rb H^_.^£X3^^f zsC-zC4 t ^ Methods of Assessing the Physical Fitness of Children Bureau Publication No. 263 UNITED STATES DEPARTMENT OF LABOR CHILDREN'S BUREAU UNITED STATES DEPARTMENT OF LABOR Frances Perkins, Secretary CHILDREN'S BUREAU • Katharine F. Lenroot, Chief Methods of Assessing the Physical Fitness of Children A study of certain methods based on anthro- pometric, chnical, and socioeconomic obser- vations made of 713 7-year-old white boys and girls in New Haven, Conn,, over a period of 19 or 20 months during 1934-36 By RACHEL M. JENSS, So. D. and SUSAN P. SOUTHER, M. D. Bureau Publication No. 263 United States Government Printing Office Washington : UHO For sale by llie Superintendent of Dcieuments, Washington. D. C. ....... Price 15 cents U. S. SUPERINTENDENT OF DOCUMENT'S MOV 25 1940 CONTENTS Page Letter of transmