as follows :—

TABLE LXXIV. — COPENHAGEN

Mean Duration of
Stay in Hospital
in Days.

Mean Duration of Stay (Days) in
Hospital of Patients

Dying in the
Hospital.

Leaving the
Hospital.

Series I. .

„ II. - -
„ HI. .

40
107
107

112

40
105
no

The reduction of phthisis in Copenhagen, therefore, has been
associated with a large amount of institutional treatment of the
disease in general hospitals. The co-variation of the phthisis
death-rate for Copenhagen during the period of 1880-1904 and of

TABLE LXXV

Prussia.

Berlin.

Rate per 100,000 of
Population of

For every
loo Deaths

Rate per 100,000 of
Population of

For every
loo Deaths

from

Tuber-

Years.

culosis the

Cases of

Number

Cases of

Number

Deaths

Tuber-

of Patients

Deaths

Tuber-

of Patients

from

culosis

with Tu-

from

culosis

with Tu-

Tuber-

treated in

berculosis

Tuber-

treated in

berculosis

culosis.

General

treated in

culosis.

General

treated in

Hospitals.

Hospital

Hospitals.

Hospital

was

1877-80 .

319

337

231

1881-85 •

311

332

255

1886-90.

291

294

282

1891-95 .

248

244 l

291 l

119

1896-1900

212

213

313

1901-02 .

I92

124

210

284

136

1 Returns for 1891 missirrg.

288 THE PREVENTION OF TUBERCULOSIS

the deaths from phthisis which occurred in the hospitals of
Copenhagen is summarised in a correlation coefficient of '57.
When segregation is measured for the same period by the pro-
portion of cases of phthisis treated in hospitals to total deaths
from this disease, the coefficient of correlation with the phthisis
death-rate is '68. These figures (Table LXXV.) express a fair
co-variation between segregation as measured above and the
phthisis death-rate.

Table LXXV. shows that, while in the whole of Prussia the
number of cases of tuberculosis treated in general hospitals has in-
creased from 14 for every 100 deaths from this disease in 1877-80
to 64 per 100 deaths in 1901-02, the death-rate from tuberculosis
has declined from 3*19 to 1/92 per 1000. Similarly in Berlin
the number of cases treated in Berlin has increased from 69
per 100 deaths from this disease in 1877-80 to 136 per 100 deaths
in 1901-02.

There is reason for believing that the duration of treatment
as well as the number of hospital patients has increased. It
will be noted (Table LXXV.) that the proportion of cases treated
in hospital was greater throughout in Berlin than in Prussia.
Collateral evidence shows that the duration of treatment of
each patient has been shorter in Berlin than in Prussia. Ap-
proximately while Berlin had 153 beds (for all patients in its
general hospitals) for every 100 in Prussia, it had 241 patients
for every 100 in Prussia, for equal populations.

The above experience is summarised in correlation co-
efficients between the annual returns of segregation and of
phthisis or tuberculosis death-rates of -95 for Berlin and -93 for
Prussia, showing close co-variation of the two phenomena.

It will be remembered that the general hospitals indicated
above are not sanatoria. The limited operation of the latter
has already been described on p. 254.

In Brussels the death-rate from tuberculosis has declined
from 3 "2i per 1000 in 1886-90 to 1*97 in 1901-03. In the two
great hospitals of Brussels (St. Jean and St. Pierre) the number
of deaths from tuberculosis to every 100 in the whole city was
12*2 in 1886-90, I5'6 Jin 1891-95, 17-3 in 1896-1900, and 38-9
in 1901-03. I am unable to obtain further information as to
the character and duration of the hospital segregation of con-
sumptive patients in Brussels, but the experience of Brussels

AMOUNTS OF INSTITUTIONAL SEGREGATION 289

appears to fit in with that of Copenhagen and of English towns.
The correlation coefficient between the annual segregation
ratios from 1888 to 1903 and the corresponding phthisis death-
rates in Brussels is 76.

In 1902, 4828, i.e. 41 per cent, of the total deaths from
tuberculosis of the lungs and larynx in Paris occurred in its
public hospitals. The average duration of stay in hospital of
all patients admitted to its general hospitals was only 23*6
days in 1901 (Dr. J. Bertillon). The institutional treatment of
phthisis in Paris is very short, and can have but little effect
in preventing infection. We have already seen that in Paris
there is probably no considerable decline of the death-rate
from phthisis, and that it remains much higher than that of any
other city for which statistics have been obtained.

There is among the medical profession of Paris an impression
that the Paris hospitals are a focus for tuberculous infection.
Thus, M. Mesurier states that the hospital attendants " suffer
cruelly from contagion in the wards, two-thirds of them be-
coming tuberculous (1905, p. 9)." He states also (1905, p. 16)
that the hospitals contain 30 to 40 per cent, of consumptives.
On the other hand, Dr. S. Bernheim, Vice-President of the
Societe Internationale de la Tuberculose (1905, p. 173), states : —

" The Paris hospitals scarcely suffice for patients suffering
from acute diseases, and can only, in view of their number,
exceptionally admit consumptives. Furthermore, all the
hospitals in our large centres of population, were they restricted
to the treatment of tuberculosis, would not suffice for a tenth
part of the consumptive poor of these towns."

The two statements here quoted can be partially reconciled
by the fact that Paris hospitals are generally so overcrowded
that consumptives make a very short stay in them.

Dr. Bernheim, in a later paragraph, says : —

" A consumptive never improves in our hospitals. We
can allow the death in one of our beds of a consumptive with
cavities ; and, on the contrary, the curable consumptive has
his fever increased in the presence of patients with serious
lesions ; and, in the inevitable overcrowding, rapidly passes
beyond the first stage of the disease, and on leaving the hospital
has no further prospect of recovery/ In this sombre statement I
leave out of consideration the contamination of the hospital ;
19

290 THE PREVENTION OF TUBERCULOSIS

and do not wish to speak of the unhappy typhoid patient who
often leaves the hospital with consumption which he has acquired
there/'

On the whole, it may be said that in balancing the possibilities
of infection in Paris homes and hospitals, it is doubtful on
which side the dangers are greatest. These hospitals, with a
few exceptions, cannot under recent conditions be regarded
as institutions tending to reduce total infection. As a whole,
neither the extent of accommodation nor the average length
of treatment is comparable with what is found in other countries.
This, coupled with the uncertainty of the death returns, would
make it unsafe to include the French statistics, even if they
were available, in the consideration of the problem.

In the cities of the United States a considerable and increasing
proportion of cases of phthisis are institutionally treated. In
Cincinnati, in 1885, 18*6 per cent., and in 1902-04, 34-6 per cent.,
of the total deaths from phthisis occurred in its public institu-
tions. In San Francisco, in 1885-87, 30 per cent., and in 1902-
04, 38 per cent., of the total deaths from phthisis occurred in
its public institutions. In New York, in 1884, the death-rate
from phthisis was 3 '86, in 1903 it was 2*40 per 1000 of popula-
tion. In 1882-84, 22*0 per cent., and in 1901-03, 26*0 per
cent., of the total deaths from all causes occurred in public
institutions. Dr. Hermann Biggs writes me that he cannot
give separately the number of deaths from phthisis in the public
hospitals of New York ; but he states that a census of tuberculous
patients in the public institutions in the boroughs of Manhattan
and the Bronx has been taken twice a year for a series of years,
and that the number of beds available for phthisis has greatly
increased. At the present time there are 2100 to 2200 beds,
chiefly for the care of advanced cases. Fifteen years ago the
number specially devoted to this purpose was scarcely more
than a quarter of this number, certainly not in excess of one-
third. He adds that in little more than a year they will probably
have over 3000 beds for tuberculous patients : though even
this number is insufficient. The number of deaths from phthisis
in Manhattan and the Bronx in 1903 was 5250. This implies—
assuming the above beds to be always occupied — that every
advanced case of phthisis in the city has had in recent years
an opportunity of being segregated in a hospital during 21

AMOUNTS OF INSTITUTIONAL SEGREGATION 291

weeks. Doubtless a smaller number, representing the poorest
and therefore the most dangerous part of the phthisical popula-
tion, were segregated for a correspondingly greater part of the
year.

During the years 1881-1903 the coefficient of correlation
between the phthisis death-rate and the proportion of deaths
occurring in public institutions was '75. This figure in itself
shows a well-marked co-variation of the phenomena in question.
Its significance is the more notable when it is considered in
connection with the amount of overcrowding in New York.

CHAPTER XXXVI

THE RELATIVE INFLUENCE OF INSTITUTIONAL SEGRE-
GATION AND OF OTHER MEASURES FOR THE CON-
TROL OF TUBERCULOSIS

results disclosed by Chapters XXVIII. to XXXIII. may
JL be said to have added nothing of practical value to the
knowledge described in Part I. of this volume. They
indicate the probability that tuberculosis is affected to a greater
or less extent by general sanitary conditions, town life and over-
crowding, and the various elements of well-being ; but the
probability disclosed in this way is not so strong as that result-
ing from the facts given in Part I., which indeed place the con-
nection beyond doubt. Neither line of investigation, however,
has succeeded in measuring the respective extent of influence
exerted by the important factors in question.

The experience of institutional segregation differs from
that of the other factors of the death-rate from tuberculosis,
both because the nature of its influence on the prevalence of
the disease cannot be inferred with certainty from the facts
given in Part I., and because not only the nature but the relative
extent of this influence is demonstrated clearly from the statis-
tical results. On theoretical grounds it has long been recognised
that the institutional segregation of patients suffering from
an infectious disease may influence its prevalence in two ways.
It may restrain the disease by segregating foci of infection
from the general population, or it may spread it by exposing to
infection from these foci persons in or about the institutions not
suffering from the disease in question. With tuberculosis it has
till recently been a moot point whether these theoretical results
actually appear in practice, and which of them is the more
important. The records of segregation analysed in the pre-
ceding pages give a decided answer to this question. Each
group of records shows, not as a matter of hypothesis or theory,

MEASURES FOR CONTROL OF TUBERCULOSIS 293

but as the teaching of actual experience, which gives the final
touchstone for final conclusions and action, that with no more
precautions than are taken in well-conducted general infirmaries
the increase of institutional segregation has been associated
with reduction of tuberculosis in the community affected by it ;
and that the segregation of a decreased proportion of the total
bulk of tuberculosis has been associated with an increase of
the disease. The scale of the observations and the number of
communities examined is so large as to eliminate the chance
that this correspondence has been due to mere coincidence ;
and it follows that these associations of segregation, with
the prevalence of tuberculosis, have not been accidental, but
have occurred because segregation has had an influence on
the disease, and because it has done more to restrain infection
than to spread it.

By comparing the several experiences of the communities
examined, we have been able to obtain information as to the
relative importance of institutional segregation and of the
other factors of the death-rate from tuberculosis. We have
examined the records of a large number of communities exhibit-
ing the respective variations of the several factors affecting
the death-rate from tuberculosis side by side with the variations
of this death-rate. Each of these factors was thus tested in
the actual experience of many large communities over the
same period of history. In the series of communities subjected
to this test, institutional segregation was the only factor of which
the variation was always associated with a variation in the pre-
valence of tuberculosis in a constant relative direction. It would
not have been surprising had the influence of institutional segrega-
tion been masked by that of opposing factors, as has been seen
(p. 221) to have occurred in many countries with the important
influence of urbanisation; or contrariwise, it would not have
been surprising if more than one influence had varied with the
prevalence of tuberculosis in a constant relation. In either
case the question as to which influence had predominated in
affecting the prevalence of tuberculosis would have been left
open. In fact, however, no influence except that of institu-
tional segregation has appeared in actual experience in a constant
relation to the amount of tuberculosis, and it must .therefore
be accepted as having been the predominant influence.

294 THE PREVENTION OF TUBERCULOSIS

The administrative consequences flowing from this result
are obvious in principle from what has been stated previously,
and further reference in detail is made to them in Part III.

(P. 394).

Some general reflections may be permitted as to the method
by which the result has been obtained. It has involved
necessarily much repetition of inquiries concerning the factors
of the prevalence of tuberculosis as the experience of each
country came under review ; in many of these experiences
questions subordinate to the main issue have had to be asked
and answered by further reference to communal experience
in order that doubts arising in the course of the investigation
might be eliminated. The presentation of the argument would
have been far simpler and easier if the number of these reitera-
tions had been reduced and the doubts ignored ; but the results
would have been inconclusive and intellectually dishonest.
Those who have read this section attentively may have found
some or all of it tedious and wearisome ; the collection, calcula-
tion, and above all the conspective criticism of its data has
certainly been far more tedious and wearisome. Such, how-
ever, is the condition upon which alone the records of com-
munities large enough to be worth studying by this macro-
scopic method will consent to give up their secrets.

The experience which these records contain is not arranged

in the orderly sequence of a text-book, but is intermingled in an

almost endless intricacy. The chief difficulty in handling it

lies in arriving at the assurance that the material examined

is sufficient for the purpose in view. The temptation to stop

short of what is necessary for sound conclusions does not lie

mainly in the reluctance to continue the protracted labour of

accumulating, arranging, and comparing data ; nor to persons

of elementary scientific honesty does it consist in the fear that

continued investigation may upset conclusions previously

reached ; but rather in the fact that many of those whom the

solution most concerns may decline to follow the more detailed

argument associated with protracted investigation, when it

becomes as intricate as it has to become if the results of the

investigation are to be trustworthy. Such investigations are

apt to be judged by summaries which are often imperfect,

misleading, or even inaccurate ; and the work is subjected not

MEASURES FOR CONTROL OF TUBERCULOSIS 295

to the welcome criticism which is based on equal labour, but
to random and often irrelevant conjectures, hypotheses, and
speculations.

Although the continued search for the full truth may, as
indicated above, even obstruct its recognition, no part of the
search can be omitted with safety. The attempt to find a
royal road to truth and to express it as a whole by suppressing
essential parts, leads too often to indolent work and slovenly
thought ; and this in the public health service is not to be
tolerated. We are not engaged in academic labours, of which
the prize shall go to the winner, and it is at the choice of each
man to neglect his preparation. The servant of public health
is working on the lives of men, and should be laying the founda-
tions of national prosperity and happiness. He belongs to an
order of sanitary priests, and if he forms or announces con-
clusions without having used fully and faithfully the material
at his disposal, he belies his vocation and abuses his trust.
' The day is short, and the work is much, and the labourers
are slothful, and the reward is great, and the master of the
house presses."

NOTE ON CORRELATION COEFFICIENTS

The coefficient of correlation between two columns of figures is a
number, never greater than unity, which expresses the closeness with
which deviations of figures in one column from their mean value follow
deviations in the corresponding figures of another column from their
mean. In the case of perfect direct correlation, i.e. when all corre-
sponding deviations from mean values vary in the same sense of excess
or deficiency and bear the same ratio to each other, the coefficient is i ;
in the case of perfect inverse correlation, where the senses of variation
in corresponding pairs of figures are opposite and the ratio of their magni-
tudes is the same, it is -i ; and it may have any intermediate values
according to the nature of the case. The closer the coefficient is to + i,
the nearer is the approach to constant co-variation of the pairs of figures ;
and where no influences but those represented by the figures are operating,
a high correlation coefficient on a sufficient number of figures is the
numerical expression of strong inductive evidence that there is some
connection — whether causal or otherwise is a matter for subsequent
discussion — between the phenomena represented by the two groups of
figures. In practice it is rare for two groups of phenomena to be free
from disturbing influences; and the correlation-coefficient measures
therefore for practical purposes the influence of one group of phenomena
on the other to such extent as it predominates over or is assisted by the
other influences in operation. Within certain limits the manner in which

296 THE PREVENTION OF TUBERCULOSIS

the deviations are measured may vary according to the circumstances
of the case. The effect of any such variation would, however, only be to
alter the final result by a relatively small amount ; and coefficients of
correlation, computed on any single system, represent the closeness
of relations between such curves as appear in Part II. far more dis-
tinctly than any general impression that can be derived from mere in-
spection of the curves. The usual form taken for this coefficient is the
ratio of the arithmetical mean of the products of corresponding devia-
tions in each group of figures from the arithmetical means of the values
in the respective groups to the product of the square roots of the arith-
metical means of the sums of these deviations squared ; that is to say

where x and y are the deviations from the arithmetical means of the
respective series.

Without discussing the precise mathematical reasons for the selection
of this form of coefficient and the processes by which its validity is demon-
strated, it is worth while to verify the fact that, by whatever mathe-
matical considerations the coefficient in question may have been obtained,
it is a quantity of which the magnitude must always depend on the closeness
with which the phenomena to which it refers stand in some relation to
each other. This may be seen very shortly. It can be shown by simple
algebra, and is here assumed to have been proved, that this fraction
can never be greater than i. If the two groups of phenomena were
unconnected by any causal link whatever, that is to say, if there was no
reason why a deviation xn of any figure in one group from the arithmetical
mean of that group should be accompanied by a deviation ±yn of dependent
magnitude and constant relative direction in the corresponding figure
of the other groups, then in any long series of pairs the deviation of figures
in each group from the arithmetical mean would be as often positive as
negative, and their values would be distributed evenly on each side of
the mean. Hence the products of the pairs of deviations (#y) of which
the sum (2#;y) forms the numerator of the fraction will be as often
positive as negative, and when added together with their proper signs
will exactly balance each other, and the sum will be o. In other words,
when there is absolutely no causal link between the phenomena, this
correlation coefficient will become o. If there is any causal link, then
to such extent as they are governed by the causal relation the figures
expressing the phenomena will always deviate from their respective
arithmetical means in a common direction or always in opposite direc-
tions ; the members of every pair of corresponding deviations will in every
case be either both greater or both less than the arithmetical mean of
their respective groups (i.e. always -f x and +y or always -x and -y),
or else in every case one will be greater and the other less (i.e. always ±x
and +y). Therefore the products of which the sum enters into the
numerator will either always be positive or always be negative, and the

MEASURES FOR CONTROL OF TUBERCULOSIS 297

sum total of the products will accordingly be either a positive or a negative
quantity of which the magnitude will depend on the number of terms
to be added. It follows therefore that the more the co-variant terms,
the larger will be the numerator ; and as the whole coefficient can never
exceed +i, the closeness with which its value approaches +i will be a
measure of the closeness with which the phenomena under examination
are connected by cause directly or inversely.

PART III

MEASURES FOR THE REDUCTION AND
ANNIHILATION OFi TUBERCULOSIS

399

CHAPTER XXXVII

GENERAL NATURE OF PREVENTIVE MEASURES :
INDIRECT MEASURES

IN Part I. and Part II. of this volume we have discussed
in full the causation of phthisis, and the factors which have
produced the decline already secured in the death-rate
from this disease. It has been seen that, on the one hand, an
infective agent, the tubercle bacillus, is the essential agent in
causation, and that, on the other hand, various influences other
than infection favour or inhibit the spread of the disease. If
our review of the factors of past decline of phthisis is correct,
the diminution of infection outweighs in importance the diminu-
tion of the conditions favouring infection, though historically
the two have been acting in combination in most countries.
To remove infection most completely we must have the earliest
diagnosis of disease. The early recognition of an infectious
disease is therefore the first step in preventive measures against
it. The cases recognised thus early must then be notified to
those whose duty it is to inaugurate and ensure the execution
of measures against further spread of infection, and to discover
its source in the notified case. This must be some other case
of the same disease, either human or animal ; and the detection
of the source, when practicable, will enable wider measures
to be taken against infection ; while at the same time the
removal or improvement of the conditions, which in the instance
in question have favoured infection, will aid in preventing the
occurrence of further cases. Around the notified case centre
our further preventive measures, which are none the less
preventive in character because they consist largely in the
most effective treatment of the patient himself. Wherever
practicable, the sanatorium treatment of the patient at an
early stage will be secured, with a view to his cure and to his

being trained in the details of the hygienic life which offers

301

302 THE PREVENTION OF TUBERCULOSIS

him the best prospect of recovery and of efficiency after re-
turning home. Should recovery not be secured, the hospital
treatment of the patient, especially if he is poor and cannot
secure good nursing at home, is indicated at a later stage ; and
if he recovers but partially, the conditions for modified work
under favourable conditions need careful consideration. All
these and many allied problems require to be studied, and
some attempt at stating the principles of action is made in the
following chapters. In this chapter we may now consider in
outline the indirect measures against phthisis, which in the
aggregate are very important in its prevention.

INDIRECT MEASURES AGAINST PHTHISIS. — The Teaching of the
Laws of Health. — Of these measures the most important of all
is the inculcation of the laws of health. Hygiene should be
one of the most important subjects in the curriculum of every
scholar in the higher classes of our elementary schools, and
every teacher should be thoroughly competent to teach it.
In a paper read before a Conference of Medical Officers of Health
in 1890, I pointed out that as the entire school population
passed through the higher standards in our elementary schools,
we had here the means of systematically teaching the science of
health to at least six-sevenths of the entire population of the
next generation ; but that for this purpose " it was necessary that
teachers competent to teach the subject should be provided."
The same opinions have been frequently expressed ; and it is
satisfactory to find it stated in a circular issued by the Board
of Education in November 1907 that that Board " are urging
the necessity of giving special instruction in the principles of
hygiene to all students in every type of training college, so that
they may be able to deal profitably with this subject in the
schools." With such teaching in schools and the correlative
practice of school hgyiene, each school will gradually become
an example of the application of the laws of health, and the
homes of the people will quickly benefit also.

Fresh Air and Cleanliness. — In such a scheme of teaching
hygiene the importance of an abundance of fresh air, of strict
cleanliness of person and environment, and particularly of avoid-
ance of dust, will be emphasised ; and thus something will be
done towards securing three great conditions for the prevention of
phthisis. The importance of nasal breathing will also be taught,

GENERAL NATURE OF PREVENTIVE MEASURES 303

as a means of filtering the incoming air, and of preventing the
formation of adenoids, which are a favourite nidus for tubercle
bacilli. If, as appears to be the case, artificial feeding with
the ordinary bottle-teat, and particularly the constant use of
the " dummy-teat," favour the production of adenoids, an
additional reason is furnished for the abolition of the latter
and the encouragement of breast-feeding of babies. The
dangers of dust illustrate the need for having school-drill and all
gymnastic exercises on dustless floors and in an atmosphere
which approximates to that of the external air.

Ill-nutrition and Fatigue. — Defective nutrition may favour
tuberculosis, either by allowing latent foci to come into activity
or by favouring new infection. Over-fatigue is a contributory
influence similar to ill-nutrition, in which the toxic effect of the
products of fatigue replaces the effect of inanition ; and in context
with over-fatigue, it is convenient to group the ordinary occupa-
tional disadvantages which combine with over-fatigue to lower
the inhibitory powers of the workers, favouring catarrhs, and
rousing into activity foci of infection, which may have remained
latent in the bronchial or other lymphatic glands for many
years (pp. 74 and 137). In the poor the two often unhappily
coincide. If food is carefully chosen, even the very poor seldom
suffer from dangerous mal-nutrition ; but if bread and tea
alone take the place of porridge, cheese, herrings, with bread
and other very cheap but highly nutritious foods, mal-nutrition
opens the way to a dangerous extent to invading tubercle
bacilli. Over-fatigue probably causes a much larger number
of attacks of tuberculosis than mal-nutrition, and much of the
excess of pulmonary tuberculosis among men as compared with
women is due probably to this. It is not suggested that there
is not abundant infection in the workshops ; nor that the dust
of workshops is not largely responsible for the result under
consideration. If a reliable test for the limits of physiological
fatigue were applicable, which would eliminate the element of
personality in the testing, and would enable work to be given
in accordance with individual fitness, much avoidable disease
might be prevented. At present we are without any such test,
capable of being used in practical life.

Alcoholism. — Alcoholism, like excessive fatigue, loads the
circulation with toxic matter, diminishes the normal phagocytic

304 THE PREVENTION OF TUBERCULOSIS

action of the body cells, and makes the individual more prone
to every form of infection, and especially to tuberculosis. As
already seen, alcoholic indulgence, when it involves the frequent-
ing of public - houses, implies increased risk of infection by
tuberculosis (pp. 159 and 181 ; and it is scarcely practicable in
most instances of phthisis among the intemperate to distinguish
between the two factors. It is fairly clear, however, that even
among those classes of intemperate persons, who have not been
exposed to convivial infection, an excessive death-rate from
phthisis prevails.

Poverty. — There is no need to reconsider in detail the relation
of poverty to phthisis, as Part II. is largely devoted to this
problem. Poverty and tuberculosis are allied by the closest
bonds, and nothing can be simpler or more certain than the
statement that the removal of poverty would effect an enormous
reduction of the death-rate from tuberculosis. It is, however,
essential in order to secure clear conceptions of causation, to
investigate differentially in various communities the separate
operation of overcrowding, ignorance, mal-nutrition, increased
opportunities for infection, as constituent elements of poverty.
This has been done in pp. 224 to 255, and the preceding remarks
as to the teaching of hygiene, the removal of over-fatigue and
mal-nutrition, the encouragement of alcoholic temperance, and
of cleanliness, represent the practical issue of this investigation.
There remains to be considered the influence of housing.

Housing Conditions. — Although the death-rate from phthisis
is not proportional to the quality of the housing accommodation
in compared communities (pp. 225 and 229), the death-rate from
this disease in any given community is always higher among
those badly than among those more favourably housed (p. 147).
That improved housing is not the main influence determining
the past decline in the death-rate from phthisis is shown by the
evidence given on pp. 227 and 228. This does not imply that
improved housing accommodation is not imperative in the public
interest, but only that such improved accommodation has not
been the predominant influence in causing the decline of the death-
rate from phthisis.

Other things being equal, however, every improvement in
conditions of housing will secure a diminution of tuberculosis.
This applies both to structural and to functional conditions

GENERAL NATURE OF PREVENTIVE MEASURES 305

of housing ; to improvement in respect of light, air, and ventila-
tion ; and to improvement in internal cleanliness of dwelling-
rooms, and diminution of overcrowding. Dwellings to which
light gains free access will always be kept cleaner than dark
and sombre dwellings ; sunlight has a special purifying action
of its own (p. 53) . But even more important than these important
structural conditions is the manner of using the dwelling-rooms.
The structural improvements owe a large share of their import-
ance to the fact that they render internal cleanliness easier, and
its absence more quickly detected. In many houses, unfortun-
ately, bedrooms are overcrowded, while other rooms remain
partially or completely unoccupied. The teaching of the laws
of health, the reduction of the waste of money on alcoholic
drinks, the elevation of the moral standard, must gradually
diminish this variety of overcrowding. As already indicated, the
best means for diminishing the risks of overcrowding is to secure
the institutional treatment of the sick (pp. 149 and 224), especially
of the tuberculous sick. This brings us back to the evil done by
overcrowding in favouring the spread of infection ; in this
chapter we are concerned with its action in lowering the resist-
ance to infection ; and although this must be placed on a lower
platform than the direct effect in spreading infection, every
effort must be made persistently to spread out the sleeping
accommodation of each family over all the rooms available for
this purpose, and to insist on the increase of this accommodation
as required. This latter problem is one of the most difficult
in practical sanitation. To secure its complete solution involves
a wider attack on the problems of poverty, and an increase of
the family income in some instances, and in others a determined
attempt to prevent the waste of the family resources in dis-
sipation and gambling (see also p. 206).

CHAPTER XXXVIII

THE EARLY RECOGNITION OF PHTHISIS IN RELATION
TO ITS PREVENTION

THE NEED FOR BETTER ORGANISATION OF MEDICAL TREAT-
MENT.— For both its successful treatment and the com-
plete prevention of spread of infection, phthisis must
be recognised at an early stage. A very large proportion
of cases, especially those occurring among wage-earners, are
not diagnosed until some such serious symptom as pleurisy or
haemoptysis (spitting of blood) occurs. Even when pleurisy
occurs, this acute disease is often treated without the phthisis
which it commonly indicates being diagnosed. Under the
present conditions of medical treatment immediate improvement
in the expedition with which phthisis is diagnosed cannot be
anticipated. For the working man can seldom afford to leave
his work until actually disabled ; and too often he cannot afford
to pay a doctor's fee for treating a cough, which he may regard
as of comparatively small importance. The provident system
of medical attendance has not been generally successful in this
country, and is not likely to become so in the absence of com-
pulsory membership. Even when adopted, its full benefits
have not been secured, in part owing to the absence of arrange-
ments for consultations, where necessary, with physicians having
special experience in chest ailments. My views on this point,
which has a most important bearing on the prevention of tuber-
culosis, are set forth in the following remarks taken from a
recent address (Sept. 1907).

Doctors have never been doing so much and such good work
on behalf of the public as at present ; but this work is being
done under conditions involving the petty worries of fee-
collecting, the stress of competitive commercialism, the strain
of work which for most doctors is excessive in order to secure

a " living wage," and the " sweating " of the medical profession

306

THE EARLY RECOGNITION OF PHTHISIS 307

by hospitals, friendly societies, and similar organisations.
The doctor earning his livelihood among the artisan and labouring
classes not only has to do excessive work under harassing con-
ditions without leisure, but he is in a large measure cut off from
consultation with doctors having special knowledge in the very
considerable proportion of complicated cases which come under
his care. To the patient in the same classes the conditions
are equally unsatisfactory. However willing he may be to
pay the doctor's fee — which may be as low as is. 6d., or even 6d.
— his limited means necessitate delay in obtaining medical aid
until compelled by urgent symptoms, and necessitate dis-
pensing with this aid at the earliest possible moment. He
realises also the absence of skilled consultation in difficult cases,
and that by attending at a hospital to which his employer has
subscribed, or to which he in his workshop has given his penny
a week, he may have an additional chance of being thoroughly
overhauled, and of securing special skill. Even if the patient
is a member of a club or provident dispensary, similar reflections
apply under the present unco-ordinated conditions, in which
facilities for skilled special consultations are not organised.
Thus, in a large proportion of the total mass of sickness, the
medical welfare of the public is not secured, partly because the
rates of remuneration of club doctors and of doctors attending
the poor are so scanty that only doctors of exceptional mental
and physical capacity can afford time or energy to examine
each patient thoroughly, and partly because medical con-
sultations cannot be secured in difficult cases.

The following are some of the principal respects in which
the present medical service frequently fails : —

i. Diagnosis is belated. This is inevitable for the largest
proportion of the population, under circumstances which involve
payment of a fee or seeking for a hospital letter and then waiting
several hours in an out-patient department. The dangers of
delaying diagnosis are too well known to need detailed con-
sideration. ... In chronic infectious diseases, like phthisis, the
difficulty of obtaining early diagnosis is nearly as great as with
acute infectious diseases, and in non-infectious diseases the
normal condition among the masses of population, especially
those who do not belong to clubs, is to shirk medic'al advice
until it becomes relatively ineffective.

308 THE PREVENTION OF TUBERCULOSIS

2. Treatment is curtailed and its efficiency diminished by
similar considerations of expense.

3. When patients are treated under present circumstances
in dispensaries and in out-patient departments, the waste of time
involves a serious economic loss to the community.

4. There are no co-ordinated arrangements for medical con-
sultations in all difficult cases.

5. Valuable information as to the incidence of disease is wasted
under the present conditions of medical service.

6. There is a great waste of information as to the existence of
conditions conducing to disease, which might promptly be re-
moved under more systematised conditions of medical attend-
ance. At the present time sanitary inspectors and health
visitors are busily engaged in inspecting houses, without medical
knowledge and with only haphazard and very occasional
information of the conditions in the households of the poor,
which the poor-law medical officer, the dispensary doctor, and
the " 6d. doctor," know to be aiding the continuance of disease
and preventing its banishment. The one set of officials, unless
indefinitely multiplied, cannot properly locate the foci of mis-
chief ; while poor-law and dispensary doctors and the doctors
generally among the poor are in possession of information of urgent
importance to the public health; information which, under present
conditions of inco-ordination, is almost entirely lost. Over-
crowding and dampness of the house occupied by a bronchitic
or consumptive patient, the uncleanly and careless nursing of
children, the numerous minor cases of food poisoning, are
examples of conditions of direct importance to the public health ;
and the present system must be regarded as both extravagant
and inefficient, inasmuch as it fails to bring all available informa-
tion concerning such conditions systematically and punctually
to the knowledge of a properly organised system of preventive
medicine. My meaning will be made clearer by giving a practical
instance of co-ordination in further detail. It must be noted
that the co-ordination required in the interests of the public
health is not solely that between all medical practitioners,
preventive and curative, but also between them and such
officials as sanitary inspectors, health visitors, and nurses ; and
the efficiency of co-ordination may be measured by the extent
to which steps taken for the control of a single disease are

THE EARLY RECOGNITION OF PHTHISIS 309

applied without cost to the direct control of general sanitary
conditions.

The experience of Brighton in the notification of pulmonary
tuberculosis is an instance of successful co-ordination of measures
for the treatment and prevention of this disease with those for
the entire public health control of the town. The Public Health
Department of the town is the focus of all the measures — pro-
phylactic, curative, and sanitary — which are taken in the treat-
ment and the prevention of this disease. The officer who visits
the notified case obtains full particulars of the sanitary condition
of the patient's home and secures the necessary disinfection
and sanitary improvements. He obtains information as to the
health of other occupants of the house, and directs them into
the avenues of medical relief, supplying hospital letters when
a private doctor cannot be afforded. He arranges the removal
of the patient to the sanatorium if the doctor considers this
desirable, and there the patient is trained and treated, so that
when discharged there is little risk of his continuing to infect
others. It will be seen that under such an arrangement — an
arrangement which would be improved under a system in which
the doctor himself would to a large extent take the place of
the inspector — one visit serves several ends, and automatically,
and without expense, the information which it affords is dis-
tributed to the departments really concerned. By this co-
ordinated arrangement an economy of time, energy, and money
is secured, which would be impracticable if separate authorities
administered the departments concerned. . . . Hospital reform,
as a measure by itself, would not cure either the grievances
of the public or of the medical profession. Even were all free
dispensaries and all out - patient departments of hospitals
abolished, the willingness and competence of patients to pay
sufficient fees would not thereby be increased, nor would the
ability of the general practitioner to do excessive work for
insufficient pay.

Yet at the present time the coexistent but unco-ordinated
systems have failed lamentably to provide what the health
of the community requires — means for ensuring effectively
the early recognition and proper treatment of all disease. I
hope and believe that what has been done already towards
securing this end is merely a phase in the evolution of the system

3io THE PREVENTION OF TUBERCULOSIS

which will attain it ultimately. The total expense under a
co-ordinated system, worked with due economy, might or might
not be greater than that entailed under the present inefficient
and unco-ordinated system ; and it may be asked whether
the increased cost can be justified economically. The economical
justification, as I have already indicated, will be found in the
decrease of sickness which must follow, with the corresponding
decrease of poverty and inefficiency and invalidity ; in other
words, the economical, like the medical, justification and com-
mendation of a complete medical service consists in its being
a branch of a general service of preventive medicine.

I see no reason to expect that such a medical service, whether
partial or general, would tend to deprave any part of the com-
munity morally, any more than the system of free (that is rate-
paid) education has tended to pauperise the parents of the
children who benefit by it. There would be, I think, no diffi-
culty in proving that each additional form of medical aid
officially given up to the present time, so far from undermining
self-help, has imposed new duties and responsibilities on the
recipients of such help ; while in the aggregate these measures
have been largely instrumental in securing the immense im-
provement in the public health already realised.

Some essential features of the medical service to which
I look forward will be obvious from my previous observations.
At present we have medical officers of health dealing with
sanitation and the prevention of infection, poor-law medical
officers dealing with sickness under the most adverse home
circumstances, school doctors and nurses knowing nothing
or next to nothing of the home conditions which baffle their
work, factory surgeons out of touch with local public health
administration, and a large body of private practitioners daily
in touch with environmental evils that they cannot remove.
The picture which this mere enumeration calls up of work which
overlaps in some directions and leaves serious gaps in other
directions, and which in both instances means an enormous
waste of knowledge of enormous value to the public health,
shows that systematic co-ordination is indispensable to medical
as well as to economical efficiency. The considerations previ-
ously advanced indicate that on all grounds the extended
medical service must be primarily a preventive service. It

THE EARLY RECOGNITION OF PHTHISIS 311

must be a medical service for the general community and not
merely for its sick members, and must call into activity every
individual and collective means for the preservation of health
as well as for the cure of disease. Information of pre-
ventive value must no longer be allowed to run as at present
into culs-de-sac, but must be utilised to the full extent for
the public welfare. This can only be effected when pre-
ventive medicine is regarded as a whole, and the many frag-
mentary portions of it — now unconnected and relatively in-
efficient— are no longer allowed to continue relatively impotent ;
and when every branch of curative medicine is included in its
scope.

THE REMOVAL OF IGNORANCE. — Next in importance to the
removal of all hindrances to early treatment comes teaching
the public the significance of the early symptoms of tuberculosis.
This will doubtless be done in connection with the instruction
in hygiene in the higher classes of elementary and other schools.
Such facts as the following if thoroughly realised would go far
towards annihilating this disease.

1. Consumption is curable, in the majority of instances, if
treated at an early stage.

2. Every cough not yielding to ordinary treatment within
a limited period, indicates the necessity for (a) thoroughly
examining the patient's chest, and (b) examining the patient's
expectoration for tubercle bacilli.

3. Every case of pleurisy must be regarded as likely to be
followed by consumption, failing persistent attention to a
hygienic life.

And there is no reason why this knowledge should not be
impressed upon every boy and girl before leaving school, as
well as upon those who have already left school. At the same
time it should be made plain that scrofulous glands, abscess
of bones, and some deformities of the spine are due to tuber-
culosis.

On the part of doctors practising among the masses of the
population much more needs to be done to ensure the early
recognition of tuberculosis. More time needs to be spent in
ascertaining the antecedents of each patient, his exposures
to infection, and the method of onset of the symptoms from
which he is at present suffering.

312 THE PREVENTION OF TUBERCULOSIS

DIAGNOSIS BY HISTORY. — Symptoms otherwise obscure are
often at once elucidated when an accurate history is obtained
from the patient. The occurrence of languor and lassitude,
of occasional "bad colds" or "bronchitis," of a persistent
cough for some weeks, of indigestion and " anaemia," — one or
more, or all of them at different times — may indicate merely
passing sickness, or may form the early symptoms of phthisis ;
and the significance of these symptoms can often be discovered
by obtaining an accurate domestic and personal history from
the patient.

The diagnosis by history, — aided by such symptoms as the
above, — is in reality a diagnosis of

THE SO-CALLED PR^E-TUBERCULOUS STAGE. — Reference to
the schemes on pp. 64-70 and 75-77 shows that there is
strong reason for believing that in many cases of phthisis
years of primary latency have elapsed between the reception
of the tubercle bacilli with the formation of the first nodule of
disease, and the first recognisable symptom of disease. In
some cases, doubtless, resistance is steadily and increasingly
lowered by the reception of further doses of infective material.
In other cases, active tuberculosis is due probably to the quicken-
ing of the long latent primary foci. This stage of primary
latency cannot correctly be called a prae-tuberculous stage, as
infective nodules are already present ; but it is known under
this name, and in it no clinical evidence of tuberculosis is found.
It is in this stage that the greatest good can be done.

The patient can be suspected of being tuberculous, and
action taken accordingly. Given a complete system of notifica-
tion of phthisis, or a system fairly complete among the classes
whose children attend public elementary schools, it is possible
to pay special attention to the children of notified cases. This
is already done to a considerable extent, but action on these
lines is capable of wide extension. In Brighton the notified
cases, chiefly parents, are removed to the Borough Sanatorium
for a month's treatment and education in the management
of their illness ; and hospital tickets are pressed on any members
of the family who show the least sign of failing health, and who
cannot afford a private doctor. Scholars from such families
should receive special preference in any scheme for providing
country holidays. They are already given special preference

THE EARLY RECOGNITION OF PHTHISIS 313

in the provision of free breakfasts and dinners for the poor in
connection with elementary schools. Extensions of action on
these and allied lines, combined with the more frequent medical
inspection of children from tuberculous families than of other
children, will gradually ensure the early diagnosis and the pre-
ventive treatment of the members of suspected families.

Loss OF WEIGHT. — In persons of tuberculous family history
periodical weighing is one of the best means of ensuring the
early recognition and treatment of disease. The weight should
be taken and recorded at least four times a year — once a month
if there is any reason for anxiety. If along with loss of weight,
or in children failure to increase in weight, the patient's tem-
perature is apt to rise for apparently small reasons, the suspicion
of tuberculosis is increased.

TUBERCULIN TESTING, ETC. — Of means for the early detec-
tion of tuberculosis, other than physical examinations and the
testing of the sputum, the use of tuberculin is the best known.
The value of this test in the detection of bovine tuber-
culosis is well established; though, as Sir J. MacFadyean has
pointed out — (i) an animal may not react for some considerable
period after infection ; (2) a distinct reaction may be unobtain-
able in some advanced cases of tuberculosis ; and (3) in a con-
siderable number of cases a second reaction is not possible for
some days or weeks after the first. It appears therefore that
the reaction when it occurs is trustworthy, but that a negative
result is less reliable. Although there are differences of opinion
on the point, its general use as a means of diagnosis of disease
in man is to be deprecated, in view of the possibility mentioned
by Dr. J. E. Squire that it seemed to him to " cause an increased
activity in the tuberculous focus."

CALMETTE'S OPHTHALMIC METHOD. — A local method of
using tuberculin as a means of diagnosis has been described
recently by Calmette, which may prove to be valuable. He
places inside the eyelid one drop of an aqueous solution of a
precipitate obtained by adding 95 per cent, alcohol to tuber-
culin. If conjunctivitis develops within twenty-four hours,
it is stated to be proof positive that the patient is suffering
from tuberculosis ; no inflammatory reaction seems to occur in
other than tuberculous patients. If more detailed investiga-
tion shows that this method of employing the tuberculin product

314 THE PREVENTION OF TUBERCULOSIS

is harmless and free from fallacy, it promises to be very valuable
in the diagnosis of obscure complaints which may be tuberculous.
If it should lead to the general adoption of an earlier treatment
of tuberculosis than has hitherto been secured, it will be an
immense boon.

OTHER SPECIAL MEANS OF RECOGNITION. — The Rontgen
ray photograph of a chest in which there is an early tuberculous
focus sometimes shows a shadow at the affected part. This is
by no means a certain means of diagnosis, and cases have been
described by Theodore Williams and others in which the physical
signs (by percussion, auscultation, etc.) revealed evidence of
disease not shown by the Rontgen rays. In fact, no special
means of diagnosis will supersede the necessity for

(a) careful physical examination of the patient, and

(b) bacteriological examination of his sputum for tubercle
bacilli.

PHYSICAL EXAMINATION. — In cases in which there is cough
with or without expectoration, in which the patient has repeated
" bad colds," or in which even without these symptoms a patient
with a tuberculous family history suffers from indigestion,
anaemia, or languor, a thorough examination of the chest by a
competent doctor is indicated. Such an examination will
frequently detect the presence of lung disease, either before
there is expectoration or before tubercle bacilli can be found
in it.

The occurrence of jerky breathing or of feeble inspiration is
suspicious. A scattered fine sibilus, often heard only on deep in-
spiration or expiration, was emphasised by Sir William Broadbent
as important. When the physical signs are more marked and
there is dulness and crepitation after coughing, the diagnosis
is relatively easy, and the disease is scarcely at its earliest stage.

EXAMINATION OF SPUTUM. — Very commonly the disease is
first recognised when tubercle bacilli are found in the expectora-
tion. This cannot be regarded as satisfactory, for the occurrence
of expectoration and the presence of tubercle bacilli in it mean
that the encapsulation of the tubercle nodule by the surrounding
tissues has ceased to be effective, and closed has been trans-
formed into open tuberculosis; non-infectious into infectious
disease. For weeks, months, or even years in very slight cases
the tubercle bacilli may not find their way out of the body.

THE EARLY RECOGNITION OF PHTHISIS 315

Thus Allbutt quotes Turban as failing to find tubercle bacilli
in the sputum in the first stage in 59*8 per cent, of 408 cases.

And yet in actual public health experience of the notifica-
tion of phthisis, surprise is frequently expressed by doctors
when sputum sent by them for examination at the public health
laboratory shows tubercle bacilli. It is clear therefore that
the possibilities of early diagnosis of phthisis are not realised
in a notable proportion of cases. It must be added, further-
more, that each year a considerable number of specimens of
thick purulent expectoration are sent for official examination,
from patients who have been treated — usually for bronchitis —
for months before this step towards complete diagnosis is taken.
I append a copy of the form of certificate of results of examination
of sputa which is in use in my own office.

PUBLIC HEALTH OFFICES,
TOWN HALL,

1 90 _

Dear Sir,

I beg to inform you that the specimen of sputum from

has been examined^ and tubercle bacilli were.

Yours faithfully,

Dr.

Medical Officer of Health.

NOTE. — The failure to find the tubercle bacillus does not, of course, prove
that the patient from whom the specimen was taken is not suffering from
pulmonary phthisis.

Tubercle bacilli can sometimes only be found after repeated examinations.
The early morning expectoration should preferably be sent for examination.
patient's address should be given when each specimen is sent.

CHAPTER XXXIX

THE MEDICAL PRACTITIONER IN RELATION TO
PREVENTIVE MEASURES AGAINST PHTHISIS1

THE PATIENT MUST NOT BE KEPT IN IGNORANCE. — When
the presence of phthisis has been ascertained, the first
duty of the doctor is to inform his patient. Anxious
relatives will occasionally urge him not to do so, but the cases
in which he is justified in withholding the information in my
opinion are few ; and both relatives and the patient can with
intelligent explanation be made to understand that it is in the
latter's interest to secure intelligent co-operation between him
and the doctor. Phthisis is an eminently curable disease. Its
cure is hastened and rendered more certain if the patient is
convinced of the necessity for and the wisdom of adopt-
ing the prescribed measures, — both the treatment in the
more limited sense of the word, and the treatment which
consists in care as to sputum, thus diminishing the danger of
re-infection.

WHAT DANGER is THERE OF INFECTION IN PHTHISIS ?—
The relative infrequency of infection of hospital nurses by tuber-
culosis is important from the medical practitioner's standpoint,
as a study of it supplies him with the main indications for safe-
guarding the health of the relatives and attendants of his own
consumptive patients. He is already aware that the channels
of infection are limited. The following scheme sets forth the
main dangers. This scheme does not pretend to be logical or
exhaustive, but it serves to draw attention to some of the
more important points : —

1 A large part of this chapter has already been published in an Introductory
Address given by the author at the Mount Vernon Hospital for Consumption,
on " The Relation of the Medical Practitioner to Preventive Measures against

Tuberculosis," Lancet, January 30, 1904, p. 282.

316

THE MEDICAL PRACTITIONER

317

(1. Dose.

I 2. Cumulative dosage.
I. The infection. -(3. Closeness of contact.

I %)*

I 4. Lack or absence of precautions.

Defective ventilation and cleansing of rooms.

II. Receptivity.

Inherited.
Acquired.

1. Exhaustion from nursing, etc.

2. Depressing emotions.

3. Insufficient nutrition.

4. Defective ventilation and cleansing of rooms.

In hospitals, long before the communicability of phthisis
was recognised, expectoration was received into spittoons and
large dosage of infection was thus prevented. Similarly hospital
wards have usually been well ventilated and kept scrupulously
clean, all surfaces both of walls and floors being washable.
Again, hospital nurses are not so long on duty as wives or other
relatives, the contact between them and the patient is less inti-
mate as well as less prolonged than that of home nurses, they
have periodical holidays, are well fed, and are not subjected to
the same extent to the influence of depressing emotions or of
insanitary house conditions. They are better trained in regard
to the washing of hands and other personal precautions. In
view of the above circumstances, the difference between the
infectivity characterising phthisis in hospital and in private
practice is easily understood.

I can imagine no better means of converting those who under-
rate the infectivity of tuberculosis than the task of administering
the notification of this disease in a large town, of interviewing
some 300 patients each year, of examining over 200 patients
who are yearly treated for a month or more each in a borough
sanatorium with a view to train them so as to diminish the proba-
bility of their continuing sources of infection, of obtaining the
family and personal histories of each of these, and tracing, as
one gradually comes to do, links of infection, which, although
individually they may not be conclusive, when connected to-
gether become as convincing as any evidence can ever be regard-
ing a communicable disease of chronic course.

DUTY OF THE DOCTOR TO THE PATIENT AND TO THE PATIENT'S
FAMILY. — The first duty of the family practitioner in relation
to a case of phthisis obviously is to do his best for the patient.
Incidentally his position by implication involves that he is,
at least partially, the guardian of the health of the 'patient's
family. Happily, the interests of both patient and relatives

318 THE PREVENTION OF TUBERCULOSIS

are identical, and the measures most conducive to the patient's
recovery will also give the maximum protection to the other
occupants of the same house.

Having (i) made an early diagnosis of the disease, and (2)
acquainted the patient and his relatives with the nature of the
disease, the further indications for the doctor are : (3) to investi-
gate and, if possible, ascertain the most likely source of the
patient's infection ; (4) to treat the patient (under this head
will come not only dietetic and medicinal treatment, but the
question of sanatorium treatment and the control of the general
hygiene of the patient) ; (5) to train the patient to control his
cough, as far as practicable to cough and to expectorate only when
means are available for preventing the dissemination of in-
fective matter, to train him to live in the open air, to eat heartily,
and to attend to every detail of personal hygiene ; and (6) to
protect the attendants on the patient from infection, from
over-fatigue, from impaired nutrition, carefully training them
on the same lines as the patient himself, whose recovery depends
largely on the state of their health.

INVESTIGATION OF SOURCES OF INFECTION. — The investiga-
tion of possible sources of infection may appear to be somewhat
remote from the duties of the family practitioner, and yet
success in the treatment of the patient may be wrapped up in
the fulfilment of this indication. The three most common sources
of infection are : (i) domestic, (2) occupational, and (3) public-
houses. So far as domestic infection is concerned, in well-to-do
families the medical adviser will have the opportunity of in-
vestigating possible unrecognised sources of infection in the
same household. In poorer houses this is not so. The patient
is treated as a club patient or at the dispensary or hospital.
Domestic sources of infection cannot then be recognised by the
medical attendant. Even if he sees the patient at home he
has no time to investigate the case fully. It has been my
frequent lot in visiting phthisical homes to find other unre-
cognised patients suffering from chronic tuberculous disease
and innocently spreading more acute tuberculous disease to
husband or wife or children.

If infection can be shown with some degree of probability
to have been acquired in a dusty workshop or shop, an indica-
tion for treatment is at once obtained. Even if the occupation

THE MEDICAL PRACTITIONER 319

cannot be altered, the conditions of the workshop may be favour-
ably changed, and if the medical officer of health and the
practitioner come into touch at this point the conditions of the
workshop can be improved and the patient's chances of recovery
increased without the slightest risk to the patient's pecuniary
welfare. At this point, however, we trench on the question of
notification of the case to the medical officer of health, and the
action which would follow such notification (p. 338).

If the patient is alcoholic, to insist on a change in his
habits in this respect, given that the patient's confidence
can be secured and that he is open to conviction, is the
best means not only of preparing him intelligently to carry out
his instructions and of enabling him to recover the resist-
ance to disease which has been lowered by alcoholic indul-
gence, but also of stopping those visits to the public-house
which, as Dr. J. Niven has indicated, are a frequent means of
infection.

RELATIVE MAGNITUDE OF THE RISKS OF EXTERNAL AND
AUTO-INFECTION. — It may be urged that once phthisis is started
its subsequent course is determined not by external but by
internal infection, and that consequently the detection of the
sources of infection or even of other cases of phthisis in the
same house is not important from the private practitioner's
standpoint. This point is one of real importance. In the
card of precautionary instructions, of which a copy is given
on p. 324, the following sentence occurs : " The patient himself
is the greatest gainer by the above precautions, as his recovery
is retarded and frequently prevented by renewed infection
derived from his own expectoration."

Is the prevention of auto-infection by expectoration, which
has been already ejected from the mouth, important ? It is
well known that tubercle travels from one part of the body
to another by the lymphatics or blood vessels. It is also agreed
that healthy persons are infected chiefly by inhalation or inges-
tion of infective dust or by direct infection by minute particles
of ejected sputum. The patient is perhaps not likely to be
re-infected directly by the spray of his own sputum, but may
if this becomes dry ; and he may receive more massive re-in-
fection if no precautions are taken to prevent the inhalation,
as dust, of desiccated sputum, or the swallowing of his own

320 THE PREVENTION OF TUBERCULOSIS

sputum. I am unaware of any exact facts as to whether such
re-infection is an important factor in the downward progress of
the consumptive when considered in comparison with the auto-
infection caused by the cross-inhalation of infective mucus
into other bronchioles than those first affected ; but whether
the danger be greater or less, the swallowing of sputum should be
prohibited, and experience shows that the improvement of the
consumptive is greatest in those cases in which there is the most
rigid care to prevent re-infection by dust, whether because
in this way re-infection by the tubercle bacillus or because
secondary infection by other micro-organisms is prevented.
I attach much importance to the value of these precautions in
preventing danger to others than the patient. Self-interest is
a potent motive for beneficence.

THE EFFECT OF SWALLOWED TUBERCULOUS EXPECTORATION.
— The occurrence of self -re-infection by swallowing expectora-
tion is well established. Various statistics give the proportion
of cases in which intestinal ulcers are found after death from
phthisis, as from one-fourth to three-fourths or more of the
total cases. The coincidence between tuberculosis of lungs
and intestines might be due to the intestinal ulcer having been
the primary seat of disease ; but that this is not the correct
explanation is indicated by the fact that intestinal ulceration
is a late phenomenon in phthisis. The intestinal disease must
therefore in most instances be due to spread of tuberculosis
from other parts of the body, or to the swallowing of large quan-
tities of tuberculous expectoration. That the last is most
usually the explanation is shown by the fact that intestinal
ulcers are much more rarely found where the lung is not impli-
cated, and very rarely in general tuberculosis. Experimental
observations point to the same conclusion. Cornet records
that out of over 3000 animals on whom he experimented other-
wise than by feeding, only in about eight cases were tuber-
culous foci found in the intestine and in isolated mesenteric
glands. The extreme frequency of intestinal ulceration in young
children and in the insane, who nearly always swallow their
expectoration, points to the same conclusion.

On the other hand, instances occur in which prolonged
swallowing is not followed by intestinal ulceration. It is likely,
also, that in a certain number of instances of such ulceration

THE MEDICAL PRACTITIONER 321

infection has been received from the blood current, and not by
the direct contact of tuberculous expectoration.

The evidence points clearly to the importance of the doctor
warning his patient against swallowing his sputum. Some
French physicians have gone so far as to advise washing out
the mouth with a mild antiseptic after each attack of coughing ;
but this does not appear to be necessary or likely to be carried
out even if recommended.

THE DOCTOR IN RELATION TO DISINFECTION. — Assuming that
a doctor is called in to a case of phthisis, and that up to that
time no precautionary measures have been taken, his duty is
not fulfilled by insisting on the adoption of all the measures
enjoined in such a set of " precautionary instructions " as those
given on p. 324. Infection has been repeatedly shown to cling
to the lower part of the wall and to the floor of the consump-
tive's room. It also hangs about his pockets, bed-hangings,
etc. If the doctor is to do the best for his patient he must rid
him of old infective material. And he cannot in the majority
of instances do this alone. He must in the interest of his patient
call in the aid of the medical officer of health, who can arrange
for efficient disinfection of the room and its belongings. Then,
with a rigid system of cleanliness, re-infection of the room and
repetition of danger from this source to patient and relatives
can be greatly diminished.

THE DOCTOR IN RELATION TO NOTIFICATION. — Such an
intimation of desire for disinfection is almost tantamount to
a voluntary notification of the case to the medical officer of
health ; and this voluntary notification can in the case of
private patients be made only with the consent of the patient
or his guardians. There are other reasons why such a volun-
tary notification is desirable.

1. The medical officer of health will probably be in a better
position than the practitioner to detect the possible source
of infection and thus to minimise any likelihood of continuance
of infection when the patient resumes his occupation, etc.

2. The medical officer of health can not only enable the
patient to " start fair/' as indicated above, but he can do much
to remove any insanitary conditions of home, workshop, or
shop tending to retard recovery. It may be urged that -sanitary
authorities already have the power to abate overcrowding

322 THE PREVENTION OF TUBERCULOSIS

and to insist on the cleansing and ventilation of houses, work-
shops, etc. But sanitary officials are neither omniscient nor
omnipresent, and their work is most productive of good when
directed especially to houses in which the presence of a case
of phthisis renders overcrowding, uncleanliness, and other
insanitary conditions supremely dangerous. Without an army
of inspectors it is impossible completely to control overcrowding
and dirtiness of houses, and the notification of this disease
gives valuable additional leverage in securing the abolition of
minor insanitary conditions, the continuance of which is detri-
mental to the consumptive.

3. The most conscientious and indefatigable doctor can
usually only ensure the carrying out of a portion of the measures
which I have ventured to bring within the range of his legiti-
mate duties. He may do so if his patient is wealthy and intelli-
gent. He certainly cannot if his patient belongs to the working
classes, who contribute the vast majority of the cases of phthisis.
Between these two extremes are patients in whose behalf a
varying degree of intervention on the part of the local authority
is required. There is no wish on the part of such authorities
or their officers to interfere, but only to help. If proper steps
for preventing indiscriminate expectoration, for destroying any
infective material already deposited by the patient, and for
tracing possible connections with other cases of phthisis, have
been taken, the less the intervention of any one between the
medical man and his patient the better. But in actual practice
most phthisical patients have medical men in attendance only
at intervals, and for a short portion of their total illness. Visits
of an educational character are certainly needed in the intervals
of professional attendance, if not also while the latter is in opera-
tion. In actual experience in Brighton, although a considerable
number of cases of phthisis have been notified in private as well
as in dispensary and hospital practice, no appreciable friction
has been caused by my visit or those of my assistants, and a
large amount of carelessness as to the disposal of sputum has
been thus stopped.

THE DOCTOR IN RELATION TO SANATORIUM TREATMENT.—
A further duty to his consumptive patient devolves on the
family practitioner. He has to decide whether he can secure
for his patient the best medical and hygienic treatment at home,

THE MEDICAL PRACTITIONER

323

or whether a temporary stay in a well-organised sanatorium is
needed. These points are more fully discussed in Chapter XL.
As a rule, it may be said that both educationally and thera-
peutically the patient is benefited, and his relatives are freer
from danger of infection if such a course of sanatorium treat-
ment and teaching has been secured.

In the preceding remarks the ideal position of the medical
practitioner in relation to tuberculosis has been indicated.
Therapeutical measures are in the widest sense measures of
prophylaxis, and the aid of measures of public and private
hygiene is as indispensable to cure as are therapeutical measures.
But the doctor in the majority of cases — i.e. those of the working
classes — can scarcely be said to be the " family " doctor. Even
in the higher social strata his efforts at prophylaxis may be
hampered by prudential and other considerations, and he cannot
undertake those wider inquiries which are required in order
most completely to stop the sources of infection. Clearly, then,
everything indicates the necessity of co-operation between
doctor and medical officer of health, and the more complete
this co-operation the greater is the benefit to the consumptive
patient and to every member of the public.

CHAPTER XL

THE CONSUMPTIVE PATIENT IN RELATION TO
PREVENTIVE MEASURES AGAINST PHTHISIS

ASSUMING that the patient has consulted a doctor who is
imbued with the ideal view of his duties suggested in
the last chapter, the duty of the patient is clear, though
it necessitates a steady persistence in well-doing, which implies
moral courage and perseverance as well as intelligent accept-
ance of the duties involved.

The patient will have handed to him a set of instructions,
of which the following may be taken as an example. They
will be amplified and explained more fully by the doctor. It
may be added that in Brighton these cards are printed by the
Corporation without any official headings or names, in order
that every doctor may distribute them to his own patients.
The instructions are as follows : —

PRECAUTIONS FOR CONSUMPTIVE PERSONS

Consumption is, to a limited extent, an infectious disease.
It is spread chiefly by inhaling the expectoration (spit) of
patients which has been allowed to become dry and float about
the room as dust, or by directly inhaling the spray which may
be produced when a patient coughs.

Do not spit except into receptacles, the contents of which are to
be destroyed before they become dry. If this simple precaution
is taken, there is practically no danger of infection. The breath
of consumptive persons is free from infection, except when
coughing.

The following detailed rules will be found useful, both to
the consumptive and to his friends : —

I. Expectoration indoors should be received into small
paper bags and burnt immediately ; or into a receptacle which is

emptied down the drain daily and then washed with boiling water.

324

THE CONSUMPTIVE PATIENT 325

2. Expectoration out of doors should be received into a
suitable bottle, to be afterwards washed out with boiling water.
If a paper handkerchief is used, this must at once be placed
in a waterproof bag, the contents subsequently burnt and the
bag washed daily.

3. Ordinary handkerchiefs, if ever used for expectoration,
should be put into boiling water before they have time to become
dry ; or into a solution of a disinfectant, as directed by the
doctor.

4. Wet cleansing of rooms, particularly of bedrooms occupied
by sick persons, should be substituted for " dusting " and
"sweeping."

5. Sunlight and fresh air are the greatest enemies of in-
fection. Every patient should sleep with his bedroom window
open top and bottom, a screen being arranged, if necessary, to
prevent direct draught.

6. The patient should, whenever practicable, occupy a
separate bedroom. Children should never sleep in the same
bedroom as the patient.

N.B. — The patient himself is the greatest gainer by the above
precautions, as his recovery is retarded and frequently pre-
vented by renewed infection derived from his own expectoration.

7. Persons in good health have little reason to fear the
infection of consumption. Over- fatigue, intemperance, bad air,
dusty occupations, and dirty rooms favour consumption.

CURE AND PREVENTION ARE INSEPARABLE. — The first point
needing to be grasped by the patient thoroughly is that
measures for the cure of and measures for the prevention of
consumption are to a large extent identical. Certain drugs
have their value in treating consumption ; cod-liver oil is equally
valuable in treating it and in preventing its development ;
most other remedial measures used in the treatment of
consumption would be still more effective if employed in
preventing it.

The essential points in the treatment of consumption are—

(1) the prevention of further infection ;

(2) the prevention of the inhalation of dust of any kind ;

(3) the improvement of nutrition of the patient ;

(4) regulated rest until the disease has become -entirely
quiescent.

326

THE PREVENTION OF TUBERCULOSIS

The first of the above points has been discussed on p. 319.
The patient, as well as those about him, gains by observance of
the precautionary measures as to coughing and the disposal of
sputum. By avoiding the swallowing of sputum, he also
minimises the chance of secondary intestinal infection.

The prevention of the inhalation of dust is an essential point
in the treatment as in the prevention of consumption. It has
been already seen that this disease is most prevalent among
those engaged in dusty occupations ; and one of the great gains
in sanatorium treatment is that the patient breathes a relatively
dustless and aseptic atmosphere.

Similarly with regard to mal-nutrition and over-fatigue,
the probability of recovery from consumption and of successful
resistance to its infection, other things being equal, are both
increased by diminishing or removing their operation.

HOME TREATMENT. — These points being settled, we may
consider in detail the part which the patient has to play in
curing his disease and preventing its spread. In this chapter
the matter will be considered from the standpoint of the treat-
ment of the disease at home. The following are the main
points : —

(1) There must be no spitting into handkerchiefs, nor should
handkerchiefs with which the mouth has been wiped be placed
under the pillow. The exact details as to the disposal of sputum
are given in Chapter XLI.

(2) If linen handkerchiefs are used at all, they must not
be allowed to get dry after being used, but placed in water to
which some washing-soda has been added. It is best, however,
to use paper handkerchiefs or rags which can be burned.

(3) During coughing the patient must always hold something
in front of his mouth.

(4) A fire in the bedroom always helps ventilation, and is
useful for burning rags, etc.

(5) Cups, knives, spoons, etc., must be placed in boiling
water containing some washing-soda before being again used.

(6) There is no need to sprinkle the floor of the room with
disinfectants. Washing with soap and water suffices.

(7) The floor should be uncarpeted except for a rug at the
bedside. The best plan is to have the floor covered with
linoleum, washing this daily. The floor should never be dry-

THE CONSUMPTIVE PATIENT 327

swept. All articles not washable should be wiped with a damp
duster. Curtains and other hangings are best discarded.

(8) The walls should be periodically cleansed, especially
the part between the floor-level and about a yard above the
level of the bed. Four methods of cleansing and disinfection
are commonly adopted ; the help of the officials of the Sanitary
Authority can be obtained in carrying out one of these :

(a) The wall-paper if dirty should be stripped off and burnt.

(b) A solution of chlorinated soda may be brushed on the

walls.

(d) The German method of rubbing down the wall with bread-

crumbs, and then burning the crumbs, may be adopted.

(9) The patient's room should be carefully chosen, so as
to be convenient for nursing, and to enable the patient to get
into the garden whenever practicable.

(10) The ventilation of the room should be specially studied.
As a rule, the window and the door should both be kept wide
open, and generally — by means of screens or otherwise — this
can be arranged without leaving the patient in a disagreeable
current of air. If the bedroom has two windows, there is no
difficulty in securing the perflation of air which is desirable.
The question of open doors and windows must be decided in
each case according to circumstances. Gradually the amount
of fresh air should be increased ; and a sanatorium-treated
patient will seldom wish to go back to the imperfect ventilation
which passes muster in most households. On the other hand,
nothing is gained by increasing the discomfort of a dying
patient.

(n) The thoughtful patient will save his nurse as much
trouble as possible. She must have a sufficiency of sleep,
exercise, and rest, and must not take her meals in the bedroom.
The patient must further protect her by always placing a hand-
kerchief in front of his face when coughing.

THE PATIENT'S OCCUPATION. — The preceding scheme of
action is concerned chiefly with the patient's home-life. It
has to be borne in mind, however, that during a large part of
his illness he is still following his occupation. Commonly,
if a wage-earner, he has drifted from the more to- the less
laborious occupations, and from the ranks of the steady

328 THE PREVENTION OF TUBERCULOSIS

wage-earners to the ranks of the casual workers. But in a
large proportion of cases, the patient for a year, or even for
many years, keeps at his work in the factory, workshop, shop,
or office. As a rule, it is better that he should do so, than that
in consequence of vague advice " to get a lighter job in the
open air " he should drift into a condition of unemployment,
he and his family suffering in consequence from ill-nutrition.
If there is a definite prospect of more suitable work, it should
be taken ; but it is of little use, for instance, to advise a clerk
to become a farm labourer or even a market gardener, unless
he is unusually strong and the disease is very early.

Assuming that the patient must keep to his present indoor
occupation, what advice should be given ? It should first of
all be urged upon him to come into a sanatorium for a month
to receive the short course of treatment and teaching which
is described on p. 349. If he continues his occupation after
a month thus well spent, he is much more likely to do so without
danger to others, and with a prospect by careful living of pro-
longed work, than would otherwise have been possible for him.

The further advice needed consists chiefly in the avoidance
of over-fatigue and of the inhalation of dust, and the proper
use of his spit-bottle. This can be used judiciously, so as not
to attract attention. In his home-life the ex-patient has the
opportunity of counteracting to a large extent the influence
of an unfavourable occupation. He can sleep in the open air,
take judicious rest, and in other ways, so far as his means permit,
follow the regime, the principles of which he has learnt while in
a sanatorium.

THE PATIENT IN RELATION TO THE SANITARY AUTHORITY.—
If compulsory notification of phthisis is in force in the town in
which the patient lives, the doctor in attendance is required to
notify the patient's illness to the medical officer of health. If
such notification is invited under a voluntary system, the patient
has it within his choice to prevent such notification. By so
doing he will be acting unwisely in his own as well as in the
public interest. This somewhat bold statement needs perhaps
elaboration and proof, which it is not difficult to supply. In
the first place, it can be made clear that the patient will suffer
no disability by having his case notified. Thus the statement
that " as soon as they made known that a man was a victim

THE CONSUMPTIVE PATIENT

329

to the disease they advertised him as a dangerous person, and
the public would continue to believe that," ignores the fact
that notifications are confidential, that the information does
not pass beyond the householder, that so long as the patient
takes reasonable precautions as to his sputum, there is no inter-
ference with his home-life or his occupation.

In the absence of grave mal-administration the notion that
notification will involve any interference with a man's occupa-
tion may be banished as unfounded. At the same time, it is
true that, quite irrespective of notification, the public have
become much more alive to the possibilities of infection in
phthisis, and have oft times taken exaggerated action concerning
it. The best means for reducing such fears to their proper
magnitude is to be able to reassure the public that every case
of phthisis is notified and the proper precautions have been
taken.

Secondly, the patient himself benefits from notification so
far as both his domestic and industrial circumstances are con-
cerned, (a) Domestically the patient has offered to him any
disinfection that may be required in the interest of himself and
his family. For the poor, sputum bottles and paper hand-
kerchiefs are supplied. Under a well-organised system of
notification, sanatorium treatment is offered (see p. 347). If any
sanitary defects are found in the house, these are remedied.
Damp walls, unventilated staircases, windows that do not open
top and bottom, all militate against the patient's recovery, and
may be remedied as the result of an official visit.

(b) Industrially the patient only benefits indirectly. No
visits to patients are made at workshops or shops, in any town
with the administration of which I am acquainted. To make
such visits would be a foolish mistake. But, quite apart from
the patient himself, workplaces are visited, and defects dis-
covered and remedied, the remedy of which might otherwise
have been greatly delayed. No Sanitary Authority possesses
a sufficiently large staff immediately to discover all sanitary
defects. Very few Sanitary Authorities have a staff of sanitary
inspectors sufficiently large to enable them to visit each house
and workplace in their district once annually. In the intervals
of such visits conditions of overcrowding, dirtiness, and dustiness
may long prevail. These conditions are much more dangerous

330 THE PREVENTION OF TUBERCULOSIS

where there is a case of phthisis than elsewhere. The notifica-
tion of cases of this disease enables houses and workplaces in
which such visits are particularly important to be visited at
more frequent intervals, a great gain to the public health being
thus secured.

Thirdly, the patient by allowing his case to be notified is
contributing to the general health of the community. The
notification of his case may lead not only to the removal of
insanitary conditions favouring the spread of disease, but also
to the discovery of other untreated cases in the same household ;
and by comparison with the official records may lead to the
discovery of particular workplaces or of particular areas of a
town in which phthisis is exceptionally rife.

CHAPTER XLI
THE PREVENTION OF INDISCRIMINATE EXPECTORATION

THE proper control of spitting and disposal of the sputum are
probably the chief problems in the prevention of phthisis.
They therefore deserve a special chapter, and by this
means repetition of instructions can be avoided in other chapters.
The closely allied question of instructions for coughing with
proper safeguards is considered on p. 326.

As already seen, consumptive patients may discharge billions
of tubercle bacilli daily in their expectoration (p. 104). This
may be dangerous immediately while being scattered as fine
spray ; or after having become dried and pulverised, it may
be subsequently suspended in the air and inhaled.

Indiscriminate spitting is much less dangerous in open places,
for instance in a road, than in houses, public-houses, or other
places of public resort, especially if these are dark and over-
crowded. Dr. H. E. Annett (1902) collected by means of
sterilised swabs 105 specimens of sputum deposited in the
streets of Liverpool. Five of these were proved to contain
virulent tubercle bacilli. Apart, however, from such actual
deposits of expectoration, it is fairly certain that tubercle
bacilli can seldom be found in the dust of streets in places
protected from direct expectoration. The explanation of this is
not far to seek. Notwithstanding the large amount of indis-
criminate expectoration in streets, many factors tend to cause
tubercle bacilli to perish within a limited period. When ^xposed
in thin layers, direct sunlight kills them in a few minutes or
hours and diffuse light in a few days. The cleansing of streets by
rain or by road watering must have a very beneficial effect, both
in washing the bacilli into the sewers and in preventing their
dissemination as dust. At the same time expectoration in streets
is an undoubted source of danger, especially when this expectora-
tion is carried home on the skirts of ladies' dresses or on boots, etc.

331

332 THE PREVENTION OF TUBERCULOSIS

How should the consumptive patient dispose of his sputum
indoors and out of doors ?

INDOOR DISPOSAL OF SPUTUM. — The problem indoors is
easily solved. A special spit-cup must be kept for the patient.
If the amount of expectoration is not very great, it is a good
plan to line this spit-cup with butter-paper, and then the
daily expectoration can be easily emptied down a water-closet
or slop-closet into the drain. A disinfectant is unnecessary in
the spit-cup under ordinary circumstances ; but care must be
exercised to ensure that the outsides of the cup are not fouled,
and that flies are not allowed access to it. The spit-cup after
being emptied should be washed out in boiling water containing
some washing-soda, and subsequently washed again, before being
used. If the expectoration is abundant and adheres to the
sides of the spit-cup, it is convenient to render it less adhesive,
and aid its removal from the spit-cup, by adding some soapy
disinfectant to it before emptying it down the drain. If there
is no water-closet system, the sputum should be burned, or if
this is impracticable it should be boiled. In a sanatorium the
spit-cups should be cleaned and sterilised with boiling soda
solution, which may be done in a special apparatus heated
by coal, gas, or steam. In this way the cleansing is effected
with less trouble, and sterilisation is rendered certain. Floor-
spittoons should never be tolerated. After expectoration, the
patient's mouth is frequently soiled, and a paper handker-
chief should be employed in wiping it. This should be at once
burnt, or if this is impracticable it should be placed in the spit-
cup. Japanese handkerchiefs suitable for this purpose are pur-
chased by the Brighton Corporation at 55. a thousand. These
measure 14 inches square, and are cut into four before distribu-
tion. The patient should also be carefully trained to hold one
of these handkerchiefs in front of the mouth while coughing.

OUTDOOR DISPOSAL OF SPUTUM. — A pocket spit-bottle is
required for outdoor use. A very good and simple form consists
of a wide-mouthed bottle, with a thick rubber stopper. It is
easily cleansed, not easily broken, and of a convenient size
for the pocket. Such spit-bottles can be obtained at 4d. to 5d.
each when a gross are bought ; and both they and the Japanese
handkerchiefs mentioned above are suitable for gratuitous dis-
tribution in public health administration. The spit-bottle

INDISCRIMINATE EXPECTORATION 333

can be cleansed thoroughly with boiling water containing some
washing-soda.

It is well to carry the pocket spit-bottle in an indiarubber
pouch or in a pocket having a detachable washable lining ;
and a similar bag should be used for soiled paper handkerchiefs.

THE DISPOSAL OF SPUTUM OF PATIENTS WITH ADVANCED
DISEASE. — It is generally recognised that the danger of infection
is greatest in advanced cases of phthisis. Objection has been
taken to this view, because the sputum of early cases often
contains multitudes of tubercle bacilli. Several points, however,
need to be borne in mind : (a) Patients with early disease spend
a large part of their day away from home, and much of the
sputum they expectorate is deposited in the open, (b) Ex-
pectoration at this stage is much smaller in amount than at
later stages, (c) The patient is not enfeebled by prolonged
illness, and he still has the courage and strength to avoid fouling
his handkerchief or his bed and body linen. There is a further
reason why the sputum of advanced cases of disease is to be
particularly feared when they are treated at home. The wife
or other attendant is exhausted by prolonged nursing, and
depressed by anxiety and sorrow, and is consequently much
more liable to be open to infection than at an earlier
period.

For these reasons a special importance attaches to the manage-
ment of the sputum of patients with advanced disease.

Bedridden patients should never be allowed to keep a hand-
kerchief under the pillow or in the bed. It should always be
placed in a cleansable receptacle outside the bed. The patient's
mouth must be covered with a paper handkerchief or rag while
coughing, the mouth wiped with the same paper or rag after
coughing, and the material where practicable at once burnt.
The attendant's hands should be washed after performing these
duties.

PUBLIC REGULATIONS AS TO SPITTING. — In recent years great
advances have been made in the control of indiscriminate ex-
pectoration. In this country the Glamorgan County Council
was the first to obtain the consent of the Secretary of State
for the Home Department to a bye-law regulating spitting in
public places. As originally drafted, the bye-law ran as
follows :^~

334 THE PREVENTION OF TUBERCULOSIS

A person shall not spit on the floor of any public carriage, or of any
church, chapel, public hall, waiting-room, schoolroom, theatre, or shop,
whether admission thereto be obtained upon payment or not.

Any person offending against this bye-law shall be liable to a fine not
exceeding ^5.

The Home Office subsequently decided that the bye-law
could not properly be made to apply to churches, chapels,
schools, and shops, and the bye-law being amended in accordance
with this decision came into operation. A considerable number
of other Local Authorities have now adopted the same bye-law,
the one commonly in force running as follows : —

No person shall spit on the floor, side, or wall of any public carriage,
or of any public hall, public waiting-room, or place of public entertain-
ment, whether admission thereto be obtained upon payment or not.

Any person who shall offend against this bye-law shall be liable for
each offence to a fine not exceeding forty shillings.

Local Authorities owning tramways have also passed bye-
laws forbidding expectoration in them, and prosecutions of
persons offending against such bye-laws have been successful.

THE PREVENTION OF SPITTING IN PUBLIC - HOUSES, ETC.—
In my local experience no difficulty has been experienced in
securing the fixing on the walls of every bar of each public-house
in the town of an enamelled iron tablet, size 6f x 4f inches,
having the following words on it : —

PREVENTION OF CONSUMPTION

YOU ARE

EARNESTLY REQUESTED
TO ABSTAIN FROM THE
DANGEROUS HABIT OF

SPITTING

The following correspondence took place before the tablets
were exhibited, and it is reproduced here, as it may be useful to
others : —

INDISCRIMINATE EXPECTORATION 335

To the Sec., Licensed Victuallers' Association.
,, Beer Sellers' Association.

,, Brewers' Association.

DEAR SIR, — I enclose herewith a draft of a circular letter which it is
proposed to send to each publican in the town.

It deals with a very important question, the importance of which with
regard to the public health is becoming more and more realised.

The likelihood of securing compliance with the suggestions made in
this circular letter would be greatly increased by your co-operation.
Would it not be practicable for you to bring the question before your
Association at their next meeting, with a recommendation that individual
members of the Association should help in bringing about this desirable
reform ?

If you have any suggestions to make as to improving the draft circular,
1 should be glad to receive them and to give them every consideration. —
Yours faithfully,

MEDICAL OFFICER OF HEALTH

To the Proprietor or Tenant of

Inn or Hotel.

DEAR SIR, — You will probably have learnt from the public press that
it is now generally realised that consumption, which is the most fatal of all
the infectious diseases, is spread by inhaling the dried spit or expectoration
of patients suffering from this disease. It may not be so well known to you
that the mortality from consumption among those engaged in public-
houses is much heavier than that of the general public. Our national
statistics show that if the deaths from consumption for the average of all
men aged 25 to 65 engaged in various occupations be represented by 100,
that of innkeepers and brewers is 140 to 148, and of male inn servants is

257.

This excess is doubtless due to the conditions to which those engaged
in public-houses are exposed, among the chief of which is the frequent
inhalation of dust derived from the expectoration of consumptives. This
danger is greatly favoured by (a) the practice of indiscriminate spitting
in the bars of public-houses, and (&) the common practice of allowing
such spitting on the floor, sawdust being frequently provided for the
purpose of receiving it. If expectoration on the floor is to be permitted,
the spit should be washed up by means of a mop several times a day,
before it has had time to become dry. Sweeping up of sawdust containing
it is one of the surest methods of distributing a very dangerous infection
to others as well as to the sweeper. The spit or expectoration is not a source
of danger (unless directly inhaled when a patient is coughing) in the wet
condition. Efforts should be therefore directed towards either causing it
to be immediately burnt in the fire, or, failing this, kept in a moist condition
until it can be destroyed.

It may be further remarked that expectoration indoors is very much
more dangerous than expectoration out of doors. In the latter case its

336 THE PREVENTION OF TUBERCULOSIS

infectious properties are soon destroyed by sunlight. Hence, customers
may fairly be asked to reserve their spitting for out of doors.

It is suggested that the accompanying tablet should be put up in the
bar. Further supplies, which it is hoped will be displayed in every public
room, may be obtained as desired. It is also strongly urged that no
sawdust should be used on the floor, and that the sweeping of floors which
may have been spat upon should be entirely discontinued, and daily
mopping or washing substituted for it.

Spittoons have not been mentioned hitherto. If not carefully employed,
they may increase the danger of infection. The floor around spittoons
becomes soiled with spit ; and, unless the spittoon contains water or other
fluid and is carefully emptied daily and cleansed with actually boiling
water, it is a possible source of danger.

I shall be glad to advise with you further on the subject if you think
this desirable. If you have any suggestions to make as to practical means
of carrying out the principle of prompt removal of the infection derived
from dried spit, you will be conferring a public favour by communicating
them to me. — I am, Sir, yours obediently,

MEDICAL OFFICER OF HEALTH

There is no difficulty in securing the exhibition of similar
notices in each room of common lodging-houses, etc. Most
railway companies now exhibit such notices in railway stations
and in each compartment of railway carriages.

SHOULD EXPECTORATION IN STREETS BE FORBIDDEN ?—
When we remember the immense change which has taken place
in our national habits as to spitting, it will be realised what
progress has already been made in preventing the spread of
infection by sputum. Not many decades since nearly every
home was supplied with spittoons, and spitting into the fire
or fireplace was common. Now spittoons are almost unknown
except in public-houses and barbers' shops, and domestic spitting
seldom occurs. If it does, the person finding it necessary to
spit retires to a lavatory or water-closet. There is still much
public nuisance from expectoration deposited on public pave-
ments and roadways, and there must be carriage of infected
material from such deposits by means of dress-skirts and boots
into houses. It would not, however, be wise to ask for regulations
forbidding outdoor expectoration, even though the operation
of these was confined to towns, for such regulations would go
beyond present public opinion, and would be systematically
evaded. It would, however, be well to regulate outdoor ex-
pectoration, restricting it to certain defined parts of each street.

INDISCRIMINATE EXPECTORATION 337

A bye-law to forbid outdoor expectoration, except over street
gully-tanks, would do much to educate public opinion and keep
the streets clean ; and a bye-law which, though less rigid than
the above, would forbid outdoor expectoration except into
the channel between the roadway and pathway would be
beneficial. These bye-laws by calling attention to the need
of frequent swilling of the street-channels would conduce to
the public health, by the prevention of dust in general as well
as in reference to tuberculosis.

CHAPTER XLII
THE NOTIFICATION OF PHTHISIS

UP to the present point we have considered preventive
measures against phthisis chiefly in their relation to
the patient and his doctor ; slightly and incidentally, but
viewed from the same standpoint, the relation of the public
to the patient and his doctor. It is necessary that this wider
aspect of preventive measures should now be more fully denned.
We need not fight over again the battle as to whether the
conditions favouring infection or infection itself are the more
important. Both are important, and no hygienist would be
willing to content himself with removing insanitary areas, im-
proving the ventilation, lighting, and cleanliness of houses,
preventing industrial dust, and increasing the nutrition of the
poor, without at the same time adopting measures against
indiscriminate expectoration, or without, where practicable,
removing advanced cases of phthisis from the midst of large
families, in which they cannot be nursed suitably without risk
to others.

The great advantage of having cases of phthisis notified is
not only that each notification enables personal preventive
measures to be taken against infection, but also that each case
becomes the point d'appui for the detection of other hitherto
unrecognised cases, and for the discovery and removal of in-
sanitary circumstances and conditions either in domestic or
industrial life. It converts the patient from a focus of infection
into a focus of prevention.

OBJECTIONS TO NOTIFICATION OF CASES

It is perhaps somewhat belated to consider these, as very
few now object to systems of voluntary notification and the
action taken thereon, and there is an increasing volume of

advocacy of compulsory notification of phthisis. It is, however,

338

THE NOTIFICATION OF PHTHISIS

339

convenient to enumerate briefly the main objections which
have been urged against notification, as their fallacy is not
always recognised as clearly as it should be.

(1) It has been commonly urged that notification of cases is
of relatively small value, because most of the cases — even in the
absence of wilful concealment — will have been infectious for
a long time before being notified, and that therefore attempts
to destroy infective material derived from the patient can have
only a partial and limited success. I can see no ground for this
reasoning. It is agreed that risk of successful infection increases
with increased dosage, and it is probable that advanced cases are
usually more bacilliferous, or at least eject more bacilliferous
sputum than early cases. It is evident, therefore, that at
whatever stage precaution is taken, it must reduce the dose of
infectious material and the risk of infection which varies with
it. But this is really an understatement of the case. The
healthy occupants of a tuberculous home may be compared
to a city which is the subject of a protracted siege, in which
the combined effects of arms, and starvation, and depressing
emotions are at work. The inhabitants of such a city may
escape with but little damage if the siege is raised at a com-
paratively early period ; but they succumb if it is protracted.
Similarly the healthy members of a tubercle-invaded house-
hold may be able to withstand infection if precautionary measures
are begun as soon as the nature of the disease is detected and
are continued thereafter ; but they eventually fall victims to
the cumulative infection if a fatalistic inertia is allowed to
prevail, and no efficient precautions are taken.

(2) In the past some use has been made of the argument
that as the tubercle bacillus enjoys a saprophytic existence
apart from its human host, measures directed solely to prevent-
ing infection from the patient will be ineffective. The same
line of answer as to the first objection holds in this case ; and
the objection involves the assumption, which should be unfounded
in actual practice, that notification is not intended to be accom-
panied by measures of disinfection and cleansing directed against
the bacillus in its exiguous saprophytic environment.

(3) The objection that equally efficient action against the
defects found after notification can be taken apart from such
notification, has already been answered (pp. 321 and 328).

340 THE PREVENTION OF TUBERCULOSIS

(4) The risk of interference with the patient's occupation
has been shown not to exist in practice (p. 329). On this point
there has been confusion between the possible but unrealised
evil effect of notification, and the independent fact that the
public on their own initiative, and apart from notification,
have occasionally had exaggerated fears as to the risks of work-
ing with consumptives.

THE IMPOSSIBLE MAGNITUDE OF THE TASK? — (5) It has
been urged also that as phthisis is, unlike the infectious diseases
now notifiable, a disease of protracted duration, the carrying
out of official preventive measures is impracticable, and would,
if attempted, involve a larger staff than is possessed by any
local Sanitary Authority. This objection can be tested by
an estimate of the number of cases of phthisis in an average
population of 100,000 persons. This will be 380 on the basis
of the data given in the table on p. 63. If we assume that there
are five cases of active phthisis, each living a year of life in the
community in which one annual death from that disease occurs,
instead of three as assumed in the table, then there will be 633
cases among 100,000 persons. Many of these cases will need
no visits from the medical officer of health or his assistant. To
ensure a quarterly visit to 400 of them, about thirty visits would
need to be made each week. The number of visits actually
needed is much reduced by having consultations at the medical
officer of health's office. By this means the cases not actually
under a doctor can be kept under supervision with relatively
little difficulty, especially when the medical officer of health
is the medium through which sanatorium treatment is secured.
In a larger population it is simply a question of additional help ;
but the above figures will show that the amount of help required
is much less than has been stated.

LE SECRET MEDICAL. — (6) The only valid objection is one
which, in theory at least, presses hard against a voluntary system
of notification. It is that, in the absence of a statutory obliga-
tion, the notifying doctor may be laying himself open to awk-
ward consequences. This is a real difficulty, and must neces-
sarily always limit the operation of voluntary notification of
phthisis to patients of the poorer classes, and particularly to
those treated in connection with the poor law or with public
institutions. Among these patients I have found that visits

THE NOTIFICATION OF PHTHISIS

500

450

400

350

300

FIG. 37. — Brighton. Showing the parallelism between the number of Con-
sumptive Patients treated in the Sanatorium, of cases of Consumption
notified, and of Specimens of suspected Sputum examined

342 THE PREVENTION OF TUBERCULOSIS

by the medical officer of health are not unwelcome, and that
they are grateful for the help they receive in having their rooms
cleansed and purified, etc. In our local experience in Brighton,
we have secured in addition, under a voluntary system, the
notification of a considerable proportion of cases of phthisis
among persons above the wage-earning classes. This is owing
partly to the fact that in a relatively small town personal in-
fluence counts to a greater extent, and partly to the provision
of sanatorium treatment for the notified cases. This is shown
clearly in the diagram on preceding page.

It will be noticed that specimens of sputum were more
readily sent for examination by doctors when sanatorium
accommodation became available. It may be added that
in 1906, when the available beds at the sanatorium were increased
from 10 to 25, a further marked increase of specimens of sputum
occurred. The number of cases notified has, I think, approxi-
mated towards the maximum ; and, in the future, I look rather
towards earlier notification of cases than to any great increase
in their number.

In the light of an experience like the above, it is plain that
voluntary notification may be practised on a large scale, and
without involving any such risks as have been feared. My
advice has always been, when consulted on the point by doctors,
that they should not notify outside of hospital and dispensary
practice, without first mentioning their intention to the patient.
When the confidence of the inhabitants as well as of the family
doctor has been gained, there is little difficulty in securing
the notification of a large proportion of the total cases.

The advantages secured by notification are sufficiently in-
dicated in the preceding pages and on pp. 321 and 328. Even
with incomplete notification, a large mass of infection can be
brought under control, and circumstances conducing to infection
can be minimised.

THE GROWTH OF VOLUNTARY NOTIFICATION OF PHTHISIS.—
Nothing is more remarkable in the history of English public
health administration than the rapid conversion of the medical
profession and of the public to the necessity for the notification
of cases of phthisis. The tubercle bacillus was discovered by
Koch in 1882, and Cornet's investigations into house-infection
were published in 1886. Very soon after this, instructions

THE NOTIFICATION OF PHTHISIS 343

began to be given to patients at several hospitals and dispen-
saries, defining the precautionary measures required. As early
as 1887 and 1888, Dr. James Niven printed and distributed
to every house in Oldham elementary directions for the pre-
vention of infection. In 1892, Mr. C. E. Paget prepared for
the North- Western Branch of the Society of Medical Officers of
Health a memorandum of instructions in methods of prevention.
At a meeting of the parent Society of Medical Officers of Health
on August 4, 1893, the following resolutions were passed unani-
mously on the motion of the present writer : —

That the Society of Medical Officers of Health, while accepting the
view that phthisis is an infective disease, in the prevention of which active
hygienic measures should be taken, think it premature to recommend the
compulsory notification of a chronic disease like phthisis. They are of
opinion that it is incumbent on medical officers of health to take such
steps as may secure — (a) the voluntary notification of cases of phthisis
by medical officers of public institutions and such medical practitioners
as agree that precautionary measures are desirable ; (6) the adoption of
such precautionary measures, including the disinfection of rooms, as can
be arranged in conjunction with the family practitioner. For this purpose
the memorandum prepared by the North- Western Branch of the Society
of Medical Officers of Health would give an excellent basis of action.

Towards the end of 1893 a scheme of notification recommended
by Dr. Niven was adopted by the Oldham Medical Society,
and by it urged, though unsuccessfully, on the Town Council.
Had it not been for this failure, the voluntary notification of
phthisis would, owing to Dr. Niven's pioneer action, have been
much earlier adopted in this country than actually occurred.
This scheme was published in the Lancet on November 18,
1893. In 1894 a voluntary system of notification of phthisis
was begun in New York ; while from 1898 onwards the notifica-
tion of cases of this disease was made obligatory on doctors
in that city.

In England the voluntary notification of cases of phthisis
was begun in January 1899 in Brighton, and in September 1899
in Manchester, and since then a considerable number of other
towns have adopted it, with very varying success. In the
following table the extent to which notification has succeeded
is shown. In Sheffield compulsory notification of phthisis
has been adopted under a special local Act, and its figures are
compared with those of other towns in Table LXXVI. It will be

344

THE PREVENTION OF TUBERCULOSIS

noted that the number of cases notified is stated in terms of the
total deaths from phthisis instead of in terms of population,
in order to give a more accurate proportion between cases
notified and total cases (which may be regarded as a constant
multiple in each town of the number of deaths from phthisis).

TABLE LXXVI

Number of Cases of Phthisis notified in each Town to every 100 Deaths
from the same Disease

New York (compul-
sory notification
from 1898) .
Brighton (voluntary
notification)

112

167

20 1

173

153

137
61

175
oq

197

T^>8

211

174

251

2OQ

265

179

2O2

Manchester (volun-
tary notification) .
Liverpool (voluntary
notification)

...

138

118

I3Q

112
16-5

H3
149

109

TT6

142
1 c.o

126
I4Q

Sheffield (voluntary
notification to
1904, compulsory
notification from
10,04) .

4.Q

IC4

IC2

ICC

Under a voluntary system of notification in Brighton we have
(December 1906) under observation and being visited at regular
intervals 667 cases of phthisis, or about four times the annual
number of deaths from this disease. In other towns than those
named above the extent to which voluntary notification has
succeeded varies greatly. In the Metropolitan boroughs dis-
satisfaction is generally expressed with the results of voluntary
notification of phthisis, and the adoption of compulsory notifica-
tion is being urged.

THE COMPULSORY NOTIFICATION OF PHTHISIS. — The risks
of notification to the patient's pecuniary or social welfare have
already been shown to be merely imaginary under a properly
administered system. The information is confidential, and
for an officer of a Local Authority to use it to the detriment
of the patient would be likely to imply serious consequences
to himself. I have never heard of any such instance of improper

THE NOTIFICATION OF PHTHISIS 345

use of the information furnished by notification. The great
advantage of compulsory notification is that it relieves the
notifying doctor of any fear that he is improperly revealing
confidential information. He is merely fulfilling his statutory
obligation. This is a great gain, and usually must conduce
to more complete and often to earlier notification of cases, and
consequent earlier adoption of complete preventive measures.
The experience of New York, however (Table LXXVL),
in which city the number of cases notified compulsorily was less
for a couple of years than it had been under the previous system
of voluntary notification appears to indicate that compulsion
may occasionally carry with it some factor tending to depress
the number of notifications. Sheffield under the guidance
of Dr. Robertson was the first town to adopt the compulsory
notification of phthisis, under a local Act, which came into force
in January 1904. Sec. 45 of the Act dealing with this subject
is as follows : —

SEC. 45, SHEFFIELD CORPORATION ACT, 1903

(1) (a) Every registered medical practitioner attending on or called
in to visit any person within the City shall forthwith on becoming aware
that such person is suffering from Tuberculosis of the Lung send to the
Medical Officer of Health a certificate on a form to be supplied to him
gratuitously by the Corporation, stating the name age sex and place of
residence and employment or occupation (so far as can be reasonably
ascertained) of the person so suffering and whether the case occurs in
his private practice or in his practice as medical officer of any hospital
public body friendly or other society or institution.

(6) Any such medical practitioner who fails to give such certificate
shall be liable on summary conviction to a fine not exceeding forty
shillings.

(c) The Corporation shall pay to every such medical practitioner for
each certificate duly sent by him in accordance with this section a fee of
two shillings and sixpence if the case occurs in his private practice and
of one shilling if the case occurs in his practice as medical officer of any
hospital public body friendly or other society or institution.

(d~) A payment made to any medical practitioner in pursuance of this
section shall not disqualify that practitioner from serving as a member
of the Corporation or as a Guardian of a Union situate wholly or partly
in the City or in any municipal or parochial office.

(2) (a) Where the Medical Officer of Health certifies that the cleansing
and disinfecting of any building (including in that term any ship, vessel,
boat, tent, shed, or similar structure used for human habitation) would
tend to prevent or check Tuberculosis of the Lung the Town Clerk shall
give notice in writing to the owner or occupier of such building that the

346 THE PREVENTION OF TUBERCULOSIS

same or any part thereof will be cleansed and disinfected by the Corpora-
tion at the cost of the Corporation unless the owner or occupier of such
building informs the Corporation within 24 hours from the receipt of
the notice that he will cleanse and disinfect the building or the part
thereof to the satisfaction of the Medical Officer of Health within the
time to be fixed in the notice. If within 24 hours from the receipt of
such notice the owner or occupier of such building has not informed the
Corporation as aforesaid or if having so informed the Corporation he
fails to have the building or the part thereof disinfected as aforesaid
within the time fixed by the notice the building or the part thereof shall
be cleansed and disinfected by the officers and at the cost of the Corpora-
tion under the superintendence of the Medical Officer of Health. Pro-
vided that any such building or part thereof may without any such notice
being given as aforesaid but with the consent of the owner or occupier
be cleansed and disinfected by the officers of and at the cost of the Corpora-
tion under the superintendence of the Medical Officer of Health.

(6) For the purpose of carrying into effect the provisions of this sub-
section the Corporation may by any officer authorised in that behalf who
shall produce his authority in writing enter on any premises between the
hours of ten o'clock in the forenoon and six o'clock in the afternoon.

(c) Every person who shall wilfully obstruct any duly authorised
officer of the Corporation in carrying out the provision of this sub-section
shall be liable to a penalty not exceeding forty shillings and if the offence
is a continuing one to a daily penalty not exceeding twenty shillings.

(3) (a) The Medical Officer of Health generally empowered by the
Corporation in that behalf may by notice in writing require the owner
of any household or other articles books things bedding or clothing which
have been exposed to the infection of Tuberculosis of the Lung to cause
the same to be delivered over to an officer of the Corporation for removal
for the purpose of disinfection and any person who fails to comply with
such requirement shall be liable on summary conviction to a penalty
not exceeding five pounds.

(&) Such articles books things bedding and clothing shall be disinfected
by the Corporation and shall be brought back and delivered to the owner
free of charge.

(4) If any person sustains any damage by reason of the exercise by
the Corporation of any of the powers of sub-sections (2) and (3) of this
section in relation to any matter as to which he is not himself in default
full compensation shall be made to such person by the Corporation and
the amount of compensation shall be recoverable in and in the case of
dispute may be settled by a Petty Sessional Court.

(5) No provisions contained in any general or local Act of Parliament
relating to infectious disease shall apply to Tuberculosis of the Lung or
proceedings relating thereto under this section.

(6) All expenses incurred by the Corporation in carrying into effect v
the provisions of this section shall be chargeable on the District Fund
and General District Rate.

(7) The Corporation shall cause to be given public notice of the effect
of the provisions of this section by advertisement in the local newspapers

THE NOTIFICATION OF PHTHISIS 347

and by handbills and shall give formal notice thereof by registered post
to every medical practitioner in the City and any other registered
medical practitioner known to be in practice in the City and otherwise
in such manner as the Corporation think sufficient and this section shall
come into operation at such time not being less than one month after the
first publication of such an advertisement as aforesaid as the Corporation
may fix.

(8) The provisions of this section shall cease to be in force within
the City at the expiration of seven years from the date of the passing
of this Act unless they shall have been continued by Act of Parliament,
or by Provisional Order made by the Local Government Board and con-
firmed by Parliament which Order the Local Government Board are
hereby empowered to make in accordance with the provisions of the
Public Health Act, 1875.

(9) The term " Medical Officer of Health" in this section shall mean
the Medical Officer of Health for the time being of the City or any person
duly authorised to act temporarily as Medical Officer of Health for the
City.

The amount of notification hitherto secured under this
local Act is, as shown in the preceding table, not materially more
than in Manchester and Liverpool and less than in Brighton
under systems of voluntary notification. It would, however,
be unwise to base on these facts inferences as to the relative
value of the voluntary and compulsory notifications of phthisis.
Notification, whether voluntary or compulsory, is but a means
to an end, and it may be that the circumstances of these com-
munities including their arrangements for treating the notified
patients differ so much as to render their statistics of notifica-
tion almost incomparable. It has to be remembered in the
first instance that Brighton has a population which is only
one-fourth that of Sheffield, and from one-fifth to one-sixth of
that of Manchester or Liverpool. This renders the personal
supervision of notified cases by the medical officer of health
relatively easy, and generally helps in smoothing the working
of the system.

In the next place, no statistics are at present available as
to the stage of disease at which cases of phthisis are notified.
The third consideration is that

The success of notification, whether voluntary or compulsory,
depends in the main on the extent to which a Local Authority and
its officers can be helpful to the notified patients. And herein
lies, I think, the success of successful voluntary notification.

348 THE PREVENTION OF TUBERCULOSIS

Notification is the necessary channel through which the avail-
able help comes. Although it to some extent anticipates what
is said in later chapters, the character of this help may be now
summarised : —

(1) Paper handkerchiefs and pocket spit-bottles are pro-
vided whenever indicated.

(2) When the visits are made at the patient's home, every
possible assistance is given in securing for the patient any help
needed. The parochial authorities, the Charity Organisation
Society, and other voluntary agencies are used as far as practi-
cable. Where the patients are poor, out-patient letters for the local
hospital or dispensary are given, in order that the patient may
not be stinted of cod-liver oil and other remedies. Further-
more, if any other member of the same family appears to be
failing in health and a doctor's fees cannot be afforded similar
letters for the hospital or dispensary are given, the importance
of early treatment of illness and of the maintenance of health
being emphasised in every possible way.

(3) Sanatorium treatment is offered in all cases suitable
for it, and in actual fact more than half of the total cases at
present under observation in Brighton have spent at least four
weeks in the Borough Sanatorium, and have there been taught
the precautionary measures needed to prevent infection, and
the personal regime indicated by their illness ; while at the
same time their families have had a temporary holiday from
the charge of the patient, the house has been disinfected, and
the patient has returned with a knowledge of the means to
avoid re-infecting it.

The chief reason for the success of voluntary notification
of phthisis in Brighton has been the provision for the sanatorium
treatment of notified cases. If the dates in the following table
be compared with the curves in Fig. 37 the coincidence between
the provision of increased sanatorium treatment and increased
notification will be evident.

BRIGHTON

Voluntary notification of phthisis begun . Jan. 1899.

Four beds reserved at a sanatorium outside

Brighton ..... May 1902.

THE NOTIFICATION OF PHTHISIS 349

Four beds opened for phthisis at the borough

isolation hospital .... July 1902.

The number of beds for phthisis at the isolation

hospital increased to ten . . . Dec. 1902.

The number of beds for phthisis at the isolation

hospital further increased to twenty-five . April 3, 1906.

At first the patients were admitted for only a month, the
principle adopted being that of training the patients in personal
hygiene, and in the general management of their illness, rather
than of attempt at cure. The wisdom of this plan has been
fully justified by experience. The majority of patients have
been found to have extensive lung disease, often with cavitation,
when admitted to the sanatorium. Such patients commonly
have several years of life before them, but the experience of other
sanatoria shows that prolonged treatment of many months, or
even over a year, is necessary to ensure anything approaching
to a cure even in cases in earlier stages of the disease. It is
much more to the public interest to pass a large number of
patients through the sanatorium and train them thoroughly in
the hygienic requirements of their disease, than to treat a smaller
number for a more protracted period. It is furthermore much
more convenient for the patients, who often find it difficult or
impossible to leave their families and work for longer than a
month. Our experience is that advice as to the deposit and
disposal of sputum given at home is commonly neglected ; and
that it is very rarely neglected by patients who have been in
the sanatorium. We welcome re-admissions to the sanatorium
of patients whose health is again flagging. By this and other
means, and by quarterly visits at the home of the patient, we
keep in sympathetic relationship with the patients, and ensure
the maintenance of precautionary measures against infection.

SHOULD THE NOTIFICATION OF PHTHISIS BE MADE GENERALLY
COMPULSORY ? — The preceding facts and considerations will
prepare the way to the conclusion that at present it would be in-
expedient, unwise, and of relatively little use to advise the general
adoption of compulsory notification of phthisis unless Local
Authorities are not ready to utilise the information thus received
to the benefit of the patient and of the public. I place the two
together, because they are substantially identical. It would,

350

THE PREVENTION OF TUBERCULOSIS

in my opinion, be premature for any community to adopt
compulsory notification of phthisis which (a) does not possess
a sufficient staff of skilled visitors, preferably medical men or
women, to visit the notified cases ; and (b) does not possess
at least a few beds available for the treatment and training of
consumptive patients. Under these circumstances compulsory
notification can be made to work even in the present state
of public opinion to the benefit of all concerned ; without such
aid, apart from the notification of all poor-law cases, the
voluntary notification of all other cases forms the best prepara-
tion for compulsory notification of all cases.

CHAPTER XLIII

THE SANITARY AUTHORITY IN RELATION TO
PREVENTIVE MEASURES AGAINST PHTHISIS

THE persons primarily concerned in the management of
a tuberculous patient are the patient himself and his
doctor. Happily preventive measures and curative
measures overlap and to a large extent are identical. Hence
when this fact is realised, the co-operation of patient and doctor
in carrying out preventive measures may be confidently expected.
Very often, however, it is not realised. Patients may be ignorant,
careless, or indifferent. In the later stages of their illness they
may be unable, unhelped, to adopt the necessary precautions.
Many doctors furthermore are too busy to explain the necessary
instructions as to precautionary measures ; and whatever the
reason, these instructions are frequently found in actual official
experience not to have been given until the visit of the medical
officer of health or his assistant is made, or, when given, not to
have been carried out. The intervention of the Sanitary
Authority is necessary, under present conditions, to ensure
preventive measures being taken to the extent required by the
necessities of public health. Some parts of the duty of the
Sanitary Authority in this connection have been already con-
sidered. Of these the first is to ensure the early diagnosis of the
disease ; and for this purpose no Sanitary Authority can be
regarded as fulfilling its duty which does not provide facilities
for the

FREE BACTERIOLOGICAL EXAMINATION OF SPUTUM. — This
is already being done in many towns, and should become uni-
versal. Further details on this point are given on pp. 52 and
314. Next comes the organisation of arrangements for the

NOTIFICATION OF CASES. — Whether this should be voluntary
or compulsory will depend on local needs and possibilities, and
on the considerations urged in Chapter XLII.

352 THE PREVENTION OF TUBERCULOSIS

BYE-LAWS PROHIBITING INDISCRIMINATE EXPECTORATION
form an important official means of preventing infection. The
extent to which these are at present practicable is indicated
on p. 334.

A case of phthisis having been notified, what action follows
as the result of this notification ?

(a) COLLECTION OF NECESSARY INFORMATION. — The method
to be employed depends on whether the patient desires sana-
torium treatment, and whether this is available. In Brighton
a very high proportion of the cases notified bring the notifica-
tions with them to the Town Hall, often with a letter from
their doctor, applying for sanatorium treatment. The patient
is then interviewed by the medical officer of health, and the
full particulars indicated on the following inspection card are
obtained. If the patient does not call at the Town Hall, the
medical assistant of the medical officer of health visits him
at home. Owing to patients being at work, or being unwilling
at the first interview to give as full information as is required,
a second or even a third visit is occasionally required before the
complete history of each patient can be obtained. The in-
formation is written on a stiff four-paged inspection card 8x4
inches. The first page is as follows : —

NOTIFICATION OF PHTHISIS
Reg. No. Sanatorium No.

N ame , Age.

Address

Date of Notification Doctor

Recommended for Sanatorium by

Notes by Doctor

Date of Admission to.

Date of Discharge from_

THE SANITARY AUTHORITY

353

Date of Change of Address.
New Address

Dates of Visit.

On the inside second and third pages information under
the following headings is obtained : —

Duration and History of Illness.

Places of Residence during Illness.

Occupation and Workplaces during last 5 years.

(a) Wages.

Work regular.

No. and Ages in same
Family.

No. in 2nd Family.

History of Cough or
Consumption among these.

Family History.

Precautions : —
(i) Card

(3) Pocket Spittoon.

(4) Habits as to Spitting.

,(2) Handkerchiefs.

354 THE PREVENTION OF TUBERCULOSIS
(5) Other Occupants of same Bedroom

(6) House.

Habits as to Food and Drink.

Further Remarks.

Likely Sources of Infection :— -

(1) Same House ; (4) Neighbour.

(2) Companion_ (5) Workmates.

(3) Public-Houses (6) Others

The fourth page deals with the sanitary condition of the
home, especially as to cleanliness and crowding, space being
left at the bottom for a summary of conclusions as to exposures
to infection, which along with the statement of likely sources
of infection at the bottom of p. 3 may lead to further inquiries
and action.

Condition of Dwelling-house as to—

No. of available Dwelling Rooms ;

Overcrowding.

rof Walls

of Ceilings.
Cleanliness-(

of Floors

.of Bedding, etc..
Dampness

THE SANITARY AUTHORITY 355

Ventilation

Lighting, especially of Staircase

Size of Yard

Any Sanitary Defects _

(a) Duration of each Case.

Latest Exposure to Infection before reputed date of onset.

(d) Previous Exposures.

The inquiry form may seem to be unnecessarily elaborate, but
it is the result of long experience in the work ; and it has to be
remembered that the information often accumulates gradually,
as our acquaintance with the patient improves.

(b) GIVING OF INSTRUCTIONS. — At the first interview with
the patient the card printed on p. 324 is given, and its contents
are explained to him verbally.

At the same interview he is instructed in the methods of
using paper handkerchiefs and a pocket spit -bottle.

(c) DISINFECTION. — The next step is to ensure cleansing or
disinfection of the patient's room as required. The following
directions, quoted from a circular prepared by Drs. Niven and
Newman and myself in 1903 and issued by the National Associa-
tion for the Prevention of Consumption, may be quoted at this
point : —

The phlegm infects everything upon which it falls — handkerchiefs,
books, papers, linen, floors, carpets, furniture, etc., and when dried and
broken into dust is then readily inhaled by healthy persons.

On these facts rests the important question of disinfection. In en-
deavouring to prevent a consumptive person from spreading the disease,
two sets of preventive measures are required : — ist. The removal or de-
struction of the infective matter disseminated by the patient's phlegm ;
and, 2nd, the prevention of future dissemination. For the latter purpose
the main object is not to permit any phlegm or discharge to become dry

356 THE PREVENTION OF TUBERCULOSIS

before being destroyed. Before the consumptive person has learned the
personal precautions which must be taken, and up to the time when he
has been trained to carry them out carefully, he has probably distributed
a considerable amount of infective matter. This is especially liable to
accumulate in a dangerous form at home, where the space is small, and
light and ventilation are defective. Infective particles will be found in
greatest abundance on and near the floors, on ledges, and in room-hangings.
But the personal clothing and bedclothes will also have become infected.
Hence it is necessary to disinfect the floor, walls, and ceiling of the rooms
occupied by the patient, as well as the furniture, carpet, bedclothes, etc.

If personal precautions are taken, the risk of infection is lessened, but
it is impossible to prevent coughed-up minute drops of phlegm from
being deposited in a room, and rooms should therefore be cleaned at least
once in a month, the floors being scrubbed with soft soap, the furniture
washed, the walls cleaned down with dough. The ceiling should also
be whitewashed every six months.

Disinfection of rooms which have been occupied by consumptive
patients may be secured in various ways, but the following are the practical
rules which must underlie any methods adopted : —

1. Gaseous disinfection of rooms, or " fumigation," as it is termed,

by whatever method it is practised, is inefficient in such cases.

2. In order to remove and destroy the dried infective discharges,

the disinfectant must be applied directly to the infected surfaces
of the room.

3. The disinfectant may be applied by washing, brushing, or spraying.

4. Amongst other chemical solutions used for this purpose a solution

of choride of lime (i to 2 per cent.) has proved satisfactory and
efficient.

5. In view of the well-established fact that the dust from dried

discharges is infective, emphasis must be laid upon the import-
ance of thorough and wet cleansing of infected rooms.

6. Bedding, carpets, curtains, wearing apparel, and all similar articles

belonging to or used by the patient, which cannot be thoroughly
washed, should be disinfected in an efficient steam disinfector.

In Brighton a formalin spray is used for disinfecting rooms.
The preceding instructions when combined with direct pre-
cautions during the act of coughing suffice to prevent risk of
infection.

(d) REMEDY OF SANITARY DEFECTS.— It is unnecessary
to detail the means used for the remedy of overcrowding or
other sanitary defects found in the consumptive's home, as in
regard to these the usual procedure of sanitary administration
will be pursued. Notification has, however, secured their remedy
earlier than would have been practicable under ordinary condi-
tions (see also p. 321).

Nor for a similar reason is it necessary to detail measures

THE SANITARY AUTHORITY 357

taken in regard to workplaces, for the removal of dust, the
prevention of daily dust, and the limewashing of walls, etc.
Notices against spitting in factories, workshops, etc., such as
the one given on p. 334, are now exhibited fairly generally.

(e) EDUCATION OF THE PATIENT. — The great difficulty is to
secure that the uneducated patient will adopt the simple pre-
cautions as to coughing and spitting which are needed to prevent
infection. Most patients, whatever their class, are uneducated
in this respect, but some patients acquire more easily than others
the habit of taking the necessary precautions. My personal
experience is that very few patients can be trusted to follow
scrupulously the instructions as to coughing and spitting given
on the card printed on p. 324, except in the light of the careful
habits inculcated and the personal benefits received at a sana-
torium. Hence I consider

(/) THE PROVISION OF SANATORIUM TRAINING AND
TREATMENT as one of the most important duties of a
Sanitary Authority in regard to phthisis. The details under
this head are described in Chapter XLVIII. ; but there is no
difficulty in seeing that a medical officer of health or other
official who goes with an offer of sanatorium treatment is in
an infinitely better position for receiving a hearty welcome
than when he merely asks questions which may be regarded
as inquisitorial, and gives instructions which to the uninitiated
may seem foolish.

(g) THE PROVISION OF MEDICAL TREATMENT FOR OTHER
MEMBERS OF THE PATIENT'S FAMILY. — The welcome of the
visitor is likely to be still more cordial when it is known that
for suitable cases he has hospital or dispensary tickets, and
can ensure continuous treatment not only for the patient, but
also for other members of his household when this is indicated
(see also pp. 318 and 348).

(h) REVISITS. — In some towns visits to consumptive patients
are made monthly. In Brighton only a quarterly visit is made,
and it is probable that more frequent visits would lead to friction.
In order to prevent removal without the knowledge of the
medical officer of health, notifications of change of address
are paid for, thus ensuring in a certain proportion of cases prompt
disinfection of the vacated rooms. With the same object, a
fee of sixpence is paid to relieving officers who notify a case of

358 THE PREVENTION OF TUBERCULOSIS

phthisis, or who notify the removal of such a patient to the
infirmary or elsewhere. The cleansing and disinfection of
vacated rooms before they are occupied by another family is
one of the most important measures in connection with the
administrative control of tuberculosis.

(i) In connection with visits and revisits to the patient, the
question of helping him in gaining his livelihood under the best
conditions arises. The subject of the after-care of consumptives
is discussed in Chapter XLVIII. There will doubtless be great
future developments under this heading, but at present this
matter is chiefly one for private enterprise and charity.

CHAPTER XLIV
EDUCATION AUTHORITIES AND TUBERCULOSIS

IN previous chapters stress has repeatedly been laid on the im-
portance of teaching the laws of health (p. 302), and parti-
cularly on the necessity of having teachers taught these laws
with special reference to the prevention of tuberculosis (p. 365).
The necessity for teaching the patient the means of preventing
the spread of the disease has been emphasised on pp. 318 and 332.
The prevention of indiscriminate expectoration, which is dis-
cussed in Chapter XLL, bears on the same subject.

In all these particulars school authorities have duties which
they cannot with propriety continue to ignore. This is true
for all classes of schools, and not less true for secondary than
for public elementary schools. The majority of children attend
the latter, and the following remarks, produced from a paper
on " The School in Relation to Tuberculosis/' contributed by
me to the International Congress on School Hygiene, August 1907,
relate chiefly to them. It is convenient to reproduce here the
remarks as to the amount of open and recognisable tuberculosis
in schools, as well as those relating to its prevention.

Happily the Education Committees governing general
elementary education in this country, although they have
important specially delegated duties and have co-opted members,
form part of the local Sanitary Authority, and there is every
reason why they should actively co-operate to the fullest extent
in securing the prevention of tuberculosis. The new machinery
for the medical inspection of scholars will be an invaluable
means to this end, especially in districts in which notification
of cases of phthisis to the medical officer of health is in successful
operation.

Elementary day - schools may be considered from the
following standpoints : — (i) Whether tuberculosis is spread in
them and to what extent ; (2) whether the conditions of life

359

360 THE PREVENTION OF TUBERCULOSIS

and work in such schools tend to bring into activity latent
tuberculosis ; and (3) as important means for teaching and
training children so that we may obtain the aid of the next
generation in the rapid elimination of tuberculosis.

THE AMOUNT OF TUBERCULOSIS AT SCHOOL-AGES. — Before
we can arrive at any definite decision on the first point, it is
necessary to know how much tuberculosis there is among children
of school-age. So far as tuberculosis terminating fatally during
school-life is concerned, the figures of the Registrar-General's
reports enable this point to be settled with some approxima-
tion to accuracy for the age-periods 5 to 10 and 10 to 15,
which may be taken as practically coincident with school-ages.
Fig. 38 gives the death-rates from pulmonary and from all forms
of tuberculosis in the aggregate per million living at each age-
period in the decennium 1891-1900 (Decennial Supplement,
R.G., Dr. Tatham). The interval between the lower and higher
space in each column represents the death-rate from all forms
of tuberculosis, excluding pulmonary tuberculosis.

It will be noted that at ages under 5 pulmonary tuberculosis
only supplies about one-ninth; at ages 5 to 10 less than one-
third ; and at ages 10 to 15 not much more than one-half
of the total registered mortality from tuberculosis. At higher
ages the proportion of pulmonary to total fatal tuberculosis
becomes greater.

It will be noted furthermore that at ages 5 to 15 the
death-rate from pulmonary and from all other forms of tuber-
culosis in the aggregate is lower than at any other age-period,
except at ages over 75. It is clear, therefore, that, as a
fatal disease, tuberculosis is relatively uncommon at school-
ages. Taking the ages 5 to 15 together, it is the registered
cause of death each year of only about seven out of every 10,000
children living, while pulmonary tuberculosis only supplies three
out of these seven.

As a means of spread of tuberculosis, pulmonary tuberculosis
is supreme, all other forms of tuberculosis being almost negligible
in this respect. How many cases of pulmonary tuberculosis
are there for every fatal case of this disease ? In adults the
proportion is usually given as three to one, though this is
probably too low (see p. 63). If we assume that there are
constantly as many as ten non-fatal cases for each annual death,

EDUCATION AUTHORITIES

then three out of every thousand children at school-ages are
suffering from pulmonary tuberculosis, on the basis of the
figures of the last decennial period.

It does not follow that all these phthisical children are in
attendance at elementary schools. Many of them doubtless will
not be.

Compare this estimate with the actual results of examination

400(7

.3000

1000

IOOQ

4oo

S3§

R-V9

^N>

••*' :«'•.'!

'•.'•:-;••'

y////////,

::/-y

''$£'/:

:.;.'.:*>";>•!
.*• * ..*"•""

;>""•_'• ••-' •

'•'.'•."•;".'•

\^\^Ov

Xy'^/

§ss

.'•*'•*" '.

»':;.-.'. ','-.'; v

V "•'"•'-.***-

'-•-.'.

*•*»"*•%

•'• ••".*•" *\

' **'• '*•

••;'.-*'» T^'

?••?.?'••

i.fe>>

• «.'•'•. """

?.!$£;

:/'S:--nj:

".*>'• .-• •

'"••,*.'. >>,\\\\s

v*"-''" '"*.'.' ^:;.;<;*;^:*

•'*."' "li

:'-''".r-: *'•*.'

'•V.*'",- --,

* •'•*" T- ".:* '*.

'*"..'*• j ••. i'.v».V'.vr-.'.

>.'':•'*

;:«-VVyX"V

:'\;>-:'^ :^.-V^^

3 0-5 5-

10-

15- 2o- 25- jjr 35 — 45 — S$— 65— /S-*-

FIG. 38. — Death-rate per million living in each Age-period from Phthisis (dotted)
and from other forms of Tuberculosis (lined)

of children in elementary schools. These are given more fully
in a paper by Drs. Lecky and Horton of Brighton (1907). I
need, therefore, only briefly summarise the results. They very
exhaustively examined 806 children, of whom 491 were attending
an elementary day-school, 241 in a parochial industrial school,
and 74 in the workhouse. These children varied in age from 4
to 17. Only three cases of phthisis were found — one in the
parochial school, one in the workhouse, and one in the elementary
school. With these results may be compared the -following,
which are summarised in the same paper. At Dundee, Dr. A. P.

362 THE PREVENTION OF TUBERCULOSIS

Low (1905) found no pulmonary tuberculosis in 517 children ;
at Dunfermline, Dr. Ash (1905) had a similar result in examining
1371 children. Dr. Mackenzie, in Edinburgh, found fourteen
cases in 600 children ; Professor Hay, in Aberdeen, three cases in
600 children ; the Charity Organisation Society results, Edinburgh
(Canongate schools), give nineteen cases in 1318 children. These
results vary greatly, and it appears likely that there has been some
confusion between bronchitis and phthisis in some of the observa-
tions, a very easy mistake unless a very careful examination is
made.

Dr. Greenwood, at Blackburn, found 6*7 per cent, of phthisis
in 1028 children referred to him, but these were children whose
fitness for schools was already in question, and rather confirm
the view, which is, I think, correct, that a child failing with
phthisis usually does not remain in school long before his ill-
health is recognised.

Omitting the above negative observations, and Dr. Green-
wood's results, which represent a selected sick population, the
proportion of children in elementary schools with revealed
phthisis appears to be i in 43 (Edinburgh), i in 69 (Edinburgh,
second series), i in 200 (Aberdeen), and i in 296 (Brighton).
Compare these figures with the estimate of i in 333 children
based on the national death-rate, and on the assumption that
ten non-fatal cases go to every fatal case, I incline to think
that there is not, on the average, more than i in 300 children in
schools showing revealed or diagnosable phthisis.

Is TUBERCULOSIS SPREAD IN SCHOOLS ? — To what extent
are these children a source of infection ? Probably very little.
Children seldom expectorate ; and a child with a troublesome
cough would not be kept long in school. It does not appear
likely that there is much spread of tuberculosis from scholar
to scholar in schools.

Teachers and caretakers are possible sources of infection.
There do not appear to be trustworthy statistics of the amount
of phthisis in teachers. Probably it is somewhat more than in
the general community, and, judging by my own experience, I
should say that it is more often laryngeal than in the averages
of consumptives. The medical examination of teachers and of
caretakers, as well as of scholars, is obviously indicated as a
precautionary measure.

EDUCATION AUTHORITIES 363

THE AMOUNT OF LATENT TUBERCULOSIS IN SCHOLARS. — The
preceding figures deal with revealed tuberculosis. Latent
tuberculosis is nearly, if not quite, always non-infectious. Such
latent tuberculosis has, however, important bearings on school
hygiene. Notwithstanding the small amount of revealed
tuberculosis among school-children, such children, if they die
of other diseases, show, in a very high percentage, evidence of
tuberculosis, especially in the bronchial glands. Thus Naegeli,
at Zurich,1 found in autopsies of children aged i to 5 that
17 per cent., and of children aged 5 to 14 that 33 per cent.,
had tuberculous lesions.

Such latent lesions are undoubtedly very frequent in children.
I cannot doubt that the true interpretation of these figures,
showing as they do heavy incidence of tuberculosis before as
well as during school-life, is that tuberculous infection in children
is nearly all domestic and not scholastic in origin.

How TO DEAL WITH LATENT TUBERCULOSIS. — The presence
of such latent foci is a constant source of danger to the children
implicated. Although there is at present no statistical evidence
to that effect, it is almost certain that in the children of adult
consumptives such lesions are present to a preponderant extent,
a fact which supplies a valuable indication for preventive treat-
ment. The children of such parents should be periodically
examined by the school-doctor, and the card giving the medical
state of each scholar should have a column for family history of
consumption, and for entering any cases of this disease that
have been or may be subsequently notified in the same house-
hold. The general notification of phthisis to the medical officer
of health thus forms an essential part of school hygiene.

The course to be adopted in regard to such children is a part
of the problem of general public health administration. Two
plans are open — the removal of the children from their homes
either temporarily or permanently to homes or schools at the
seaside or in the country ; or the institutional treatment of the
consumptive parent. The former plan has been adopted on a
considerable scale in France and elsewhere, and occasionally is
the best or the only available line of prophylaxis ; the latter
plan is the one which has been chiefly employed in England,

1 Quoted by Dr. H. Mery, Rapports pr&senUs au Congrks International de
la Tuberculose, Paris, 1905, p. 298.

364 THE PREVENTION OF TUBERCULOSIS

not intentionally, but incidentally in the relegation of a very
large proportion of consumptives among the poor to the work-
house infirmary and to other institutions. Judging by inter-
national statistics, action on the latter line is more effective than
any other. It brings the greatest relief to the family, both from
privation and from infection. Supplemented by earlier treat-
ment and training of consumptives in sanatoria, it will effect
still more good ; and if there is to be a choice of remedies, the
balance of good lies on the side of measures directed towards
removing the patient himself rather than of measures for re-
moving the children from the infected domestic circle. It is
evident, however, that both remedies are excellent, and that
each consumptive family will need to be considered on its merits,
and the most practicable line of action taken. It may be re-
peated, however, that, given the choice between measures for
increasing resistance to infection, and measures for diminishing
or abolishing exposure to protracted infection, the latter must
always occupy a supreme position.

HOW TO PREVENT SCHOOLS FROM PROVOKING LATENT TUBER-
CULOSIS TO ACTIVITY.— Both in regard to the children under
special suspicion of tuberculosis, and in regard to all other
children, much can be done to prevent the school from becoming
a place in which latent tuberculosis is brought into activity.
Overcrowding is the rule in schools. A larger floor-space
should be required. Classes are too large, thus straining the
voice of the teacher, and making him much more prone to tuber-
culosis. Ventilation is usually very defective ; and the methods
of cleansing, involving the raising of dust, need reform. These
are obvious points of hygiene. In school hygiene they are pro-
minent because of the grossness with which they are neglected.
In the boarding-schools of the middle and upper classes we are
familiar with the overwork and over-fatigue due to excessive
games, as well as with the insufficient sleep to which Dr. Acland
has drawn attention. In England the children of the great
majority of the population almost certainly do not suffer from
over-fatigue due to games ; but there is little doubt that many
of these suffer from over-fatigue and want of sleep, due to
domestic and sometimes to industrial demands, and to defective
domestic arrangements. These factors cannot fail to aid in
setting ablaze the smouldering fire of latent tuberculosis. In

EDUCATION AUTHORITIES 365

each of these particulars, there is much need for detailed medical
supervision of our schools and scholars, and for the adoption of
preventive measures, on the lines that have been briefly indicated.
If these and similar reforms are secured, the school may be
made a most important centre for the prevention of tuber-
culosis. I think that the principal measures needed for this end
may be summarised as follows : —

1. The medical examination of all children on admission to

school and periodically afterwards, supplemented as it
must be to attain its full value by information system-
atically acquired in regard to the health conditions of
their homes and all living in them.

2. The exclusion of children found to have open or revealed

tuberculosis.

3. Special care as to the feeding and general hygiene of

children from tuberculous families, including avoidance
of fatigue.

4. The frequent wet cleansing of schools.

5. The reduction of overcrowding.

6. The improvement of arrangements for the ventilation and

warming of schools.

7. Careful attention to the personal hygiene of all scholars,

especially in relation to the removal of adenoids and of
carious teeth.

8. The periodical examination of caretakers and teachers,

and the avoidance of excessive strain on the voice of

the latter, or over-fatigue in general.

THE FORMATION OF PUBLIC OPINION ON TUBERCULOSIS IN
THE SCHOOLS. — Public opinion is formed in the schools ; and if
each teacher and scholar is taught to practise the laws of health,
a much more rapid decline of tuberculosis can be secured. What
has been said about the supreme importance of domestic infection
illustrates this. The inculcation of good habits as to coughing,
expectoration, and scrupulous domestic cleanliness, and of
knowledge as to the relative value of foods and the dangers of
alcoholic drinks, will go far towards making the school a valuable
aid in preventing tuberculosis.

CHAPTER XLV

THE BOARD OF GUARDIANS AND THE PREVENTION OF

PHTHISIS

IN previous chapters we have discussed in relation to the
prevention of phthisis the functions of the doctor, of
his tuberculous patient, and of the Sanitary Authority and
the Education Committee as at present constituted in this country.
One local governing body remains whose present functions in
this connection are not less important than those of the two
bodies already mentioned. This is the Board of Guardians,
whose duties are to relieve the destitute, giving food, lodging,
and medical aid when required. The importance of such aid
in preventing phthisis and in helping to diminish the danger
of its spread is at once evident. The fact that the help given—
especially the domestic medical aid — is ofttimes belated and
insufficient (see p. 307) is well known ; while the importance
of the institutional relief given by Boards of Guardians has
not been sufficiently realised in the past. Its bearing on the
past prevalence of phthisis has been fully discussed in Part II.
If there is one point that I am more desirous of making common
property than another, it is that in the improved and more
general institutional treatment of advanced cases of phthisis
we have the means ready to hand from which the greatest
quickening of the rate of decline of the death-rate from this
disease can be expected.

THE INSTITUTIONAL TREATMENT OF ADVANCED CASES.—
So long as Boards of Guardians remain a separate local govern-
ing body and are hemmed in by present regulations in giving
indoor medical relief, this timely and general treatment cannot
be obtained. It is to be hoped, however, that ere long sickness
will be the sole and sufficient condition of prompt and efficient
medical treatment for all requiring it. This will imply the
removal of the parochial stigma from treatment in a workhouse

THE BOARD OF GUARDIANS

367

infirmary. The infirmary will, in fact, no longer be an annexe, —
except perhaps structurally, — of the workhouse. Until this
reform is secured, the local problem for administrators is to
secure for cases of phthisis in the workhouse infirmary the most
abundant and the most efficient use of separate wards consistent
with present regulations. There is no compulsory power of
removal or detention in these wards. The best policy is, by
provision of sufficient and palatable food, by good medical
attendance and nursing and general comfort, to make the con-
sumptive patients unwilling to go home. This advice may
appear to be contrary to the first principles of poor-law adminis-
tration. It is, however, actually calculated to diminish [pauperism,
which ought to be the object of every one concerned. The
return of consumptive patients to small homes, in which due
precautions are not likely to be taken, is an effective means
of growing a later crop of consumptive paupers. The general
conditions of treatment of advanced consumptives in the wards
of infirmaries do not differ materially from those in sanatoria.
The wards will, however, in view of the more serious illness
of the patients, be kept warmer ; lighter and more easily
masticated food will be required ; and precautions as to the
coughing and expectoration of the bedridden patients will
need to be precise and rigidly carried out. Much can be done
even for advanced patients to increase their comfort and to
smooth their path during progressively increasing weakness.

The medical superintendent of the infirmary occasionally
has to deal with another class of consumptive, who is extremely
difficult to control. He is not very ill, he has a troublesome
cough, and is addicted to indiscriminate spitting. He is occasion-
ally obstreperous, and the temptation then, and even short of
this if the patient is dirty in his habits, is to relegate him to
the able-bodied part of the workhouse as a punishment. This
is obviously unfair to the able-bodied paupers, and some other
means, such as separate warding, ought to be devised.

At this stage comes in the difficulty that the patient will
probably " take his discharge," and leave the institution, going
back to a common lodging-house, where he will continue to
disseminate infection. For such patients, — and for such patients
only in my opinion, — the power of

COMPULSORY REMOVAL TO AND DETENTION IN AN INSTITUTION

368 THE PREVENTION OF TUBERCULOSIS

is indicated. We are much more timid on this subject than
our cousins in the United States, as shown by the following
remarks made by Dr. Knopf at a recent Conference of Sanitary
Officers of the State of New York :—

New York was the first city in the world which enacted the compulsory
removal law in regard to tuberculosis. That is to say, if in the opinion
of the inspector, the physician in charge, or the visiting nurse, the tuber-
culous patient is a menace to his fellow-men, he is removed to a hospital
whether he likes it or not. Now you may think that those patients are
refractory and might not do well in the hospital. Not at all. It is my
privilege to be on service as attending physician for six months in
the year at the Riverside Sanatorium for Consumptives, which is
in charge of the New York City Health Department. Half of these
patients are there against their will, and you would be surprised what a
change it makes in their condition to remove them from the dark, dreary
tenement houses — where they have neither light, air, nor decent food —
into a clean bed, plenty of air day and night, and give them good food,
including eggs and milk. We never lock up the eggs. We tell the patients,
" Go and help yourselves." They can drink all the milk they wish. You
would be surprised what results we obtain there in spite of the cases
being, in the majority, far advanced, and in spite of their being forced
to go there. If they recover, in not a few instances they become better
men and women. The results as a whole are most satisfactory. Thus I beg
of you not to be alarmed when you hear the words compulsory removal.
It is the most humane and scientific way of treating the consumptive
poor, who are a menace to their neighbours, without food and air, or
entirely homeless.

This experience in New York is interesting ; but it would
be a mistake to conclude that any such practice would be wise
in this country. Resort to compulsion, if it were thought
advisable, should undoubtedly be hemmed in by special con-
ditions, such as special investigation and a magisterial decision.
There are, however, cases of the nature indicated above, of
persons lodged in common lodging-houses or in crowded dwell-
ings who cannot secure proper nursing and attention, and who
are suffering to an unnecessary extent themselves, and inflicting
suffering and unnecessary danger on those about them ; persons,
again, who are already in the infirmary but wish to return to
the above conditions ; in whose cases there is need of com-
pulsory removal or detention. In the vast majority of cases
there is no need for compulsion, and the power to enforce it
against them is undesirable. For them, the one thing necessary
is to make the institutional treatment satisfactory to the patient

THE BOARD OF GUARDIANS

369

as well as conducive to the public interests. As has appeared
so often in considering questions relating to phthisis, this means
of protecting the community is identical with the best treat-
ment for the patient, whose cure will usually be the more rapid
and more probable if the circumstances in which he is treated
are attractive to him.

SANATORIA AND BOARDS OF GUARDIANS. — Liverpool and
Bradford have been pioneers in providing for the treatment
of comparatively early cases of phthisis through Boards of
Guardians. It is to be hoped that other Boards will follow
their example. It must be noted, however, that when a patient
becomes ill enough to be a pauper, he is usually suffering from
well-established or advanced disease, and that the chief medical
function of the Board of Guardians under present arrange-
ments is the treatment of patients who are so ill as to be com-
pletely unable to work. While infirmary treatment involves
the stigma of pauperism, far more patients will struggle against
the disease till they are past recovery, in the hope of avoiding
the workhouse, than will apply for infirmary treatment at a
stage at which it can have a fair chance of producing
recovery, and before they have sown widespread infection
in their environment. At present, therefore, workhouse
infirmaries cannot usually cover so wide a field as the local
Sanitary Authority, which may succeed in obtaining patients
for treatment at a stage before tuberculosis has produced actual
disablement. The Boards of Guardians have, in fact, the
accommodation and arrangements for treatment without being
able to secure the patients at the most favourable time ; the
Sanitary Authority can secure the patients, but seldom or never
has the accommodation and arrangements for treating them.
This inefficient state of things points to the need for finding
a way of combining the resources and functions of the two
Authorities in respect to the treatment of the sick. Such a com-
bined Authority would then be able to carry out the complete
institutional treatment of this disease among the poor, namely :—

1. The protracted sanatorium treatment of suitable early cases.

2. The shorter treatment of cases of longer duration, among
patients still able to earn their livelihood, with a view to tem-
porary improvement, and to training in the management of
their illness (pp. 357 and 391).

370 THE PREVENTION OF TUBERCULOSIS

3. The protracted institutional treatment of advanced cases,
when the home conditions are unfavourable.

THE HOME TREATMENT OF PAUPER CASES OF PHTHISIS.—
The Board of Guardians is frequently faced with the problem
of giving outdoor relief to the family of a consumptive patient,
to enable medical treatment and nursing of the patient to be
continued at home. If any general rule is to be followed in
such cases, it should be to the effect that outdoor relief ought
never to be given to consumptive patients. Exceptional
cases may occur, as, for instance, when the household consists
only of the patient and his wife ; but even then it is usually
wiser to admit the patient to the infirmary, release the wife
from the constant and unrelieved stress of nursing, night and
day, and when necessary give her outdoor relief after her husband
has been placed in the position of receiving proper medical
aid in the infirmary. If there are children in the family, under
the domestic conditions in which those needing parochial aid
live, such aid ought seldom if ever to be given except on the
condition that the patient becomes an in - patient at the
infirmary. In the light of the past history of phthisis in this
country, and of the important part which has been played by
these infirmaries in securing the past decline of the death-rate
from this disease, no other course is justifiable either in the
public interest or with a view to safeguarding the patient's
family.

THE RELIEF OF THE CONSUMPTIVE'S FAMILY. — The fatigue
and chronic mal-nutrition in the families of the poor associated
with the nursing of a consumptive are powerful influences
favouring the active development of tuberculosis ; and there
is no doubt that the provision of food, clothing, etc., at the public
expense, when required, would tend to diminish this risk for
the patient's family ; and would diminish the risk of relapse in
patients who have been sent home from a sanatorium after
favourable treatment. Dr. Niven has specially drawn attention
to the need for a fund from which assistance can be given to
households in which the breadwinner is struck down with phthisis
while the children are too young to earn wages, and recommends
that this fund should be administered in connection with the
official scheme of notification.

This is a problem in which Boards of Guardians and private

THE BOARD OF GUARDIANS

371

philanthropy can both bear a part. In my opinion the medical
officer of health or his subordinate should not have a direct
share in the administration of such relief ; but he should be
responsible solely for such relief as can be given by medical
and sanitary measures. The most efficient means of relieving
the family, and the means which most effectively removes the
risk of further cases of tuberculosis, is the provision of satisfactory
institutional treatment for the patient, the disinfection of the
home, and the removal of insanitary conditions. At the same
time the medical officer of health can set in operation both
official and private charity for the rest of the household when
the need for these is indicated.

ate

CHAPTER XLVI

INSURANCE AND FRIENDLY SOCIETIES IN RELATION
TO THE PREVENTION OF PHTHISIS

LIFE insurance and particularly insurance against sickness
forms one of the most effective means of combating
tuberculosis. The sick-pay received by a member of a
friendly society gives him the means of entering a sanatorium,
and provides his family with food in his absence, assuming
that he is treated without payment. In Germany the system
of insurance against sickness has been developed on an enormous
scale. All wage-earning workmen in Germany have been
compulsorily insured against sickness, employer and workman
contributing to provide an annuity to all persons unable to
support themselves or over seventy years old. ' This insurance
is effected (Bielefeldt, 1901) under the supervision of the Imperial
Insurance Department, State Insurance Departments, thirty-one
insurance institutions territorially limited, and nine special club
institutions of the Invalidity Insurance." These offices and
institutions have a financial interest in postponing invalidity,
as contributions cease when invalidity begins. Hence accurate
investigations of causes of invalidity have been made. The results
up to 1901 showed that of male workers employed in mining,
metal works, factories and the building trades who became
invalided up to the age of 30, more than half suffer from phthisis.
Of persons engaged in forestry and agriculture, who became
pensioners at ages 20-25, 350 out of every 1000 pensioners are
consumptive. Death statistics similarly showed that at ages
15-60 in the German Empire, out of every 100 deaths 33 were
due to phthisis. Hence it was evident that one of the most
important tasks of the officers of the German Workmen's Insur-
ance was to battle successfully against tuberculosis. Obligatory
insurance against sickness has been enforced in Germany since
June 1883 among industrial employees, the sick employee having

372

INSURANCE AND FRIENDLY SOCIETIES 373

the right to free medical attendance and the payment of half
his wage for thirteen weeks, or in the alternative to free treat-
ment in a hospital. In January 1891, insurance against chronic
invalidity and old age was made obligatory ; and six years later
it was found that out of 60,000 pensions given, 8500 were
given to consumptives. Hospital treatment has been made
obligatory in certain cases, and the duration of compulsory treat-
ment has been extended to twenty-six weeks, a fourth of the
patient's wages being paid during this period to his family. If the
patient relinquishes the treatment without good reasons, and thus
incurs the risk of becoming a permanent charge on the pension
funds, the pension may be refused either wholly or partially.
The extent to which sanatorium treatment has been carried out
in Germany is set out on p. 254.

The general system of insurance in Germany has helped to
reduce the death-rate from tuberculosis in three ways : firstly,
patients are able to afford treatment earlier than was formerly
possible ; secondly, the importance of keeping down grants for
sickness and invalidity has led to assiduous education of con-
sumptive patients and of the entire German public in the means
of prevention and cure ; and thirdly, there has been institutional
treatment on an extended scale, and for a much longer period.
A very high proportion of consumptives have been treated in
the general hospitals of Germany both before and since the
sanatorium treatment was introduced (p. 287). Any measure
enabling earlier treatment to be secured by patients, and bringing
home to the general population the importance of hygienic precau-
tions against this disease, must greatly aid in reducing its amount.
It is unlikely that any system on the exact pattern of the
German system will be adopted in this country. The machinery
is complicated and elaborate ; and, in part at least, a rate- or tax-
supported system of medical attendance for those needing it,
on the lines on which " free " education has already been given,
is probably more in accord with our national trend of social
evolution and with our special needs.

Pending any such great national movement as that suggested
by the action of Germany, how can Insurance Societies and
particularly Friendly Societies be utilised in the campaign against
tuberculosis ?

INSURANCE SOCIETIES do their best to eliminate consumptives

374

THE PREVENTION OF TUBERCULOSIS

from the list of the insured by careful inquiries into family and
personal history and by physical examination of the candidate.
That they do not completely succeed is shown by the following
table, taken from Dr. Muirhead's report on the experience of the
Scottish Widows' Fund, 1874-94 :—

Phthisis. — Annual Death-

Ages.

rate per 100,000 Males

living at each group of

20-25.

2S-35-

35-45-

45-55-

55-65.

65-75-

(i) England and Wales,

1881-90

234

304

358

351

292

182

(2) Scottish Widows' Fund

experience, 1874-94

104

163

117

The difference between the insured and the general population
is partly due to the benefits of selection, though average social
condition has also much to do with it. Mr. Hoffman (1901)
has discussed whether, especially in connection with the work
of Industrial Insurance Companies, it would pay to aid those
insured by providing sanatorium treatment, etc., for them. He
points out that the financial interest of the companies is limited to
the increased duration of the policy-life or the increased premium
income in consequence of prolonged life ; and estimating the
prolongation of life by sanatorium treatment at five years, and
taking as the basis of his computation the experience of his
own insurance company among industrial policyholders, he
concludes that the additional income secured by prolonged
life will not provide by increased premiums one-half of the cost
of treatment. In the present state of matters it cannot be
expected that private insurance companies should subscribe
heavily to sanatoria for consumptive persons whose lives are
insured with them. They undoubtedly will gain not only by
sanatorium treatment, but also by improved housing, increased
cleanliness and temperance, the increasing avoidance of pro-
miscuous spitting, and all the measures of hygiene and education
now being pushed forward.

FRIENDLY SOCIETIES are more closely concerned than

INSURANCE AND FRIENDLY SOCIETIES 375

Insurance Companies in the diminution of phthisis, for they
give sickness as well as burial benefits. About fourteen millions
of the population of the United Kingdom belong to such societies,
and more than a million and a half belong to Trade Unions
which have sick benefits, etc. Many more belong to slate clubs
and similar less satisfactory organisations. Mr. J. L. Stead
has collected the experience of the Ancient Order of Foresters,
with the results shown in Table VIII. p. 16.

Some figures collected by Mr. Garland (1905) , based on some-
what scanty data, indicate that the sick pay of consumptive
members costs three times as much as (£14 more than) the
average sick pay to members dying from other causes. The
Friendly Societies are very deeply concerned in reducing the
sickness caused by tuberculosis, and even if ultimately they
do not find it financially advantageous to provide sanatoria
for workers on their own account, they would benefit greatly
by active propaganda against tuberculosis, educating their
members in every possible way, helping in securing the promptest
diagnosis of disease, and in obtaining better conditions of housing
and industrial employment for their members.

An interesting scheme has been launched by the National
Association for the Establishment and Maintenance of Sanatoria
for Workers suffering from Tuberculosis. Mr. Garland and
Dr. T. D. Lister, in a description of the objects of this Association
and of the sanatorium recently opened in connection with it
at Benenden, emphasise the educational aspect of this sanatorium.
By the graduated employment of the patient, they hope to avoid
the demoralisation which occasionally occurs at the convalescent
home and at hospital. They evidently intend the Benenden
Sanatorium to fulfil the functions which the Brighton Sanatorium
has exemplified since 1902 of being " really a training school
for the would-be- well." In their own words —

The palatial building and the liege-halle must give place to the simplest
home-like institution and organised training for the resumption of wage-
earning. If possible, the patients in whom the disease may be believed
to be arrested should be retained in an after-care colony connected with
the sanatorium. Here full work and wage-earning can be resumed
gradually, while yet not entirely out of touch with the medical authorities
of the sanatorium, though not directly under medical control. For the
success of such a scheme propagandist work among all the friendly, labour,
and trade societies affiliated to the movement must be continuouslv

376 THE PREVENTION OF TUBERCULOSIS

pursued, and the co-operation of the medical profession in the selection
of suitable cases must be anxiously sought. The members of all the
affiliated organisations must be taught the means of recognising early
consumption as well as the necessity of seeking treatment before being
completely incapacitated. The importance of the educational value of
a term of residence in a sanatorium is inversely proportional to the magni-
ficence of the buildings and surroundings. Every patient must leave a
working-class sanatorium convinced that there is nothing in the accom-
modation or in the life which he experienced there which is incapable of
being copied in his own simple home. If he be of the fortunate majority
in whom the disease becomes arrested, he must realise how much his
future will depend upon himself, and how much he can do of good to his
fellows by inducing them to live the cleanly, sober, busy, regular life of
a workers' sanatorium.

If the Association succeeds in training those sent to its
sanatorium on the lines here indicated it will be doing admirable
work, with which it is to be hoped that Friendly Societies will
see the advisability of associating themselves.

Meanwhile, apart from the provision of sanatorium treat-
ment, there is much work for Friendly Societies to do in diminish-
ing the present drain on their resources through tuberculosis.
They can ascertain and inform the medical officer of health of
any insanitary circumstances, and particularly of any dusty
occupations to which their members are exposed. They can
start a crusade in every workshop and factory against indiscrimin-
ate expectoration. They can encourage and almost insist on
any of their members who are losing weight or who have per-
sistent cough being thoroughly overhauled, and having their
sputum examined bacteriologically ; and in these and other
ways they can help to the early recognition of disease, to its
treatment while curable, and to the prevention of infection.

CHAPTER XLVII
DISPENSARIES AND THE PREVENTION OF PHTHISIS

SO far we have been concerned with the measures which the
patient himself and his doctor, the different local authorities
of the community in which the patient lives, and friendly and
similar societies can take in the prevention of tuberculosis. Dis-
pensaries and sanatoria may be either municipal or voluntary
in their organisation, and together they hold a high place in the
list of measures against this disease.

The French hygienists have especially developed dispensaries
and the Germans sanatoria as a means of fighting tuberculosis,
and the discussion as to their relative utility has been prolonged
and sometimes heated. Thus Dr. Calmette of Lille, with whose
name the French dispensary system is especially associated,
says that the sanatorium cannot be regarded as a means of
prophylaxis, but only as the one great means of cure. Dr. Savoire
of Paris, speaking on the same point, minimises the importance of
sanatoria because these establishments reject more advanced
cases and only isolate tuberculous patients " at the stage of the
disease in which they are least dangerous." These and other
writers claim that dispensaries, on the contrary, are important
means for combating the spread of the disease. The relative
value of the two can best be discussed dispassionately after
the two institutions have been described.

It is generally agreed that on the Continent Dr. Calmette
first realised completely the ideal of a dispensary which would
be self-contained, not only treating the patients medically,
but watching over their welfare, visiting them at their homes,
j giving them all the necessary hygienic instructions, and providing
material and aid when needed. His dispensary, as described by
MM. Courtois-Suffit and Ch. Laubry (1905), consists of a large
[waiting-hall, two consultation rooms, a dark room for laryngo-
:opic examinations, a laboratory, and an office for the assistant

378 THE PREVENTION OF TUBERCULOSIS

investigator. The chief doctor is assisted by a staff of doctors
and bacteriologists. Their complete medical investigation of
each case is supplemented by a social inquiry entrusted to a
special officer, who visits the home, inquires into urgent needs,
emphasises the hygienic advice already given, and arranges for
supplying cod-liver oil, antiseptics, spit-cups, and, where needed,
food. The dispensary is thus a centre of prophylaxis, thanks to
its educative work, and to the means of disinfection used by it..
Dr. Calmette estimates the cost of an establishment helping
100 families at about 72,000 francs per annum, not including
the cost of installation. The work of the dispensary does not
preclude, of course, the recommendation of suitable early cases
for sanatorium treatment, and the sending of the children of
tuberculous parents to seaside resorts, etc.

The work thus described does not differ materially so far as
the homes of the patients are concerned from that carried out
under an efficient system of notification of phthisis in England.
Such a dispensary as described above does not gather to it all
the patients in a town, and almost certainly not so large a pro-
portion of their total number as are notified in an English town
to the medical officer of health under a fairly successful system
of notification. The preventive measures that can be taken
by a medical officer of health have a wider sweep than those
of the dispensary physician or of his domiciliary visitor. Dis-
infection is better done, sanitary defects can be effectively
remedied, and removal to a suitable institution of patients
housed badly for themselves and their families can more easily
be arranged. The chief point in which the French dispensary
system appears to be better than the English system of voluntary
notification is in the giving of material aid. This under the
English system can be, and is partially in process of being,
remedied by co-operation with voluntary helpers, the Charity
Organisation Society, etc.

TUBERCULOSIS DISPENSARIES IN ENGLAND. — The out-patient
departments of certain British hospitals and certain dispensaries
have for many years past carried on similar work to that of
the French dispensaries, apart from the home visits. Even
these have been arranged at Edinburgh in the pioneer work
of Dr. Philip. The Victoria Dispensary for Consumption was
founded by him in 1887, and, with the exception of the giving

DISPENSARIES 379

of food, etc., to necessitous patients, the method of procedure is
identical with that of Dr. Calmette's dispensary. Dr. Philip
(1906) describes the present arrangements of the Victoria
Dispensary as follows :—

The Victoria Dispensary, as at present arranged, contains —

Two consulting rooms, a laryngoscopic room, one large waiting-room,
two dressing-rooms (male and female), a general office where names are
entered, a laboratory for bacteriological examinations, a drug and food
store.

The dispensary is open thrice weekly for three or four hours.

The staff consists of —

1. Four qualified physicians who attend when the dispensary is open
for the purpose of examining and instructing patients. Three of the
physicians are honorary.

2. One of the medical officers receives a salary of £60 a year, and devotes
a large amount of time to the work. In addition to examining patients
at the institution, along with the honorary physicians, he pays domiciliary
visits to the dwellings of patients in co-operation with the trained nurse.
He makes bacteriological examinations of expectoration and other suspect
discharges. By arrangement with the city authorities, he notifies all
cases of tuberculosis which he meets. He advises regarding the disin-
fection of houses during illness and after the removal or death of the
patient. He supervises treatment of patients at their own home when
this is desirable. He selects suitable patients for the sanatorium. In
co-operation with the city authorities, he drafts the more advanced or
dying patients to a hospital now dedicated to such cases in the neighbour-
hood of the city.

3. A nurse who has been carefully trained in modern open-air methods
at the Royal Victoria Hospital for Consumption, Edinburgh — the sana-
torium in connection with the dispensary — visits the homes of the patients.
She readily wins their confidence by her interest in their welfare. She
instructs the patients, or their friends (wives, mothers, etc.), both as to
treatment and prevention. In co-operation with the visiting physician,
she reports regarding the patient's residence and other conditions according
to the annexed schedule of inquiry. The reports, when completed, are
vouched for by the signature of both doctor and nurse.

SCHEDULE OF INQUIRY REGARDING DISPENSARY PATIENTS
No. in Ledger Date of Report ,

Name Age

Address Married or single ?

Occupation Has patient changed occupation ?

Able to work full time ? Or part time ?

If unable, confined to bed ?

How long ill ?

Situation of house (area, ground floor, first, etc.) ?

Number and ages of inmates ?

380 THE PREVENTION OF TUBERCULOSIS

Number and description of rooms ?
General aspect of house (clean, damp, dusty, smelly) ?
Number of windows ? Can they open ?

Are they kept open (a) by day ?

(&) by night ?

Have they always been kept open ?
Does patient sleep alone (a) in bed ?

(6) in room ?

How is washing of clothes done ?
How long in present house ?

If has moved within two years, previous addresses ?
Have there been illness or deaths in house ?

(a) In own time ?

(ft) In previous occupancy ?
Exposed to infection (a) at home ?
(ft) at work ?
(c) among friends ?

Present health of other members of household ?
What precaution taken to disinfect ?
T. B. in sputum ?
T. B. in dust of room ?

General dietary ? Teetotal ?

General condition (well-to-do, badly off) ?
Proximate income of household ?
Assisted by societies, church, friends, rates ?

Signed Reporter.

Medical Officer.

4. A volunteer Samaritan Committee of ladies, in conference with
the doctors, take charge of more distressing cases, where, through pro-
longed illness, the financial conditions have been much reduced. In
many cases they visit the patients' houses. With the assistance of the
numerous charitable and parochial organisations which exist in the city,
they are enabled to adapt the relief necessary to the particular case.
The members of the Samaritan Committee further occupy themselves
with the question of suitable employment for tuberculous persons fit for
some effort, although unable to work an entire day. In some cases they
arrange likewise for persons who have been discharged from the sana-
torium. Attention is also paid to the case of school children affected
with the disease, so as to have their education supervised on more physio-
logical lines. The operations of the Committee are regulated at fortnightly
meetings, and a minute of the business is kept.

5. An officer — a working-man who gives his entire time to the dis-
pensary— lives on the premises. This man receives and enters the names
of the patients on the afternoons when the dispensary is open. When
the dispensary is not formally open, he attends to requests from patients
or other persons. The officer is conversant with the home and work
conditions of many of the patients, and is a valuable lieutenant both to
the doctors and nurse.

DISPENSARIES

Dr. Philip holds that such a dispensary as the above " should be,
for every city or district, the uniting point of all other agencies."
In the strictly medical sense, this is true. The dispensary is
the receiving-house, the clearing-house for patients. It feeds
the list of official notifications and it enables official preventive
measures to be taken. But it does not act — in this country, at
least — as a complete receiving-house, and is not likely to do so.
A municipal dispensary, and much less a dispensary under the
control of private charity, will not draw to itself all the con-
sumptives needing preventive measures as well as curative
help, though it may be the largest agent to this end. Many
consumptives will remain under the medical care of private
practitioners, of club doctors, of private dispensaries, or in the
out-patient departments of various public hospitals and dis-
pensaries. Under a system of notification of phthisis the
medical officer of health forms the centre from which in a well-
governed community the various measures against phthisis
start and are co-ordinated and made complete. He is almost
certain to know of more cases of phthisis than the physician of
the dispensary, and he has the further advantage that he can
secure for each patient the removal of insanitary conditions of
home and workshop, and the necessary disinfection. He can also
provide handkerchiefs and spit-bottles ; and we hope shortly
will be able in very many towns to arrange for sanatorium
treatment and for the hospital treatment of advanced cases.
The ideal cannot be better stated than in Dr. Philip's words : —

It cannot be too strongly emphasised that the strength of such a scheme
lies especially in its organisation and co-ordination. Each factor is
doubtless of value. Each department has its own sphere of operations.
As isolated elements their possibilities are relatively limited. In pro-
portion as the various departments are intimately connected and co-
ordinated, they each become more serviceable. The key to complete
success in the campaign against consumption lies in the harmonious
co-ordination of well-directed measures.

CHAPTER XLVIII

THE R6LE OF SANATORIA IN THE TREATMENT AND
PREVENTION OF PHTHISIS

A SANATORIUM, as its derivation indicates, is a place
for the cure of disease, in the present connection of
tuberculosis. By Trudeau and others the word is used
to denote also a hospital or asylum for hopeless cases, in which
they can be cared for and treated under conditions preventing
infection to others. There is some convenience in accepting
this wider meaning of the term, in view of the difficulties likely
to be encountered in the future in the institutional treatment
of advanced cases of disease. If these are relegated to a separate
" hospital," they will probably refuse in many instances to
enter ; if only to a separate ward of a " sanatorium," consent
to institutional treatment is much more likely to be secured.

It is not difficult to define the respective r6les of sanatoria
for early cases and for advanced cases of disease. The former
are primarily concerned with the effective arrest, if not actual
cure, of the disease ; the latter with the sympathetic care of
the progressively sick, and with the prevention of infection.
For I quite agree with Dr. Philip's and the general dictum that
" there can be no manner of doubt that the far advanced or
dying cases constitute the greatest source of infection " (see also
pp. 103 and 257). The functions of the two classes of sanatoria
overlap, for the effective arrest of disease in the individual
is an excellent way of stopping infection ; and for this reason,
if for no other, the sanatorium for early cases is also a means
of prophylaxis of great importance. Its importance in this
respect is enhanced by its educational influence. No self-
respecting or even self-regarding patient, after being trained
in a good sanatorium, will continue to spit without due pre-
cautions, and his general life in regard to cleanliness and venti-
lation is likely still further to reduce any possible risk of

THE ROLE OF SANATORIA

383

infection. Hence it is a great mistake to regard sanatoria as
merely cure-places. They are schools of national importance.

OBJECTS OF SANATORIA FOR OTHER THAN ADVANCED CASES.
—i. In early and suitable cases a cure may be expected.

2. Short of cure in a large number of cases, arrest of disease
occurs, the patient possibly continuing to have a small amount
of sputum daily, but being able to resume his work. In a still
larger number of cases, although the disease is not completely
arrested, the patient's condition is improved, his sputum dimin-
ished, he is able to resume his work at least to a modified
extent, and his working life is much prolonged.

3. While the patient is in the sanatorium his home is dis-
infected, his relatives are free from recurring infection and
have time to recover their full measure of resistance to infection.

4. On his return home and to his work the patient is much
less likely than before, even though he continues to have sputum
containing tubercle bacilli, to be a source of infection to others.

Before considering these points in further detail, it will be
well briefly to consider the

HISTORY OF THE OPEN-AIR TREATMENT OF PHTHISIS. — It was
an English village doctor named George Bodington who first
seriously practised the treatment of this disease by what he
called " the natural method/' He described his treatment
in the following words (1840) :—

To live in and breathe freely the open air, without being deterred
by the wind or weather, is one important and essential remedy in arresting
its progress.

The cold is never too severe for the consumptive patient in this climate ;
the cooler the air which passes into the lungs the greater will be the benefit
the patient will derive.

The common hospital in a large town is the most unfit place imagin-
able for consumptive patients, and the treatment generally employed
there very inefficient, arising from the inadequacy of the means at com-
mand.

Dr. Henry MacCormac of Belfast, writing in 1855, emphasised
the value of open windows and cold air in the arrest of phthisis ;
and Sir B. Ward Richardson, writing in 1857, quoted by Dr.
Kelynack (1904), used the following words :—

In a cosy room the consumptive is bound never to live, nor in any
room, indeed, for great lengths of time. So long as he is able to be out
of doors, he is in his best and safest home.

384 THE PREVENTION OF TUBERCULOSIS

Stoves of all kinds, heated pipes, and, in a word, all modes of supplying
artificial warmth, except that by the radiation from an open fire, are,
according to the facts which I have been able to collect, injurious.

If special hospitals for consumptives are to be had, they should be
as little colonies, situated far away from the thickly populated abodes
of men, and so arranged that each patient should have a distinct dwelling-
place for himself. They should be provided with pleasure-grounds of
great extent, in which the patients who could walk about should pass
every possible hour in the day ; and with glass-covered walks overhead,
where the open air could be freely breathed, even if rain were falling.

Opinion gradually grew in favour of an open-air life for
consumptives, but the main impetus to systematic sanatorium
treatment has come from Germany, especially from the methods
employed by Brehmer at Gorbersdorf and by Walther at
Nordrach. Brehmer, who first began to write on the subject
in 1856 and opened his sanatorium in 1859, held that tuber-
culosis was an infectious disease, and, judging by his experience
of the population at Gorbersdorf, that high altitude had an
inhibitory influence against it. Arguing from this experience,
he inferred that anything protecting one person from becoming
a victim to tuberculosis must, if properly employed, be able to
cure another person of the same disease ; and on these lines
he built up his sanatorium treatment, including in it

1. Living in the open air under conditions which appear to
give immunity to tuberculosis.

2. Ensuring freedom from debilitating influences or any-
thing likely to cause recrudescence of disease.

3. Methodical exercises, particularly hill-climbing, when
the patient's condition permitted it.

4. An abundant diet, especially comprising fatty food,
milk, and vegetables.

5. Constant systematic medical supervision, and various
hydro- therapeutic measures.

It is unnecessary to follow the recent history of the evolution
of sanatorium treatment, or the principles embodied in it. In
the words of Dr. F. Rufenacht Walters (1905, p. 41) " the essence
of Brehmer's and Dettweiler's methods is the elimination of
haphazard treatment and the prescription of absolute repose
or of various degrees of exercise according to definite medical
indications."

STRUCTURAL CONDITIONS AND ARRANGEMENTS OF SANATORIA.

THE ROLE OF SANATORIA

385

— A very short summary on this subject must suffice, the reader
being referred for details to Dr. Walters' exhaustive work on
Sanatoria, and to Dr. Latham's Essay on the same subject.
Here we are only concerned with the principles that should
guide local authorities in the matter, and with advice as to the
avoidance of unnecessary expense. Sanatorium treatment can
be carried out successfully in any place where the air is pure,
though a position sheltered on the north and east is preferable.
If the soil is drained and has a slope, it is unnecessary to select
a sandy or other porous soil, though this is preferable when
accessible. The main desiderata as to the site are that

1. The air must be free from dust. Hence nearness to
main roads is inadvisable.

2. Shelter is desirable to the north and east, and there should
be sheltered walks in the grounds.

3. The aspect should be sunny.

The grounds should have shelters suitable for patients to
lie out of doors during a greater part of each day, and the walks
should suffice for graduated exercise.

The arrangements of bedrooms will vary with the class of
patient. It is always desirable to have a number of bedrooms
for single patients, but the exclusive provision of single bed-
rooms in large institutions supported by charity is in my opinion
an extravagant use of charitable gifts. My experience is that
six or even twelve consumptive working-men can be treated
with success in one ward, small rooms being provided for those
whose coughs are particularly troublesome. There is the further
point that in such wards absolutely complete perflation of air
can be secured ; whereas in separate bedrooms as usually
arranged in sanatoria, a corridor is needed opening from each
bedroom door. However well-ventilated is this corridor, it
does not permit as good cross-ventilation as in a hospital ward
of which the two opposite walls are outside walls with windows
between each bed ; and single bedrooms on the plan just men-
tioned are seldom so light and cheerful as a cross-lighted ward.
If there is a verandah outside the single bedroom, the defective
lighting becomes a still greater detriment for acute cases con-
fined to bed.

Of other structural arrangements it is only necessary to
say that they need not be expensive to secure efficient treatment
25

386 THE PREVENTION OF TUBERCULOSIS

of the patient. A linoleum flooring is as sanitary as parquet
and much cheaper. Ledges and corners for dust should be
avoided. Furniture should be simple and free from unnecessary
coverings and hangings. Walls may be covered with a wash-
able distemper. There is much to be said in favour of these
in preference to well-painted cement walls, as the latter favour
the condensation of moisture, and clothes hung in the room
are on humid days cold and damp. Walls and floor and furni-
ture should be cleansed daily with a damp cloth, a broom or
brush only being permitted under special conditions.

Unless in a few special instances for particular purposes,
the cost of construction should be kept down to £200 or £300
per patient to be accommodated. It can seldom be justifiable to
spend £800 to £1200 per bed, as has occasionally been done.

PRINCIPLES OF TREATMENT. — Some of the essential points
have been already indicated, both in dealing with the home treat-
ment of cases (p. 326) and earlier in this chapter (p. 382). In a
sanatorium, treatment is more systematic, the patient is removed
from temptations to depart from the necessary regime, and he
avoids the risks of catarrhal infection and of mental or bodily
fatigue or harass which are apt to occur at home. The atmo-
sphere at the sanatorium is usually purer and freer from dust
than at his home, and the patient gains the advantages associated
with a complete change of environment. Specific treatment by
tuberculin, controlled by opsonic testing, is more easily managed
at a sanatorium than at home. Hygienic rules can be more
easily enforced, rest in bed can be controlled in accordance with
exact observations of the patient's temperature and other
conditions ; and, where the appetite is deficient, the more or less
forced feeding which is an important part of sanatorium treatment
can be efficiently carried out. Although too rapid accumulation
of fat is undesirable, the indication is to press feeding sufficiently
to ensure in non-febrile cases a weekly gain of weight of at least
i lb., better 2 to 3 Ib. (Walters), "until the natural full weight
is reached, and to ensure this being maintained afterwards."
The patient can often digest large quantities of meat, even when
he is feverish. Many feverish patients begin at once to improve
as soon as they sleep out of doors, or at least stay out during the
entire day. Complete rest and open-air life give the best prospect
of reducing the fever of acute phthisis. The regulation of

THE ROLE OF SANATORIA

387

amount of exercise is one of the most important duties of
the sanatorium physician, and it is on this point that the
superiority of sanatorium over home treatment is most evident.
As Dr. Latham (1906) remarks : " What the patient learns at
a sanatorium, and only at a sanatorium, is the fact that fatigue
kills the majority of consumptives and causes the frequent
relapses of the disease. The avoidance of fatigue is therefore of
primary importance." This leads to the consideration of the
chief practical objection urged against sanatorium treatment for
working-men. The problem for them is a serious one. As
frequently sent out from sanatoriums they are much improved
in health, but their muscles are soft, and they are unable to
bear the normal fatigue associated with their daily work. Even
when able, they are often unwilling. Dr. Walters (1906) may
be quoted here : —

It is justly argued that prolonged idleness is apt to foster lazy habits
and to make the patient less capable of steady work. The remedy for
this is to substitute other forms of useful occupation as soon as the patients
are fit for it. Hard manual labour is unsuitable for something like two
years after the breakdown, but many forms of light work are permissible
as a rule, such as hoeing, raking, sweeping, pruning, poultry feeding,
chopping up thin pieces of food, and some of the work in which hand
machinery is used. The spare time should, however, be chiefly employed
in education. At Dr. Weicker's sanatorium for artisans in Silesia and
in some others the patients have regular courses of instruction in short-
hand, foreign languages, cooking, and the like. Many of the applica-
tions of science and art to manufacture would also be permissible, such
as designing, photography, the reproduction of designs, some methods
of decorating pottery, and some of the applications of microscopy and
chemistry. A conference of medical men with technical instructors
in various branches of handicraft would probably bring to light many
useful occupations open to convalescent consumptives. The chief point
to bear in mind would be the substitution of delicate for laborious work,
brains for brawn. That hygienic teaching bearing upon the disease
itself would be given is taken for granted ; but the addition of suitable
technical teaching would make the sanatorium a valuable educational
centre, would add to the happiness and usefulness of the inmates, and
greatly diminish the difficulty in finding work for discharged patients.

Short of the change 01 occupation wisely advocated above,
wherever practicable, much can be done for the industrial patient
while in the sanatorium to prepare him to return to his own
work. On this point I will quote somewhat fully Dr. Kingston
Fowler's (1906) description of the methods adopted at Frimley,

388 THE PREVENTION OF TUBERCULOSIS

the Brompton Hospital Sanatorium, which have been organised
and successfully carried out under the care of Dr. M. S. Paterson,
the medical superintendent : —

Each batch of patients on arrival from the parent hospital at Brompton
— through which they must all pass — is addressed by the medical super-
intendent on (i) discipline, (2) fresh air, and (3) feeding. As they have
already been trained at Brompton for the lesson they have to learn, they
find but little difficulty in falling in with the more complete open-air
life followed at Frimley. It was, however, not an easy task to establish
the tradition of absolute obedience to orders which now prevails ; the
conviction as to the wisdom of the regulations came to the patients as
they found themselves steadily improving in health and strength. Now
everyone cheerfully goes about his appointed exercise or work irrespective
of the weather, and if told off to roll the lawn for two hours he does it,
and is not found after five minutes sitting upon the handle of the
roller.

As an illustration of the thoroughness of the treatment, so far as " open
air " is concerned, I may state that the desire of the majority of the
patients whose bedrooms on the upper floor are without a balcony is
to be promoted to a room with a balcony, or to one on the ground floor,
so that they may be able to pull out their beds and sleep in the open
air. I was told when at Frimley in December 1905 that most of the
patients at that time slept in the open air when it was not raining. During
the recent frosty weather the patients were told that they could close
their windows for an hour whilst they were dressing, but it was found
that none of the windows were closed. Hats and caps are not worn except
when walking outside the grounds. The appetite developed by an open-
air life is surprising ; as most of the staff voluntarily lead the same life,
they experience a similar increase of appetite.

Daily Routine. — 6.50 a.m. : Rise and turn down beds and proceed
according to "Morning Routine." 8.15 a.m.: Breakfast for tables
i,2,and3. 8. 30 a.m. : Breakfast for tables 4, 5, and 6. 9.30 to 9.55 a.m. :
Indoor work. 10 a.m. : Outdoor work or exercise. 10.50 a.m. : Lunch,
ii a.m.: Outdoor work or exercise. 12 to 12.45 p.m.: Absolute rest
for tables i, 2, and 3. 12 to i p.m. : Absolute rest for tables 4, 5, and 6.
i p.m. : Dinner for tables i, 2, and 3. 1.15 p.m. : Dinner for tables 4,
5, and 6. 2 to 2.45 p.m. : Absolute rest for tables i, 2, and 3. 2.15
to 2.45 p.m. : Absolute rest for tables 4, 5, and 6. 2.45 to 4.35 p.m. :
Work or exercise in grounds. 5 p.m. : Tea for tables i, 2, and 3. 5.15
p.m. : Tea for tables 4, 5, and 6. 5.50 p.m. : Temperatures taken for
tables i, 2, and 3. 6.5 p.m. : Temperatures taken for tables 4, 5, and 6.
6 to 7.45 p.m. : Read papers, write letters, play indoor games, etc.
7.45 p.m. : Supper for tables i, 2, and 3. 8 p.m. : Supper for tables
4, 5, and 6. 8.40 p.m.: Prayers. 8.45 p.m.: Bed. 9.15 p.m.: Lights
out. 9.30 p.m. : Silence.

A quarter of an hour is allowed for smoking after each meal. A
quarter of an hour is allowed before each meal for washing. Patients
are not allowed indoors except for meals and rest hours until 6 p.m. without

THE ROLE OF SANATORIA

389

special permission. Patients may use the concert-hall and reading-
room from 6 p.m. until prayers.

Sunday Routine. — The routine is the same, with the following differ-
ences : There is no work. 9.30 to 10.35 a.m. : Patients walk two miles
in all weathers. 1 1 a.m. : Divine service. 1 2 noon : Rest hour.
2.30 p.m. : Those patients who have permission may walk outside the
sanatorium until 4.45 p.m.

The patient's day is thus so completely occupied that he has little
leisure for introspection, and I am informed that the only common com-
plaint is, " We are kept so busy we have no time for anything."

Graduated Labour. — The new feature which Dr. Paterson has intro-
duced at Frimley is graduated labour, a feature which appears to me
to go far to solve the question as to the applicability of sanatorium
treatment to the poorer classes. County authorities and the public are
naturally asking : " Are the patients whom you call ' cured ' able to work
and earn their own living ? " (I deprecate the use of the word " cure,"
but the public will have it so.) Upon the answer which we are able
to give to this question the provision of adequate sanatorium accom-
modation for the poor depends. I believe we can state that the patients
classed as " arrested " after treatment at Frimley are fit for work.

The gradation of exercise and labour is as follows : Exercise and
labour are for two periods daily, each of two hours' duration, (i) Slow
walking exercise, beginning at two miles a day and gradually increasing
up to ten miles a day. (2) Picking up fir cones and firewood in the grounds
and carrying a " half-basket " (weight 1 1 pounds) to the stack. (3) Carrying
a full basket of firewood and cones (weight 16 pounds). (4) Carrying
a " half -basket " of gravel or stones from the gravel pit to the place
where paths are being made or repaired (weight 21 pounds). (5) Carrying
a basket of gravel or stones, the weight of which is gradually increased
up to 38 pounds. (6) Rolling the grass or gravel. Sixteen men pull a
roller weighing 1 5 cwt. (7) Digging ground already broken. (8) Mowing
grass with a lawn mower. (9) Digging unbroken ground. (10) The same
as under (9) but for six hours daily instead of four hours — i.e. the hours
usually spent at rest are spent in labour.

The indications accepted as evidence of the arrest of the disease are :
(i) absence of fever ; (2) absence of adventitious sounds, except such
as are indicative of fibrosis ; (3) absence of cough and expectoration ;
(4) continuous gain of weight or maintenance of the patient's highest
known weight ; and (5) ability to perform labour incidental to grade
No. 9 as above.

The point to which I wish especially to draw attention is that no
patient is classified on discharge as " arrested " unless for three weeks
continuously he can pass one or other of the following tests : —

Test A. — For patients who earn their living by manual labour : To
be able on an ordinary diet and without rest hours to use a pick and
shovel of the full size and weight for six hours daily and to maintain
his health. The shovels and spades are in three sizes, weighing 2, 4,
and 6 pounds respectively. The picks vary from 3 to 7 pounds in weight.

Test B. — For patients who do not earn their living by manual labour,

390 THE PREVENTION OF TUBERCULOSIS

e.g. clerks, shopmen, or salesmen : To be able on an ordinary diet to
perform the labour of grade No. 6 or No. 7 for six hours daily for three
weeks and to maintain his health. These patients are, as a rule, gradually
brought up to No. 9, and when it is found that they can do this work,
they are put back to No. 6 or No. 7. The theory is that a man doing
the work described under No. 9 or No. 10 who on discharge will engage
in work involving but little bodily exercise, would suffer in health from
such an abrupt transition. Further experience is, however, necessary
upon this point. In some cases it is found that patients are unfit for
No. 9 but that they can be raised to a standard of labour which is equal
to their ordinary work. These patients are tested before discharge
on the grade to which they have attained, but they are not, as a rule,
classified as " arrested."

The system has been gradually evolved and has not yet been in opera-
tion for a sufficient time to justify the expression of a final opinion as
to its value, but there appears to be every reason for anticipating that
it will prove successful.

MEDICAL RESULTS OF SANATORIUM TREATMENT. — After care-
ful consideration, I have decided not to utilise any of the many
published statistics as to sanatorium treatment. So much
depends upon accurate diagnosis, upon accurate tabulation of
figures, and upon the lapse of a sufficiently long and uniform
interval before results are tabulated, that I doubt if many of the
published figures can be trusted for comparative purposes.

I am completely convinced that the sanatorium treatment
is most beneficial to patients, and enables a large proportion of
them to resume their ordinary life. This is true even for cases
in which there is consolidation, and occasionally also for cases
with considerable cavitation of lungs. Although similar cures
occur apart from sanatorium treatment, clinical experience
indicates that they are more frequent and occur earlier under
sanatorium treatment, and I have no doubt that were exactly
comparable data available, this would be found to be so. As
Professor v. Ziemssen, quoted by Dr. Walters, says : —

The possibility of treatment outside a sanatorium with equally good
results cannot be denied, but it requires much more prolonged rest and
much more time on the part of the physician, and has by no means so
certain a result.

The general results of sanatorium treatment have been well
summed up by Dr. J. E. Squire as follows :—

I. It can, he says, be " reasonably expected that of the cases
of pulmonary tuberculosis which are recognised sufficiently

THE ROLE OF SANATORIA 391

early and commence sanatorium treatment without delay, some
may be cured and return to work in three months."

2. Three months' treatment being rarely sufficient for the
stage in which " early " cases are generally admitted to the
sanatorium, " we are justified in stating that early cases of
pulmonary tuberculosis may be expected to recover under
sanatorium treatment if persisted in sufficiently long," but six
or even twelve months may be required.

3. There is a further justifiable expectation that by " sana-
torium treatment, even in acute and somewhat advanced cases,
arrest may be anticipated provided the patient is able to continue
the treatment sufficiently long." This generally means at least
twelve months' treatment and a further period under supervision
before " cure " can be spoken of.

CLASS OF PATIENTS SUITABLE FOR SANATORIUM TREATMENT.—
The great desire of all physicians at sanatoria is to secure patients
at an early stage of disease, and their general lament is that this
desire is not achieved. Not all the cases with physical signs of
early disease do better than cases of disease of longer standing,
much depending on the acuteness and febrile reaction of the
patient. The three months usually allowed for sanatorium
treatment often does not suffice for cure or arrest of disease.
The choice of patients in most sanatoria is made from the point
of view of the individual. Can the disease be arrested or not ?
is the question asked from this side. It is not identical with the
view of the public health administrator, whose question in relation
to sanatoria is, By the sanatorium treatment of what patients,
and of these for what length of time, can I secure the greatest
amount of prevention of infection ? This question is sufficiently
important to be dealt with in a separate chapter. Meanwhile,
we may add here a few words as to the training of sanatorium
patients, and as to their after-care.

THE TRAINING OF SANATORIUM PATIENTS. — An important
element in the treatment of each patient is that he should know
the nature of his disease, and should receive exact instructions
as to the hygienic precautions necessary for aiding his cure,
for preventing relapse, and for obviating infection. Whatever
differences of opinion there may be as to the economic gain of
the sanatorium treatment of wage-earning patients, there can
be none as to the great gain to the community secured by this

392 THE PREVENTION OF TUBERCULOSIS

training. The principles of it are sufficiently obvious, and they
have been stated on pp. 348 and 357. The following card is
given to each patient leaving the Brighton Sanatorium : —

ADVICE TO PATIENTS LEAVING THE SANATORIUM

1. The spit-bottle should always be carried in the pocket,
and daily washed out with boiling water after emptying its
contents down theW.C. At home, if the bottle is not used, spit
into paper or rag, and burn this at once.

2 . Be careful not to cough directly opposite to any other person .
Always hold a handkerchief to your mouth when coughing.
Change your handkerchief every day, and put the soiled one into
water.

3. In order to maintain a condition of good nourishment,
take a glass of milk with each of the three chief meals, in addition
to the ordinary food.

4. Keep on taking cod liver oil each day until you have no
cough, unless otherwise ordered by your doctor.

5. Do not take beer or other alcoholic drinks. Money thus
spent is wasted.

6. Keep up the practice of sleeping with your bedroom door
and window wide open. One of these without the other does not
suffice. To keep warm, wear plenty of woollen clothes.

7. It is imperative that you should sleep in a separate bed,
and if possible have a separate bedroom.

8. Do not run the risk of inhaling dust if you can avoid it,
either in the house, or when at work, or in the street. Always
insist on the " wet cleansing " of rooms, instead of dry dusting or
sweeping.

THE AFTER-CARE OF SANATORIUM PATIENTS. — The per-
manence of cure or of arrest of disease depends greatly on the
training which the patient has received while in the sanatorium,
and his intelligence and assiduity in living up to it. Ofttimes,
however intelligent and willing he may be, he cannot live the
life best calculated to maintain his ground. He is obliged, for
instance, to return to hard manual labour in a dusty workshop.
The general considerations applying in this matter are stated on
p. 327. If alongside these considerations be placed those quoted

THE ROLE OF SANATORIA 393

from Dr. Walters on p. 387, we have a statement of possible
alternatives, of which the resumption of previous work most
frequently occurs. The difficulty as to subsequent occupation
is even greater for patients whose expectoration continues,
often fairly abundant, but who have before them several years in
which they are still able to work. For these the month's sana-
torium training mentioned on p. 395 is particularly indicated.
After this, what is to be done with them ?

INDUSTRIAL COLONIES have been advocated for them.
During the patient's stay in the sanatorium itself, something
may be done in this direction, as indicated on pp. 387-390, and
the sanatorium may be arranged so as to merge into the industrial
colony. There is little doubt that a year's life on a farm or farm
colony after leaving the sanatorium would in many instances
which now soon relapse mean permanent recovery. There are,
however, difficulties which prevent one from being very sanguine
in regard to them. Dr. Jane Walker (1906, p. 365) draws
attention to three of these : the patients are mostly town-dwellers,
they are often married men, and they have generally learnt a
trade, and will not therefore make up their minds to take the
wages of an agricultural labourer. The subsequent development
of schemes in this direction will be watched with interest, but it
cannot be said at present that the establishment of such colonies
otherwise than by private charity is to be recommended.

CHAPTER XLIX

THE INSTITUTIONAL TREATMENT OF PHTHISIS FROM
THE PUBLIC HEALTH STANDPOINT

THE subject of this chapter necessarily traverses ground
already partially covered in previous chapters. It is
desirable, however, to summarise from the standpoint of
public health administration the question of the institutional
treatment of phthisis ; and this chapter may be regarded there-
fore as an annexe to Chapter XLVIIL, as well as an attempt at
the practical application of the argument of Part. II.

Three classes of consumptive patients need to be considered :
first, those in an early and probably curable stage ; second, those
who, though showing marked disease, are still able to work
either continuously or with intervals of inability, and who
are likely to have several further years of life, whether treated
or untreated ; and third, advanced cases, unable to work,
commonly confined to the house except in warm weather, and
often bedridden.

Which of these is most dangerous to the public health ?
Reasons have been already given for the view that the advanced
cases do most harm ; for not only are they unable to control
so perfectly the disposal of their more abundant sputum, but
they require that intimate and protracted personal attention
which in the ordinary circumstances of domestic life among
the poor especially favours infection. Against this is to be
set the fact that the early and the intermediate patients
have a wider field for scattering infection. The balance of
evidence is nevertheless strongly against their being the chief
source of infection. Whatever view be taken on this point,
evidently the wise course is to ensure the due disposal of ex-
pectoration by each of the three classes of patient. The training
of the early patient, when it can be secured, holds good during
a longer period of infectivity than that of the intermediate or

394

INSTITUTIONAL TREATMENT OF PHTHISIS 395

advanced patient. Hence it should be the rule to ensure the train-
ing of consumptive patients from the earliest practicable period.

SANATORIUM TRAINING OF EARLY AND INTERMEDIATE CASES.
—Early experience of notified cases of phthisis showed me —
what has been confirmed by later experience — that even when
I had given definite instructions, both verbally and printed
(see p. 324), as to care in spitting, on subsequent visits it was not
infrequently found that these were not being effectually followed.
Sometimes the instructions had been misunderstood, more
often they had been neglected. The patient's self-interest as
well as his conscience needs to be utilised. If he can be taught
heartily to believe that his own welfare and that of his family
is favoured by the precautionary measures recommended to
him, we may usually rely on his co-operation. How to secure
this educational influence became, then, an important question
early in my local experience of the notification of phthisis. The
plan eventually adopted — the success of which in this respect
has exceeded my anticipations — was the treatment on open-
air principles of all patients who could be persuaded to consent
to such treatment. This was carried out in a detached pavilion
of our hospital for acute infectious diseases which is locally
known as the sanatorium. The difficulty in getting patients
to come into the sanatorium was greatly diminished by the
fact that only very short terms of treatment were proposed,
which could in most instances be managed, without the patient
risking loss of his livelihood. The Fig. on p. 341 shows how
greatly the number of cases of phthisis voluntarily notified
in Brighton has increased since sanatorium treatment became
available. The details of the system adopted in Brighton
have been regarded with considerable interest, and I therefore
give here certain fuller particulars which may be of assistance
in other towns.

MUNICIPAL SANATORIUM TRAINING AND TREATMENT AT
BRIGHTON. — The earlier details of our local efforts at sana-
torium treatment are stated on p. 348 in their relation to the
notification of cases. Further details will now be given. The
first point aimed at was to avoid any new capital expenditure
on buildings ; and in order to do so, to utilise an empty pavilion
of our present isolation or fever hospital. Epidemics of scarlet
fever and diphtheria are intermittent, and of enteric fever are-

396 THE PREVENTION OF TUBERCULOSIS

very rare ; and yet hospital accommodation in most communities
is kept ready for the contingency of their occurrence. This
accommodation it was proposed to utilise for phthisis patients ;
and there did not seem to be any serious difficulty in doing so,
as, with the possible exception of an occasional milk outbreak
of one of the above acute diseases, plans can be made for several
weeks ahead, and phthisical patients can easily be sent home
when necessary. Events have proved this forecast correct.
Not only has it been unnecessary to cease treating consumptives
at the hospital up to the present time, but we have been able
to increase our beds for this disease from four to ten and then to
twenty-five. This increase is in part owing to a charitable bequest
(the Hedgcock Bequest), which enabled the Town Council to devote
a yearly income from this source of £600 to £700 to the endow-
ment of further beds. This fund enabled the number of beds
for the use of consumptives to be increased from ten to twenty-
five, including three beds for paying patients, twelve to be
maintained by the Hedgcock Bequest, and ten provided directly
by the Town Council. The Town Council provides the entire
accommodation for these twenty-five patients in its isolation
hospital.

The directly municipal patients are usually admitted for a
month each, and are by preference men and women still able
to work, and in connection with whom a month's rest, treat-
ment, and training, can effect the greatest good to the patient
and to others in preventing infection, both of fellow-workers
and of family. No charge is made for the admission of these
patients, who are chiefly labourers, artisans, clerks, etc., and
their relatives.

The Hedgcock patients belong to the same classes. They
must be unable to pay for their own maintenance in the sana-
torium. Some of them are very advanced, or even dying
cases, for whom continuance at home is undesirable owing to
difficulties as to nursing, or because there is a large family
and much danger of infection. Where practicable, advanced
cases are treated in separate rooms. It is not, in my opinion,
necessary to have a separate institution for them ; and the
objection mentioned on p. 382 is strongly against this. Hedgcock
patients are kept in the sanatorium for several months or for
a shorter time, according to individual requirements.

INSTITUTIONAL TREATMENT OF PHTHISIS 397

THE METHOD OF USING ISOLATION HOSPITAL BEDS1
(i) Accommodation available

The isolation hospital consists of four main pavilions for
infectious cases — an administrative block, the borough dis-
infecting station, a laundry, and a small destructor. Three of
the main hospital pavilions were originally used for scarlet fever,
diphtheria, and enteric fever, and the fourth for cases needing
special isolation.

In the scarlet fever pavilion (two storeys) 68 beds.
,, diphtheria fever pavilion „ 56 „

„ enteric „ „ . 22 „

,, isolation „ 14 „

Total 160 „

The population of Brighton estimated to the middle of 1907
was 129,023, the proportion of beds to population being about
I to 800.

FIG. 39. — Block Plan of Isolation Hospital.

A. Discharge Room ; B. Porter's Lodge ; C. Administrative Block ;
D. Isolation, Pavilion ; E. Diphtheria Pavilion ; F. Phthisis Pavilion ;
G. Laundry and Disinfecting Station ; H. Scarlet Fever Pavilion ;
I. J. Phthisis Shelters

(2) Isolation of the Consumptive Patients from other Diseases

Visitors from other towns frequently ask the question : " Do
the phthisical patients run any risk of contracting the infectious
diseases treated in the hospital ? " The answer is that the
possibility of the spread of infection depends on the standard
of administration, and that an experience of six years shows a

1 The following particulars are taken from a joint paper with Dr. H. C.
Lecky published in Tuberculosis, June 1907.

398 THE PREVENTION OF TUBERCULOSIS

complete absence of such infection. Infection might be spread
in any of the following ways : (a) By contact between patients ;
(b) by the carriage of infection by nurses, or (c) by the doctors ;
(d) by infection from the laundry or kitchen.

(a) Contact between patients in different pavilions. — It being
impossible completely to shut off one portion of the grounds
from another, the keeping of the prescribed bounds depends
upon the supervision by nurses of children and on the honour
of patients who have reached years of discretion. Consumptive
patients are as desirous not to contract another disease as the
doctor is to prevent it, and patients suffering from diphtheria
and scarlet fever are under the strictest supervision. In
practice, therefore, this difficulty scarcely arises, and the erection
of impassable barriers between areas allotted to the different
diseases is found to be unnecessary.

(b) Infection by nurses. — It is customary in isolation hospitals
for the nurses from the various wards for acute infectious
diseases to have their meals in a common dining-room in the
administrative building. In my experience infection has never
been caused by the adoption of this plan. The experience of
other isolation hospitals is to the same effect.

The nurses for the consumptive wards use a separate table
in the dining-room, and sleep in separate rooms on the first
floor of the administrative building. All other nurses dine
at another table in the same room. The nurses for diphtheria
sleep on the second floor of the administrative building, and
those for scarlet fever sleep in the dormitories over the scarlet
fever pavilion with a separate means of access. The nurses
for different diseases are allowed to go out together, and they
occasionally use a common sitting-room.

To enable scarlet fever and diphtheria to be intercommuni-
cated under the above circumstances by the nurses attending
these diseases, infection would need to pass through two inter-
mediaries— a highly improbable event. If infection does not
spread under these circumstances from scarlet fever to diphtheria,
or conversely, it is unreasonable to expect that it would spread
from either of these to consumptive patients, and our confidence
in this anticipation has been justified by events.

(c) Infection by the doctor. — The precautions adopted are
those which every careful practitioner adopts? in his everyday

INSTITUTIONAL TREATMENT OF PHTHISIS 399

rounds. The consumptive patients are visited first, and overalls
are used when going into the other wards.

(d) Infection from the laundry. — The washing from the whole
hospital is done in one common laundry. Special precautions
are taken with the soiled linen from the scarlet fever and
diphtheria pavilions, articles only being sorted after having
been in soak for a certain time. A definite routine is main-
tained, so that when the linen has once been washed no soiled
linen is taken into the laundry during the same week. The
chances, therefore, of spread of infection in this laundry are
less than in an ordinary general laundry, and infection, in fact,
has not occurred.

(e) Infection from the kitchen. — The food for all the wards is
distributed from a central kitchen. Every article to be returned
from the various wards is washed first. No food is ever returned.

The above summary of our procedure shows that no risk is
involved in the treatment of consumptives in a well-administered
hospital, in pavilions properly separated from those for scarlet
fever and diphtheria. Experience has justified the advice
given as to the d priori improbability of such spread, for during
the last six years, in which 730 consumptives have been treated
for an average period of five weeks for each patient, not a
single case of an acute infectious disease has occurred among
these patients.

(3) The Principles on which Beds in the Sanatorium are

allocated

Not every patient notified to be suffering from phthisis is
offered treatment at the sanatorium. Since the average time
that the patients can afford to stay is from four to six weeks,
the main factor determining the admission of patients to other
hospitals and sanatoria, namely, the possibility of permanent
benefit or cure, obviously is the factor of least importance in
deciding as to the admission of patients to our sanatorium.
The benefit to be derived from the short treatment of patients
has been summarised on p. 349. From the public standpoint
it may be summed up in the word education or training : (a) The
patient is taught that he is in part responsible for his own cure,
and he is shown the best way of living with this end iii view ;
(b) he is trained so to manage his cough and expectoration that

400 THE PREVENTION OF TUBERCULOSIS

he is no longer a source of infection to others. These being the
chief objects at present attempted, each of the following circum-
stances is taken into account in considering the suitability of
cases for admission : —

(a) The age of the patient. — People at the working years of
life are those who can derive the greatest benefit from the sana-
torium treatment and training. Children, whose home circum-
stances are in the hands of others, obviously cannot carry out
a given line of treatment of their own accord. Furthermore,
children are seldom sources of infection to others, owing to the
absence of expectoration. Old people suffering from phthisis
frequently drift to the workhouse infirmary, and every effort is
made to facilitate their admission to this institution, though in
the event of their not coming within the legal limits of the poor
law they are admitted to the sanatorium if they are likely
sources of infection.

(b) The size of the family. — If a family consists of a mother
and father and several children, and one of the parents has been
notified, every inducement is offered to get the patient into the
sanatorium. If, at the same time, the cases of a parent and one
of the children have been notified, an endeavour is made to get
them into the sanatorium together. On several occasions two or
more members of the same family have been treated at the same
time. If the family consists only of a married man and his wife,
past middle age, and one of them is notified, there is less necessity
to urge sanatorium treatment than if other and younger people
are living with them.

(c) The occupation. — This is an important factor. Pre-
ference is always given to consumptives working in factories
or workshops with a large number of other men or women.

(d) The stage of the disease. — As mentioned above, this factor
by itself is of minor importance in determining the suitability
of notified cases for admission. It is of extreme urgency to
educate the young adult, especially if he is a bread-winner and
a parent, both from the standpoint of cure and of prevention
of infection. Patients with advanced disease are admitted as
readily as patients having earlier disease, the one condition of
admission being that the possibilities of infection can be reduced
by the training of the patient.

(e) The social position of the patient. — Under our present

INSTITUTIONAL TREATMENT OF PHTHISIS 401

system of voluntary notification information is rarely received
of cases where the family has an income of more than -£2 a week.
Yet, although there is a great difference between the positions
of a family with an income of 353. and one with an income of
253., the need for sanatorium treatment is almost as urgent
for the one class as for the other, and no social distinction is
therefore drawn in admitting patients. The only partial ex-
ception to this rule is in regard to patients who come within
the purview of the poor law. If these patients are possibly
curable they are admitted to the sanatorium. If their disease
is advanced they are urged to go into the Workhouse Infirmary.
The arrangements in the thirty beds of that institution reserved
for phthisis are good, and patients who would otherwise be a
source of serious domestic infection are well segregated in these
beds.

It will thus be seen that the suitability of a patient for
admission to the sanatorium depends on the answer to the
following questions : (i) " Will the treatment begun at the
sanatorium, if subsequently continued, give a reasonable chance
of a cure ? " (2) " Even if there is no reasonable chance of a
cure, will the treatment and training diminish and possibly
prevent the spread of infection to others when the patient leaves
the sanatorium ? "

The preceding sketch of local arrangements is given in full
not as representing an ideal, but as an illustration of what can,
in many districts, be done without expenditure on new build-
ings. In other districts, if the isolation hospital accommodation
is insufficient, new buildings will be required. It is, however,
most desirable that local authorities should not unnecessarily
incur heavy capital expenditure, when by possible adaptation
of already available accommodation the interest on the same
money might be utilised for the actual treatment of further
patients. It is possible that in a few years interchange of accom-
modation for consumptives may be possible between the public
health and the parochial authorities. If the parochial regula-
tions could be relaxed for the sick, there is in many workhouse
infirmaries excellent accommodation for

ADVANCED CONSUMPTIVES WHO ARE NOT PAUPERS. — The pro-
vision of accommodation for the patients of this class is the most
urgent problem in the prevention of tuberculosis. The way

• 26

402 THE PREVENTION OF TUBERCULOSIS

to this provision in most districts will probably lie through the
removal of parochial restrictions, and the consequent increase
of popularity of the consumptive wards of the infirmary. This
question is dealt with to some extent on p. 394. There can be
no doubt, as stated in the admirable circular issued by the Local
Government Board of Scotland (March 1906) on the " Adminis-
trative Control of Pulmonary Phthisis," that " the isolation of
such dangerous cases is a primary duty of the local authority."
The view taken on p. 382 is that these cases may properly
be treated, though in a separate ward, in the same institution
as earlier cases of phthisis. The removal of parochial restric-
tions in respect of the treatment of the sick, it may be hoped,
will ere long remove the chief difficulty in successfully coping
with this problem.

The following estimate by Dr. Rushton Parker gives some
guidance as to the possible expense involved in the further
provision of hospital beds for advanced cases of phthisis : —

As two-thirds (or, strictly, 70 per cent.) of any population usually
belongs to the working class, and as during the last ten years there have
been about 42,000 deaths annually from consumption in England and
Wales, we may assume that 28,000 persons will annually qualify for
admission into such homes. At those which already exist the applica-
tions for admission far exceed the vacancies ; the duration of stay is
about six months ; and the annual cost of maintenance is about £65
per bed. We may assume, therefore, that we shall require 14,000 beds,
at an annual cost of £1,000,000 a year. About one-sixth of the cases
would be paupers ; so that one-sixth of the cost would be chargeable
to the guardians. As it has been calculated that one-eleventh of all
the pauperism of the country, costing in England and Wales £11,500,000
a year (1900-1901), arises from consumption, the million pounds a year
proposed to be so spent should produce much more profitable results
than the million pounds a year already spent in merely relieving the
pauperism caused by neglected consumption.

In every population of 100,000, about 120 die annually of consump-
tion, of whom 80 require accommodation in a home of 40 beds, at a cost
of £2600 a year, roughly equivalent to a penny rate for such population .

CHAPTER L

THE PREVENTION OF TUBERCULOSIS DUE TO
INFECTED FOOD

THE degree of danger from the flesh of tuberculous animals
has been already indicated, and it has been seen that
on present evidence it is much smaller than that from
milk and its products. Both these dangers might conceivably
be removed by action along one or other of the following
lines : —

1. The extermination of tuberculous cattle and of other
tuberculous animals used for food.

2. The prevention — apart from their complete extermina-
tion— of the use of such animals or their products as human
food.

3. The sterilisation of food derived from tuberculous
animals.

The first of these lines of action is not within the range
of immediate practical policy. The Legislature could not be
expected to undertake the enormous initial expense of the
destruction of all animals found by means of tuberculin testing
to be diseased. Short of such wholesale condemnation of
diseased cattle, more stringent regulations are undoubtedly
indicated, and there is much room for better enforcement of
already existing regulations. Thus at the present time it is
punishable to sell milk derived from cows suffering from tuber-
culosis of the udder ; but this power is at present in the
hands of authorities who are usually rural authorities, of whose
members farmers form a large proportion. If the administra-
tion of the powers relating to this disease were in the hands of,
or powers of action in default were given to, larger authorities,
they would be more likely to be enforced. It is desirable also
to increase the power of such authorities, enabling them to

test by means of tuberculin if necessary any cow showing

403

404 THE PREVENTION OF TUBERCULOSIS

symptoms suspicious of tuberculosis, whether in the udder or
not. Further power is needed to prevent the same cow from
being used for feeding calves or passed on to another farm,
after its milk has been stopped on the farm where the disease
was first discovered. At present the farmer can evade the
results of this discovery, by selling the cow in question. Some
unobjectionable method of marking such cattle permanently
would be useful in preventing this traffic. Compulsory slaughter
is indicated in some cases. Whether limited fractional compen-
sation should be given in such cases may be left open for con-
sideration. It is difficult to devise a local scheme for such
compensation which would work equitably.

Apart from specific action in respect of tuberculosis in cattle,
much could be done by improved sanitation in cowsheds to
diminish the amount of infection from cow to cow.

MEAT FROM TUBERCULOUS CATTLE. — The evidence connect-
ing tuberculous meat with the possibility of infecting man has
already been considered (p. 140). In the words of the First
Royal Commission (par. 22 of their report, April 1895), " any
person who takes tuberculous matter into the body as food
incurs some risk of acquiring tuberculous disease." The cooking
of meat affords a considerable measure of protection, as all
except under-done parts would be sufficiently sterilised. With
uncooked meat; which is often given in the form of pounded
meat or meat juice to weakly children, there must be considerable
risk ; and doctors prescribing such meat should give preference
to meat derived from animals known to have been slaughtered
at a public abattoir.

The second Royal Commission on the same subject (1898)
laid down the following principles in the inspection of the
tuberculous carcasses of cattle : —

(a) When there is miliary tuberculosis of

both lungs,
(6) When tuberculous lesions are present on

the pleura and peritoneum,
(c) When tuberculous lesions are present in

the muscular system or in the lymphatic

glands embedded in or between the be s
muscles,

(d) When tuberculous lesions exist in any
part of an emaciated carcass,

all the organs may

PREVENTION ARISING FROM FOOD 405

(a) When the lesions are confined to the lungs
and the thoracic lymphatic glands,

(6) When the lesions are confined to the liver,

(c} When the lesions are confined to the
pharyngeal lymphatic glands,

(d~) When the lesions are confined to any
combination of the foregoing, but are

The carcass, if other-
wise healthy, shall not
be condemned, but
every part of it con-
taining tuberculous
lesions shall be seized.

collectively small in extent,

They add that

in view of the greater tendency to generalisation of tuberculosis in the
pig, we consider that the presence of tubercular deposit, in any degree,
should involve seizure of the whole carcass and of the organs. In respect
of foreign dead meat, seizure shall ensue in every case where the pleura
have been " stripped."

These rules, where adopted, give a fairly good guarantee
against the entry of tuberculous meat into the market. They
are fairly well enforced in all public abattoirs, and possibly
in a majority of private slaughter-houses in towns ; but in
rural districts there is no efficient control. It is not even
obligatory that animals should be slaughtered in a registered
or licensed slaughter-house ; and when an animal is killed on
the farm, there is no enactment compelling the submission
of the carcass to inspection by a competent inspector. Such
inspectors often do not exist in rural districts. A large amount
of diseased meat is prepared for the market on unlicensed
premises in country districts, and is smuggled into towns. The
one essential for improvement is that no meat should be allowed
to be exposed for sale, or to be conveyed from place to place
(except when it is consigned to a clearing house or public abattoir
for inspection), unless it is stamped in some way, to vouch that
it has been properly inspected.

The following extracts from the above report (1898) em-
phasise as strongly as is needful the evils of the present state
of things : —

So long as private slaughter-houses are permitted to exist, so long
butchers, from use and wont, will continue to use them, and so long
must inspection be carried on under conditions incompatible with
efficiency ; besides other disadvantages and risks to health which lie
beyond the scope of our reference.

Nor is there anything lacking in thoroughness in the recom-
mendations of the Royal Commission, which were as follows : —

406 THE PREVENTION OF TUBERCULOSIS

We recommend that in all towns and municipal boroughs of England
and Wales, and in Ireland, powers be conferred on the authorities similar
to those conferred on Scottish corporations and municipalities by the
Burgh Police (Scotland) Act, 1892, viz. : —

(a) When the local authority in any town or urban district in England
and Wales and Ireland have provided a public slaughter-house, power
be conferred on them to declare that no other place within the town
or borough shall be used for slaughtering, except that a period of three
years be allowed to the owners for existing registered private slaughter-
houses to apply their premises to other purposes. The term of three
years to date, in those places where adequate public slaughter-houses
already exist, from the public announcement by the local authority that
the use of such public slaughter-houses is obligatory, or, in those places
where public slaughter-houses have not been erected, from the public
announcement by the local authority that tenders for their erection
have been accepted.

(6) That local authorities be empowered to require all meat slaughtered
elsewhere than in a public slaughter-house, and brought into the district
for sale, to be taken to a place or places where such meat may be in-
spected, and that local authorities be empowered to make a charge to
cover the reasonable expenses attendant on such inspection.

(c) That when a public slaughter-house has been established, inspectors
shall i be engaged to inspect all animals immediately after slaughter,
and stamp the joints of all carcasses passed as sound.

We recommend, further, that it shall not be lawful to offer for sale
the meat of any animal which has not been killed in a duly licensed
slaughter house.

Up to the present time, however, no legislation has been
passed rendering the above practical and important re-
commendations operative.

MILK FROM TUBERCULOUS CATTLE. — I cannot better sum-
marise the dangers and the remedies for the dangers arising
from tuberculous milk than in the words and recommendations
of the same Royal Commission (1898). They state their agree-
ment with the opinion of the previous Royal Commission on
Tuberculosis, that " no doubt the largest part of the tuberculosis
which man obtains through his food is by means of milk containing
tuberculous matter." They then go on to say that " even
local authorities, which exert themselves to prevent the sale of
tuberculous meat, are without sufficient powers to prevent the
sale within their districts of milk drawn from diseased cows."
It appears clear that the danger of infecting the milk arises
chiefly, if not solely, when the tuberculosis affects the udder of
the cow ; but inasmuch as " tuberculosis of the udder can rarely

PREVENTION ARISING FROM FOOD 407

be differentiated from other forms of udder disease by the
ordinary stock owner or dairyman, ... all udder diseases
should be forthwith notified to the local authority."

Since the above recommendation was made, tuberculosis of
the udder has been placed among those diseases of cattle
where the sale of the milk for human food is forbidden. It is
unfortunate that the recommendations of the First Royal Com-
mission have not been also adopted.

Town dwellers and the local authorities appointed to protect
their health are in most instances completely impotent in respect
of public measures against tuberculous milk. On this point the
report of the same Commission (1898) may be again quoted : —

It will be seen how futile are the restrictions on the sale of tuberculous
milk produced within a city in the absence of any safeguard against
its introduction from without. Clearly there is the most urgent necessity
for powers being conferred on and exercised by local authorities to make
periodical inspection of all cows of which the milk is offered for sale within
their districts.

They draw attention, furthermore, to the fact already men-
tioned, that "the spread of tubercle in the udder may be very
rapid," becoming manifested "between fortnightly inspections
carried on along with a veterinary surgeon." Notwithstanding
these facts, they were of opinion, having regard to the extent
of prevalence of the disease, that " direct action for the elimina-
tion of all tuberculous cows from dairies should proceed tenta-
tively." They recommended at once that

(i) Systematic inspection of the cows in dairies and cowsheds should
be made by the officers of the local authorities within whose district
the premises are situated ; (2) that the authorised officers of local author-
ities within whose districts milk is supplied should have power to inspect
the cows in any dairy or cowshed, wherever situated ; (3) that power
should be given to a medical officer of health to suspend the supply of
milk from any suspected cow for a limited period, pending veterinary-
inspection ; (4) that power should be given to prohibit the sale of milk
from any cow certified by a veterinary surgeon to be suffering from such
disease of the udder as in his opinion renders the animal unfit to supply
milk ; and (5) the provision of a penalty for supplying milk for sale from
any cow having obvious udder disease.

The powers enumerated under (2), (3), and (4) remain a dead
letter in most urban districts. The nearest approach -to them is
contained in the " model milk clauses" possessed by a few large

408 THE PREVENTION OF TUBERCULOSIS

towns in local Acts of Parliament. It is unnecessary to describe
these clauses in detail ; but subject to tedious regulations they
enable the veterinary inspector and medical officer of health of
the town possessing the above powers to inspect the cattle of a
suspected farm, and if tuberculosis of the udder is found, to
prohibit the supply of milk to that town from the infected cow.
There is no power to prohibit its supply elsewhere, and no power
to prevent the infected cow being sold to another farmer for
milking purposes. The recommendation of the Royal Com-
mission on this point is that

when, under the certificate of a veterinary surgeon, the sale of milk from
a given cow is prohibited, the local authority should slaughter the
same, and if on post-mortem examination it appears that the cow was
not so affected, the local authority should pay compensation to the
extent of the full value of the cow immediately before slaughter. If,
on the other hand, the animal be found to be so suffering, the carcase
should be sold by the authority, and the owner thereof should receive
the proceeds of the sale.

This recommendation has not been embodied in legislation.

In the light of the facts described above it seems clear that
the enforcement of much more efficient public health administra-
tion in rural districts than has hitherto been the rule is needed.

Failing efficient protection of the public against the supply
of foods which are sometimes contaminated by tubercle bacilli,
the public still have it within their power to protect themselves
by refusing to eat uncooked foods derived from the farm. They
may at the same time, by bringing pressure to bear on the
purveyors of meat and milk, aid in securing the commercial
protection which is the subject of the next paragraph.

COMMERCIAL PROTECTION AGAINST BOVINE TUBERCULOSIS.—
Apart from the enforcement of public health regulations, public
protection might be entirely secured under the ordinary
conditions of commercial life, if the public were willing to
pay a little more for their milk and milk-products. There
is in my opinion great scope for commercial enterprise in
this matter ; and it is not unlikely that the additional ex-
penditure at first incurred by the enterprising large farmer, in
eliminating all cattle that reacted to tuberculin, in cleansing
and disinfecting his sheds, and in giving ample light and air in
them, would eventually be recouped by the more permanent

PREVENTION ARISING FROM FOOD 409

healthiness of his herd. Some doubt may be entertained on
this point of expense, in view of the large proportion of the cattle
that would in the first instance need to be eliminated (p. 139),
and in view of the difficulty in replacing the slaughtered cows
by others reacting negatively to the tuberculin test.

The ideal would be that each dairyman should be in a position
to issue a guarantee to his customers that all the cows from
which his milk is supplied had been proved to be free from
tuberculosis by means of the tuberculin test ; and at the same
time to certify, by means of expert evidence, that all other
sanitary requirements had been fulfilled. It must be confessed
that in very few districts is it practicable at the present time to
purchase milk under an efficient guarantee to the above effect.

The next alternative is for the dairyman to supply pasteurised
milk, and this is now largely done on a commercial scale. Often
it is done to preserve stale milk, and the slight taste of pasteurised
milk is concealed by mixing the milk with fresh unpasteurised
milk. This obviously gives little protection to the purchaser.
Furthermore, the dairyman is only concerned in pasteurising at
the lowest temperature which will prevent souring of the milk, a
temperature which, as will be shortly seen, does not suffice to kill
the tubercle-bacillus. If, therefore, commercial pasteurised milk
is to be regarded as safe in respect of tuberculosis, the temperature
and duration of the heating process must be specified. The
following experimental results throw light on this question : —

THE THERMAL DEATH-POINT OF THE TUBERCLE-BACILLUS. —
In 1887 Sternberg showed that tuberculous expectoration sub-
jected to temperatures at and above 60° C. (140° F.) was rendered
harmless. From this date onwards there has been considerable
disagreement as to the exact temperature fatal to the tubercle
bacillus. Theobald Smith in 1897 found that the variable
results as to the death-point of the tubercle bacillus in milk were
probably due to the formation of the milk pellicle in which
bacilli were caught, and thus artificially protected against further
heat. Russell and Hastings in 1900 found that exposure of tuber-
culous milk to 60° C. (140° F.) in a tightly closed commercial
pasteuriser for ten minutes always destroyed tubercle bacilli,
while, when milk was heated under conditions allowing a pellicle
to form, exposure to the same temperature (60° C.) for consider-
ably longer times did not kill the bacilli.

4io THE PREVENTION OF TUBERCULOSIS

# DOMESTIC PROTECTION AGAINST BOVINE TUBERCULOSIS. —
As domestic pasteurisation is not likely to be carried out under
scientific conditions, it would not be safe to adopt a temperature
lower than 85° C. (185° F.) in domestic life. Probably, although
home sterilisers are to be obtained, the safest plan for most
households is to boil the milk in accordance with the following
directions given in a pamphlet issued by the National Association
for the Prevention of Consumption. If these are carried out
exactly, the " cooked " flavour objected to by many individuals
will be found to be comparatively slight, and little if any surface
scum will be formed.

1. Use a double milk saucepan ; x if, however, this cannot
be obtained, put the milk into an ordinary covered saucepan
and place it inside a larger vessel containing water.

2. Let the water in the outer pan be cold when placed on
the fire.

3. Bring the water up to the boil, and maintain it at this
point for four minutes without removing the lid of the inner milk
pan.

4. Cool the milk down quickly by placing the inner pan in
one or two changes of cold water without removing the lid.

5. When cooled down, aerate the milk by stirring well with a
spoon.

THE PROTECTION OF OTHER DAIRY PRODUCTS. — Butter and
cheese may also contain tubercle bacilli. The first is the more
important, as it bulks more largely in the dietary of children.
Some of the results as to the presence of tubercle bacilli in butter
may be exaggerated, owing to possible confusion with other acid-
fast bacilli. They are, however, sometimes present, and the
only safe protection is by partially cooking the butter ; which,
however, loses much of its palat ability by this process.

1 Obtainable from any ironmonger.

CHAPTER LI

THE CO-ORDINATION OF MEASURES AGAINST
TUBERCULOSIS

REFERRING to the tabular statement on p. 317 it will be
seen that preventive measures against tuberculosis must
have regard to the receptivity of the patient, as well as
to the prevention of infection. The measures against receptivity
have been almost sufficiently indicated in previous chapters.
Every improvement in cleanliness and ventilation, every ap-
proach towards better nutrition, every avoidance of excessive
fatigue and of other depressing influences undoubtedly tends
to diminish active infection. Whether to these should be added
measures directed against the marriage, and especially the inter-
marriage, of those with a strong family history of phthisis is
a subject of much greater difficulty. As already indicated
(p. 189), each family history would, in the event of advice on
this point being given, need to be considered as a separate
problem ; and the opportunities for infection in the family, as
well as the possible inheritance of innate weakness, would need
to be carefully weighed.

In this chapter, we propose to endeavour to summarise
and obtain a conspective view of all those measures against
tuberculosis which public authorities and the governing bodies
of hospitals, dispensaries, and friendly societies may be able to
adopt. Evidently the greatest efficiency of result is likely to be
secured by first obtaining a complete view of the measures which
are practicable, and then by bringing the scattered efforts in
posse as well as in esse into active relationship with each other.

The following schemes, which to a certain extent overlap,
show the main official measures and the operations of hospitals
and dispensaries in the prevention of phthisis. In each scheme
I have placed the medical officer of health as the agent for
originating and co-ordinating preventive measures ; and although

412

THE PREVENTION OF TUBERCULOSIS

w
w
u

a <

c r! ?- 05^^ w -•'td oTi! •»•* »-3

O 3 •£ £ £X <u —i fl-5-^rtc/3*o

'5 g O U'2^ ^ O ^O^^^-g

Hfi !i? Ll,li^fi

lltl^il^FWo:

^3
^

»31

o 8

1«

J3 <U

M 5

? unlinked

fiCJ
M

*-.&

.a
^^
ft

•S c

«|

i/)

.52

•a

fa K

• «orT. ^

THE CO-ORDINATION OF MEASURES

413

personal, domestic, and industrial measures of prevention are
practicable, and are occasionally practised, apart from notifica-
tion of cases to the medical officer of health, it is none the less
true that they are commonly neglected and cannot in the com-
pletest sense be carried out apart from such notification.

The second scheme indicates from the point of view of the
individual patient as well as of the public health what is practi-
cable under present conditions.

SCHEME II

I. Patient is treated at HOME.

(1) Under the charge of his own doctor, the dispensary,

out-patient department of the hospital, etc.

(2) Home visits are made by the medical officer of health,

or his assistant, in connection with which

(a) Cleansing and disinfection are arranged.

(b) Instructions are given as to general hygiene,

and as to the special hygiene of the disease.

required.

(d) Material aid is given in conjunction with

voluntary agencies, friendly societies, and
the poor-law organisation.

(e) Regular visits to the doctor or dispensary are

urged.
(/) Dispensary or hospital tickets are given to

other members of the same family who

appear to be failing in health.
(g) Free bacteriological examination of sputum

from these or from any other suspected

patients is provided.
II. Patient is admitted to a SANATORIUM.

1 i ) Disinfection of the patient's home is arranged.

(2) Aid is organised as required for the patient's family,

hospital tickets provided for suspected cases, etc.
III. Patient is admitted to a HOSPITAL FOR ADVANCED CASES.

At present in most districts the only hospital available for
advanced patients is the workhouse infirmary, which
is only available for pauper patients.

The preceding schemes display the imperfections of our
present official measures and the reforms which are indicated.
Thus there are insufficient encouragements to early treatment
of this most curable disease. We have no system of sickness
insurance of a national character as in Germany, and medical
aid is not so readily obtainable as to compensate in part for
the absence of this. Friendly Societies do not completely fill
the gap here indicated. We have no universal system of com-
pulsory notification of phthisis, nor, it may be added, is public
opinion — without which it would be inoperative — completely
ripe for such a measure. Sanatorium accommodation for early

PATIENT. WITH
PHTHISIS
NOTIFIED

TO THE

MEDICAL OFFICER
OF HEALTH.

4i4 THE PREVENTION OF TUBERCULOSIS

cases among wage-earners is very deficient. There is a still
more serious deficiency of institutional treatment for advanced
patients who are not paupers, but who cannot afford to provide
suitable treatment at home. The arrangements for providing
suitable occupation, or part-time employment, for patients dis-
charged from a sanatorium partially cured, need to be organised
on a larger scale, and the practicability of industrial colonies
will require to be considered.

But even under present conditions a study of the two pre-
ceding schemes indicates how much admirable work — beyond
what is done in most communities — can be done under present
conditions by the full employment of official machinery and by
its co-operation with voluntary agencies. By proceeding on
the tried lines described in the preceding chapters, by further
experimental advance from the points of vantage already
reached, and above all by the earnest and combined efforts of
voluntary and official workers, there is, in my opinion, no reason
why, within a relatively short period, tuberculosis should not
follow the closely allied disease of leprosy towards extinction.

BIBLIOGRAPHY

ABRAHAM, P. (1896). Discussion on Latency, etc. Proc. Med. Chi.
Soc. Lond., 1896.

ALLBUTT, CLIFFORD (1899). On the Preventive and Remedial Treat-
ment of Tuberculosis. Brit. Med. Journ., Oct. 28, 1899, pp.
1149-1151.

ANNETT, H. E. (1902). Tubercular Expectoration in Public Thorough-
fares. Vol. iv. pt. ii. Reports of Thompson Yates Laboratories.

ARLIDGE (1892). Diseases of Occupations, p. 246.

ARMSTRONG, H. (1902). A Note on the Infantile Mortality from
Tuberculous Meningitis and Tabes Mesenterica. Brit. Med.
Journ., vol. i. p. 1024.

BARTHET AND STENSTROM (1905). The Action of Heat on the Viru-
lence of Tuberculous Milk. Le Bulletin Veterinaire, 1905,
p. 510, and Journal of Compar. Pathol. and Therapeutics, vol. xix.
pt. i. p. 62.

BEEVOR, H. (1899). Hunterian Oration on the Declension of Phthisis.
Lancet, April 1899, P- IQo8.

- H. (1900). Sex Constitution and its Relation to Pulmonary
Tuberculosis. Med. Magazine, June 1900.

(1901). Maps, Charts, and Tables illustrating the Associations

of Phthisis in England. Descriptive Catalogue, Brit. Congr. on
Tuberculosis, 1901, p. 158.

- (1905). Discussion on Paper by A. Newsholme. Epidem.
Soc. Trans., p. 130.

BIELEFELDT, PRIVY COUNCILLOR (1901). The Battle against Con-
sumption ... by means of the German Workmen's Insurance.
Trans. Brit. Congress on Tuberculosis, vol. ii. p. 336.

BIGGS, HERMANN (1903-04). First Ann. Rep. of the Henry Phipps
Institute, p. 191.

(1903). Tuberculosis : its Causation and Prevention in a

Handbook on the Prevention of Tuberculosis (Charity Organisa-
tion Society, New York, etc.).

BODINGTON (1840). The Treatment and Cure of Pulmonary Con-
sumption ; reprinted in Selected Essays, etc.. New Sydenham
Soc., 1901, p. 125.

415

416 THE PREVENTION OF TUBERCULOSIS

BOWDITCH, H. I. (1862). Paper on the Topographical Distribution
and Local Origin of Consumption in Massachusetts. Read
before the Mass. Med. Society, May 28, 1862.

BROADBENT, Sir WILLIAM (1905). Discussion on A. Newsholme's
Paper. Trans. Epid. Soc., p. 118.

BROUARDEL (1901). Address to the British Congress on Tuber-
culosis, vol. i. p. 66 and p. 48.

BUCK (1879). Manual of Hygiene and Public Health, vol. ii. p. 29.

BULSTRODE, H. T. (25, vii. 1903). Milroy Lectures. Lancet, ii. p. 208.

BURTON-FANNING, F. W. (1902). On the Etiology of Pulmonary
Tuberculosis. Practitioner, vol. 68, pp. 317-326.

(1904 ?). The Open- Air Treatment of Pulmonary Tuberculosis.

CALMETTE AND GU£RIN (1905). Origine intestinale de la tuberculose

pulmonaire. Annales de VInstitut Pasteur, tome xix. No. 10,

p. 601.

CHEYNE, WATSON. Practitioner, April 1883.
COATES, J. (1891). Tuberculosis viewed as an Infectious Disease :

its Prevalence and the Frequency of Recovery from it. Sanitary

Journal, No. 189, p. 343.
COHNHEIM (1890). Lectures on General Pathology. New. Syd. Soc.,

vol. iii. p. 1030.
CORNET, G. (1904). Tuberculosis in Nothnagel's Encycl. of Practical

Medicine.
COURTOIS-SUFFIT ET CH. LAUBRY (1905). Role des Sanatoriums

et des Dispensaires dans la Lutte Anti-Tuberculeuse. Rapports

presentes au Congres Internal, de la Tuberculose, p. 503.
DEBOVE AND ACHARD. Manuel de Medecine, tome ix. p. 271.
DELEPINE, S. (1898). Tuberculosis and the Milk Supply. Lancet,

vol. ii. p. 736.

(1899). Prevention of Tuberculosis in Cattle. Veterinarian,

July and August 1899.

FAGGE (1886). Medicine, vol. i. p. 983.

FAGGE AND PYE SMITH. Principles and Practice of Medicine, 3rd ed.,

vol. i. p. 639.

FARR (1885). Vital Statistics, p. 513.
FLINT, AUSTIN (1882). The Self-limited Duration of Pulmonary

Phthisis. Brit. Med. Journ., Sept. 30, 1882.
FLUGGE C. (1898). Die Verbreitung der Phthise durch staubfor-

miges Sputum und beim Husten versprizte Tropfchen. Ztsch. f.

Hygiene u. Inf. Kr. xxx. 107.
FOWLER, J. K. (1896). Discussion on Latency, etc. Proc. Med. Chi.

Soc. Lond.

BIBLIOGRAPHY

417

FOWLER, J. K. (1906). The Therapeutic Value of Sanatorium
Treatment in Pulmonary Tuberculosis. Lancet, Jan. 6, 1906.

(1898). Diseases of the Lungs, p. 305.

Fox WILSON (1891). Diseases of the Lungs, p. 563.

Treatise on Diseases of the Lungs and Pleura, ed. 1891.

FRANKEL (1906, Mar. i). Deutsch. Med. Woch., quoted in Brit.

Med. Journal Epitome (1906), p. 53.
GANGHOFNER, F. (1905). Preservation Scolaire contre la Tuber-

culose. Rapports presentes au Congres Internal, de la Tuber-

culose, Paris, p. 315.
GARLAND, C. H. (1905). Assurances et Mutualites dans la Lutte

contre la Tuber culose. Rapports presentes au Congres Internal.

de la Tuber culose, Paris, p. 495.
GRAY, E. G. (1906). An Unusual Case of Typhoid Infection. Lancet,

July 1906.

GREENHOW (1869). Path. Trans., vol. xx. p. 57.
GREENHOW, KNAUFF. Archiv fiir path. Anat. und Physiol. und fur

Klin. Med. von Virchow, Bd. xxxix. S. 442.
GUTHRIE, L. G. (1899). The Distribution and Origin of Tuberculosis

in Children. Lancet, vol. i. pp. 286-290.
HARRIS, T. Brit. Med. Journ., vol. ii. p. 1385.
HAYWARD, T. E. (1904). On the Construction of Life Tables, p. 27.

Victoria University Reports, edited by Professor Delepine.
HENSCHEN, S. E. (1905). La Lutte contre la Tuberculose en Suede.

Ouvrage dedie au Congres Internal, de la Tuberculose a Paris, 1905.
HEYMANN (1901). Versuche iiber die verbreitung der Phthise durch

ausgehustete Tropfchen und durch trockenen Sputum. Zeits.

fur Hyg. und Infektionskrankheiten, xxxviii. 20-93.
HILLIER, A. (1903). The Nature of the Infectivity of Phthisis : A

Study of the Views of Koch, Fliigge, and others. Brit. Med.

Journ., vol. i. p. 593.
HIRSCH. Geographical and Historical Pathology, vol. iii. p. 203.

Syd. Soc. Trans.

vol. iii. pp. 197-198.

HOFFMAN, F. L. (1901). Industrial Insurance and the Prevention
of Tuberculosis. Trans. Brit. Congr. on Tuberculosis, vol. ii.

P- 348.
HUTCHINSON, JONATHAN (1896). Discussion on Latency, etc. Proc.

Med. Chi. Soc. Lond.

KELYNACK (1904). The Sanatorium Treatment of Consumption, p. 7.
KINSFORD, L. (1904). The Channels of Infection in Tuberculosis

in Childhood. Lancet, vol. ii., Sept. 24, 1904.
27

418 THE PREVENTION OF TUBERCULOSIS

KNOPF. Pulmonary Tuberculosis : its Modern Prophylaxis, p.

55-
KOCH, R. Etiology of Tuberculosis, translated by Stanley Boyd,

vol. cxv. New Syd. Soc.

(1901). Address to British Congress on Tuberculosis. Trans.

Brit. Congr. on Tuberculosis, vol. i. p. 52.

(1906). Nobel Lecture on " How the Fight against Tuberculosis

now stands." Lancet, vol. i. pp. 1449-1451.

KOSSELL, H. (1905). A Report on Human and Bovine Tuberculosis.

Brit. Med. Journ., 1905, vol. ii. p. 1445.
LARTIGAU (1901). On Tuberculosis. Twentieth Century Practice

of Medicine, vol. xx.

LARTIGAU AND NICOLL. Amer. Journ. Med. Sciences, June 1902.
LATHAM, A. (1900). Pulmonary Tuberculosis in Early Childhood.

Lancet, 1900, vol. ii. pp. 1785-86.
— — (1903)- Prize Essay on the Erection of a Sanatorium

for the Treatment of Tuberculosis in England. Lancet, Jan. 3,

1903.

(1906). The Economic Value of Sanatoriums. Lancet, Jan. 6,

1906.

LECKY AND HORTON (1907). Revealed Tuberculosis in Children

at School Ages. Lancet, Dec. 28, 1907.
LISTER, J. (1868). Address on the Antiseptic System of Treatment

in Surgery. Brit. Med. Journ., vol. ii. p. 55.

Louis, P. C. A. (1844). Researches on Phthisis. Sydenham Society.
MACFADYEAN, J. (1901). Address on Tubercle Bacilli in Cows' Milk

as a possible Source of Tuberculous Disease in Man. Trans. Brit.

Congr. on Tuberculosis, vol. i. p. 83.
M ARFAN , A. B . ( 1 905 ) . Preservation de F Enfant contre la Tuber culose

dans sa Famille. Rapports presentes au Congres Internal, de

la Tuberculose, Paris, p. 255.
MATHESON, R. E. (xi. 1903). The Housing of Ireland during the

Period of 1841-1901. Journ. Statist, and Social Inquiry, vol. xi.

pt. Ixxxiii.
MEMORANDA, etc., prepared by the Board of Trade. Cd. 1761, pp.

215 and 224.

Second Series. Cd. 2337.

MORGAN, G. (1899). Remarks on Tuberculous Adenitis. Brit. Med.

Journ., Aug. 19, 1899.
MOTT, Report of Pathologist to London County Council for year ended

March 1904, p. i.
MOXON (1885). Brit. Med. Journ., vol. i. p. 130.

BIBLIOGRAPHY 419

MULLER, D. (1905). Milk and Dairy Products as Sources of Infection

in Tuberculosis. Journ. Compar. Path, and Therapeutics, vol.

xix. pt. i. p. 19, and Proc. 8th Internal. Vet. Congress, Budapest.
NEWMAN AND SWITHINBANK (1903). Bacteriology of Milk, p. 268.
NEWSHOLME, A. (1896). On the Study of Hygiene in Elementary

Schools. Public Health, vol. iii. p. 135.
— (1901). The Influence of Soil on the Prevalence of Pulmonary

Phthisis. Practitioner, New Series, vol. xiii. p. 206.

- (1903). Public Health Authorities in relation to the Struggle
against Tuberculosis in England. Journal of Hygiene, vol. iii. p.
461 ; also Compt. rend. XI IP Congres International d' Hygiene et
de Demogr., Bruxelles.

(1904). Protracted and Recrudescent Infection in Diphtheria

and Scarlet Fever. Med. Chi. Trans., vol. 87.

- (1905). A Study of the Relation between the Treatment of
Tuberculous Patients in General Institutions and the Reduction
in the Death-rate from Tuberculosis. Reports to Internal. Congr.
on Tuberculosis, Paris, p. 427.

- (1905). The Relative Importance of the Constituent Factors
involved in the Control of Pulmonary Tuberculosis. Trans.
Epidem. Soc., New Series, vol. xxv. p. 32.

- (1906). An Inquiry into the principal Causes of the Reduction
in the Death-rate from Phthisis during the last Forty Years, with
special Reference to the Segregation of Phthisical Patients in
General Institutions. Journal of Hygiene, vol. vi., No. 3, p. 304.

- (1907). The Co-ordination of the Public Medical Services. An
Address given at the Meeting of the State Medicine Section of the
Meeting of the British Medical Association at Exeter, July 1907.
Brit. Med. Journ., Sept. 14, 1907.

- (1907). Poverty and Disease as illustrated by the Course of
Typhus Fever and Phthisis in Ireland. Presidential Address,
Epidemiological Section, Roy. Soc. Med., Oct. 1907.

OSLER, W. (1901). The Principles and Practice of Medicine, pp. 258,

338.

PARKER, W. R. (1903). Sanatoria plus Homes for Consumption.

Mar. 14, 1903.
PEARSON, KARL (1907). A First Study of the Statistics of Pulmonary

Tuberculosis. Drapers' Company Research Memoirs.
PHILIP, R. W. (1906). The Public Health Aspects of the Prevention

of Consumption. Brit. Med. Journ., Dec. I, 1906.
POWELL, DOUGLAS. Lecture on the Prevention of Consumption.

Journ. San. Inst., Aug. 1904, vol. xxv. pt. ii. p. 353.

420 THE PREVENTION OF TUBERCULOSIS

QUAIN'S Dictionary of Medicine. Ed. 1894, vol. ii. p. 414.
RANSOME, A. (1890). The Cause and Prevention of Phthisis, p. 50.

(1895). Consumption a Filth Disease. Lancet, Jan. i, 1898, p. 15.

(1902). The Intercommunicability of Human and Bovine

Tuberculosis. Proceedings Patholog. Soc. of Philadelphia, May
1902.

(1905). Comparative Study of Various Forms of Tuberculosis.

Rapports au Congres Internal, de la Tuberculose, Paris, pp. 135-148.

RINDFLEISCH (1875). On Chronic and Acute Tuberculosis, in von

Ziemssen's System of Medicine, vol. v. p. 649.
ROBERTS, F. T. (1902). On the Comprehensive Study of Thoracic

Phthisis. Lancet, vol. i. pp. 867-874.
RUCHLE (1875). Pulmonary Consumption. Von Ziemssen's Medicine,

vol. v. p. 508.
SANTOLIQUIDO (1903). Compt. rend. XIII* Congr. d'Hygiene et de

Demogr., Bruxelles, vol. vii. p. 45.
SHADWELL, A. (1905). Industrial Efficiency, vol. ii.
SMITH, THEOBALD (1904). A Study of the Tubercle Bacilli isolated

from Three Cases of Tuberculosis of the Mesenteric Lymph

Nodules. Amer. Journ. of the Med. Sciences, Aug. 1904.

(1905). Studies in Mammalian Tubercle Bacilli : III. Description

of a Bovine Bacillus from the Human Body. Journ. of Med.
Research, vol. viii. No. 3, pp. 253-300.

(1905). The Reaction Curve of Tubercle Bacilli from Different

Sources in Bouillon containing different Amounts of Glycerine.
Journ. of Med. Research, vol. xiii. No. 4.

SQUIRE, J. E., C.B. (1906). A Lecture on Pulmonary Tuberculosis in
Children. Brit. Med. Journ. 1906, vol. ii. p. 133.

(1906). The Results of Sanatorium Treatment of Consumptives.

Tuberculosis, vol. iv. No. 3.

STENGEL. Manual of Pathology, p. 255.

STILL, G. F. (1901). Tuberculosis in Childhood. Practitioner, vol. 67,

pp. 91-103.
TATHAM, J. W., Report of Royal Commission on Tuberculosis,

part. ii. Appendix C., and Annual Reports of Registrar- General of

Births and Deaths.
THOMSON, ST. CLAIR (1901). Tubercular Infection through the Air

Passages. Practitioner, 1901, vol. ii. pp. 80-90.
THOMSON, ST. CLAIR AND HEWLETT (1895). Path. Soc. Trans.,
. vol. Ixxviii.

(1896). The Fate of Micro-organisms in Inspired Air. Lancet,

vol. i. pp. 86-87.

BIBLIOGRAPHY 421

THORNE, R. T. (1888). The Progress of Preventive Medicine during

the Victorian Era, p. 51.
TYNDALL (1876). The Optical Deportment of the Atmosphere in

relation to the Phenomena of Putrefaction and Infection. Phil.

Trans. Roy. Soc., vol. clxvi. pt. i. p. 27.
VALLEE, M. H. (1905). De la Genese des lesions pulmonaire dans la

Tuberculose. Ann. de VInstitut Pasteur, tome xix. No. 10,

p. 649.
VON BEHRING (1904). The Suppression of Tuberculosis. Cassel

Lecture, September 1903, American Translation, p. 14.
WALKER, J. (1906). Employment of Consumptive Patients. Tuber-
culosis, Jan. 1906.
WALSHAM, HUGH (1904). The Channels of Infection in Tuberculosis,

p. 6.

WALSHE, W. H. (1871). Diseases of the Lungs.
WALTERS, F. R. (1905). Sanatoria for Consumptives.

- (1906). Lancet, Jan. 6, 1906.
WASHBOURNE (1896). Discussion on the Latency of Parasitic

Germs or Specific Poisons in Animal Tissues. Proc. Med. Chi.

Soc. Lond., 1896.
WATSON, A. W. (1903). An Account of an Investigation of the

Sickness and Mortality Experience of the I.O.O.F. Manchester

Unity.
WEBER, H. (1874). On the Communicability of Consumption from

Husband to Wife. Clin. Soc. Trans., 1874, vol. vii. p. 144.
WEST, S. (1902). Diseases of the Organs of Respiration, vol. ii.

p. 436.

WILLIAMS, DAWSON. Trans. Path. Soc., vol. xxxv. p. 413.
WOODHEAD. Report of Royal Commission on Tuberculosis, 1895,

P- X45.

INDEX OF PLACES

Aberdeen, 195, 362

Alnwick, 196

America, 17, 218, 222, 229, 236, 241, 290

Axminster, 200

Belfast, 148, 218, 383

Belgium, 243, 256

Berlin, 132, 144, 184, 228, 287

Birmingham, 223, 274

Blackburn, 362

Bradford, 369

Brighton, 1 8, 29, 71, 102, 259, 309,

312, 324, 334, 342, 343, 348, 352,

36i, 395

Brompton Hospital,. 152
Brussels, 288
Brynmawr, 197

Carlisle, 196
Chelmsford, 196
Chich ester, 199
Cincinnati, 290
Copenhagen, 145, 286
Croydon, 199, 274

Denmark, 145
Dover, 199
Dublin, 148, 213, 283
Dundee, 195, 361
Dunfermline, 362

Edinburgh, 148, 195, 362, 378
England, throughout

France, 200, 222, 231, 235, 242
Frimley, 387

Germany, 211, 223, 229, 231, 237, 241,

254, 372, 413
Glamorgan, 333
Glasgow, 148, 195, 275
Gorbersdorf, 384
Gottenburg, 285
Greenock, 195

Holland, 200

Ireland, 172, 212, 218, 222, 226, 237
243, 247, 254, 259, 269, 281

Kensington, 274, 284

Leeds, 274

Leicester, 196

Leith, 195

Lille, 377

Liverpool, 30, 144, 148, 331, 369

London, 147, 148, 246, 269, 277

Manchester, 49, 98, 100, 144, 148, 193,

213. 343
Massachusetts, 171, 195

Naples, 56

New York, 17, 24, 228, 290, 343, 345,

368

Nordrach, 230, 285, 384
Norway, 212, 222, 231, 243, 254, 259

Oldham, 343

Paisley, 195

Paris, 50, 106, 228, 235, 253, 289

Penzance, 197

Providence, 171, 219

Prussia, 171, 222, 253, 287

Salford, 274

Salisbury, 196

San Francisco, 290

Scotland, 171, 19?, 222, 227, 229, 236,

239, 250, 269, 280
Sheffield, 223, 275, 343, 345
Stockholm, 285
Sussex, 20 1
Sweden, 285
Switzerland, 253

United States. See America

Ventnor Hospital, 153
Victoria Park Hopsital, 152

Zurich, 363

423

INDEX OF NAMES OF PERSONS

(See also under Bibliography}

Abraham, 75

Achard, 181

Acland, 364

Allbutt, 48, 163, 179, 315

Annett, 331

Arlidge, 109, no

Armstrong, H., 30

Ash, 362

Ashby, 31

Baillie, 36

Baldwin, 88

Baumgarten, 61, 79, 96, 183

Bayle, 36

Beevor, H., 53, 102, 168, 230, 232, 258

Behring, von, 86, 129, 135, 136

Beninde, 91

Bennett, H., 37, 45, 46, 48

Bernheim, 289

Bertillon, 289

Bielefeldt, 18, 372

Biermer, 146

Biggs, Hermann, 17, 228, 290

Bodington, 383

Bowditch, 195

Brehmer, 150, 384

Broadbent, 256, 314

Brouardel, 48, 181

Buchanan, 195, 202

Buhl, 37, 79

Bulstrode, 153, 252

Burton-Fanning, 177, 188

Buschke, 81

Cadeac, 54, 116
Calmette, 116, 313, 377
Cameron, C., 283
Cars-well, 48
Carter, V., 264
Chantemesse, 81
Chapin, 219
Charcot, 48
Charrin, 177

Chauveau, 40

Cheyne, W., 60

Chopin, 57

Coates, 23, 49

Coates, H., 98

Cohnheim, 40, 60, 109

Colman, 119

Cook, C. W., 109

Cornet, 59, 78, 88, 95, 96, 146, 184 320

Debove, 79

De Jong, 135

Delepine, 53, 98, 117, 143, 144

Dettweiler, 59, 384

Devlin, 283

Dieulafoy, 108

Donkin, 31

Dreschfield, 193

Dudgeon, 81

Elliott, T., 139

Fagge, 31

Falk, 115

Farr, 20

Fischer, 53, 54

Fleming, 273

Flint, 49

Fliigge, 91, 92, 95

Fowler, J. K., 78, 162, 183, 387

Fox, Wilson, 60, 152, 185

Fox, Wilson, C.B., 241

Frebelius, 164

Galen, 35
Ganghofner, 80
Garland, 375
Greenhow, no
Greenwood, 362
Gresswell, 74
Griffith, A. S., 124
Guerin, 116
Guthrie, 113, 118

INDEX OF NAMES OF PERSONS

425

Harris, T., 48
Harvey, G., 56
Hay, 362
Hayward, 14
Hervieux, 164
Heuss, 135
Hewlett, no
Heymann, 92
Hildebrandt, no
Hippocrates, 35, 48
Hirsch, 194, 201, 202
Hoff, 286
Hoffman, 17, 374
Horton, 361

Kanthack, 53, 144

Kelly, 20 1

Kelynack, 383

Kingsford, 120

Kitasato, 78, 103

Klebs, 60

Klemperer, 129

Klencke, 38

Knauff, no

Knopf, 368

Koch, R., 41, 52, 59, 60, 89, I2i, 128,

131, 146, 154, 252
Kossel, 125, 132, 135, 136

Laennec, 36, 49
Landouzy, 23, 181
Lartigau, 54, 88, 131
Laschtschenko, 92
Latham, 109, 120, 385
Lebert, 37
Lecky, 71, 361, 397
Lister (Lord), in
Lister, T. D., 24, 375
Loomis, 79
Lorenz, 136
Louis, 49
Low, 362
Lyon, G., 24

MacConkey, 79

MacCormac, H., 383

MacFadyean, J., 139, 142, 182, 313

Macfadyen, A., 79

Mackenzie, L., 362

M'Weeney, 144

Magnetus, 36

Marfan, 81

Martin, H., 46

Martin, S., 61, 108, 116, 126, 140

Matheson, 148, 227

Mesurier, 289

Metchnikoff, 44

Milroy, 195

Moller, 135

Morgagni, 36, 55

Morgan, 108

Mott, 178, 274

Moxon, 53

Muirhead, 374

Miiller, 80, 116, 143

Murphy, S., 147

Naegeli, 363

Newman, 162, 355

Niemeyer, 37

Niven, 319, 343, 355, 370

Nocard, 135

Nuttall, 104

Ostertag, 143

Paget, 343
Parker, 302
Pasteur, 41, 75
Pearse, 200
Pearson, K., 187
Philip, 378, 381, 382
Powell, 230

Rabinowitch, 144

Ransome, 54, 89, 180, 193

Ravenel, 125, 130, 131, 132, 135

Reid, 36

Reinhardt, 37

Ribard, 48

Richardson, 383

Rindfleisch, in

Robertson, 153, 165, 345

Rokitansky, 37

Romer, 136

Russell, H. W., 118

Sand, George, 57
Sanderson, B., 39
Santoliquido, 255
Savoire, 377
Schultze, in
Schiitz, 125, 128
See, 162

Shadwell, 222, 228, 241
Simon, J., 40

426

THE PREVENTION OF TUBERCULOSIS

Smith, Theobald, 124, 121, 134, 409

Smollett, 56

Spengler, 129

Spillmann, 88

Squire, 109, 113, 185, 313, 390

Stafford, 180

Stead, 375

Stengel, 77

Sternberg, 409

Still, 119

Strauss, 53, 106, 115

Swithinbank, 162

Sylvius, 35

Tappeiner, 41, 96

Tatham, 3, 7, 155, 158, 165, 172, 213,

360

Taute, 135

Thomson, St. Clair, 106, 108, no, 112
Thorne, 32, 201
Trudeau, 192, 382
Turban, 315
Tyndall, 89, 105

Vagades, 131

Valsalva, 55
Villemin, 38, 41, 59, 89
Villoret, in
Virchow, 37
Volland, 114

Walker, 393

Walsham, H., 113

Walshe, 59, 185

Walters, F. R., 384, 387, 393

Walther, 384

Washbourne, 79, 81

Watson, 15

Weber, 135, 150

West, 185

Wesener, 115

Williams, C. T., 49, 89, 152, 184, 314

Williams, D., 60

Wright, 31

Wright, A., 44

Woodhead, Sims, 109, 113, 142, 144

Zenker, no
Ziemssen, 390
Ziehl-Nielsen, 60

INDEX OF SUBJECTS

Adenoids, 108, 302

Advanced cases, treatment of, 366, 401
After-care of patients, 392
Age-incidence of tuberculosis, 6, 29,

118, 164, 219

Agricultural labourers' wages, 241
Air, expired, and infection, 88
Alcohol and phthisis, 181, 304, 319
Asylums and phthisis, 178, 274
Attendance on sick and infection,

ip
Auto-infection, 319

Bacillus tuberculosis, 5 1

number in expectoration, 104
bovine and human, 124, 134

Bovine tuberculosis, 121

Bronchial glands and infection, 112

Bronchitis, confusion with phthisis, 24
relation to phthisis, 1 79

Bye-laws as to spitting, 334

Catarrhs and phthisis, 178
Cattle and tuberculosis, 139
Children, latent tuberculosis in, 80, 363
Cleanliness, 302
Climate and phthisis, 194
" Colds " and phthisis, 178
Common lodging-houses, 368
Comparative mortality figures, 157
Compulsory notification, 344, 349

removal of patients, 367
Congenital phthisis, 182
Co-ordination of measures, 411
Correlation coefficients, 295
Coughing and infection, 97
Cows' milk. See Milk

Dairy products and tuberculosis, 145
Death-rate. See Mortality
Decadence, 217
Definition of tuberculosis, 3
Desiccation and life of bacillus, 52

Diagnosis, accuracy of, 23
importance of early, 306
Diseases predisposing to phthisis, 178
Disinfection, 321, 327, 355
Dispensaries and prevention of phthisis

377
Doctors and infection, 155

and preventive measures, 316

and disinfection, 321

and notification, 321
Domestic infection, 146
Drainage of soil, 196
Droplets, infection by, 93
Duration of life and phthisis, 14
Dust infection, 91, 97, 105, 120

in rooms, infectivity of, 94
Dwelling and infection, 104, 146, 225,
304

Economics of tuberculosis, 13, 17
Education and phthisis, 252, 302

of patient, 357

authorities and phthisis, 359
Elimination of susceptible strains,

216

Emigration and phthisis, 218
Entry of infection, 108
Environmental conditions,

lowering resistance, 191, 215
Expectoration in phthisis, 102

and tubercle bacilli, 103, 314

swallowing of, 320

prevention of, 331

disposal of, 332

regulations as to, 333
Experimental investigation, 38

evidence as to infection, 89, 1 10
Expired air and infection, 89, 105
Extra-corporeal life of bacillus, 104

Family infection, 149

Fatigue, 177, 303

Financial loss by phthisis, 15

428

THE PREVENTION OF TUBERCULOSIS

Flies and infection, 88
Food cost and phthisis, 236

amount and phthisis, 243

and tuberculosis, 403
Friendly Societies' experience, 1 5

and phthisis, 374

Gastric juice, 115
General tuberculosis, 27, 37, 114
German insurance scheme, 18
Guardians, Boards of, and phthisis,

366
and sanatoria, 369

Hands and infection, 87
Heredity and phthisis, 182
History of phthisis, 3 5

views on infection, 5 5

importance in diagnosis, 312
Home treatment, 326

of pauper cases, 370
Hospitals and dust infection, 98

and infection, 153
House. See Dwelling
Housing, 104, 146, 225, 304
Hygiene, teaching of, 302

Ignorance, removal of, 311
Income of family and phthisis, 241
Incubation period, 75
Industrial infection, 157

prevention, 329

colonies, 393
Infection, history of views on, 55

experimental evidence of, 59

channels of, 60

statistical and clinical evidence, 62

sources of, 86

limitations to, 101

by inhalation, 106

Infirmaries and phthisis, 18, 246, 273
Influenza and phthisis, 178
Ingestion of infection, 115
Inhalation of infection, 106
Injury and phthisis, 178
Inoculation of tuberculosis, 87
Institutional treatment of advanced

cases, 366

Instructions for patients, 324
Insurance Societies' experience, 17

and phthisis, 372
Intestinal infection, 1 16
Isolation hospitals and phthisis, 397

Latency, duration of, 73, 257
pathological evidence as to, 77
in scholars, 363

Latent tuberculosis, treatment of, 363

Leprosy and phthisis in Norway, 259,
263

Limitations to infection, 101, 105

Living, cost of, 238

Lunatic asylums. See Asylums

Lungs, direct infection of, 109
indirect infection of , 1 1 2

Magnitude of the evil, 4
Malnutrition, 179, 303
Married life and infection, 149
Meat and tuberculosis, 139, 404
Medical treatment, 306

service, 310
Milk and tuberculosis, 130, 141, 144,

406
Mortality from tuberculosis, 4

according to age, 4

according to sex, 7

Notification of phthisis, 338

and decline of phthisis, 253, 301
and doctors, 321
Nursing and infection,
Nutrition and phthisis, 179, 230

Occupation and phthisis, 158

of consumptives, 327
Open-air treatment, history of, 383
Ophthalmic diagnosis, 313
Overcrowding and phthisis, 147, 191,

224, 229
Overfatigue and phthisis, 177

" Parochial " statistics, 207
Pauperism and phthisis, 243
Phagocytosis in phthisis, 44
Phthisis. See Tuberculosis

confusion with tuberculosis, 26

symptoms and progress of, 43

varieties of, 47

curability of, 48

duration of, 49

Precautionary instructions, 324
Prevention and cure inseparable, 325
Portals of infection, 106
Poverty and phthisis, 179, 219, 243,

304
Prae-tuberculous stage, 312

INDEX OF SUBJECTS

429

Predisposition, 162

diseases producing, 178

hereditary, 184, 216
Proclivity, 161

Proteid food and phthisis, 231
Public-house and infection, 181, 304,
319

Railway carriages and infection, 100
Regulations as to spitting, 333
Relief for consumptive families, 370
Royal Commissions on Tuberculosis,

121, 126, 132
Rural life and phthisis, 220

Saliva and infection, 88

Sanatoria and decline of phthisis, 254

and Boards of Guardians, 369

structural arrangements of, 384
Sanatorium treatment, principles of,
386

medical results of, 390

patients suitable for, 391

and doctors, 322

and prevention of phthisis, 382
Sanitary Authority and patient, 328

and preventive measures, 351
Sanitary measures and phthisis, 192,

211, 215, 356

Scholars, latent tuberculosis in, 363
School-ages and phthisis, 360
Schools and spread of tuberculosis, 362

and public opinion, 365
Secret medical, le, 340
Segregation and phthisis, 149, 256, 266

domestic and institutional, 258

ratio, 267
Sex and tuberculosis, 164, 226

and urban or rural phthisis, 165

differences between boys and girls,
1 68

changes in incidence according to,

171
Sickness, amount of, from tuberculosis,

relation of, to mortality, 20
Soil and phthisis, 194
Speaking and infection, 95
Spitting. See Expectoration
Spray infection, 92, 120

Sputum, examination of, 102, 314

swallowing of, 320

disposal of, 332
Statistics, trustworthiness of, 22

" parochial," 207

migration and, 208
Streets and tubercle bacilli, 104, 331

and expectoration, 336
Subsoil drainage, 196
Sunlight and infection, 53, 193
Susceptibility, 161, 216. See also Pre-
disposition

Tabes mesenterica, 23, 31, 118
Teeth and infection, 109
Temperance and phthisis, 210
Temperature and life of bacillus, 53, 409
Thrift and phthisis, 210, 231
Tonsils and infection, 109
Training of patients, 391, 395
Transformation of types of bacilli, 134
Treatment, need for organisation of, 306
Tubercle, nature of, 45
Tubercle bacillus. See under Bacillus
Tuberculin testing, 3 1 3
Tuberculosis. See also under Mortality
Tuberculous meningitis, 29, 118

peritonitis, 31
Typhus fever and phthisis in Ireland,

259, 262
Typus humanus, 124, 134

bovinus, 124, 134

Ubiquity of tubercle bacillus, 101, 104,

33i

Udder disease and tuberculosis, 143
Urban life and phthisis, 220, 223

Virulence of tubercle bacillus, 2 1 5
Visits to consumptives, 359
Voluntary notification, growth of, 342

Wages and phthisis, 240
Weight, loss of, 313
Well-being and phthisis, 230
Wheat prices and phthisis, 232
Wife contrasted with nurse as to infec-
tion, 154

Workers, sanatoria for, 375
Workhouse. See Infirmaries

Printed by

MORRISON & GIBB LIMITED
Edinburgh

University of Toronto
library

DO NOT

REMOVE

THE

CARD

FROM

THIS

POCKET

Acme Library Card Pocket

Under Pat "Rel. Index File"

Made by LIBRARY BUREAU

Internet Archive Audio

Featured

Top

Images

Featured

Top

Software

Featured

Top

Books

Featured

Top

Video

Featured

Top

Mobile Apps

Browser Extensions

Archive-It Subscription

Save Page Now

Full text of "The prevention of tuberculosis"

See other formats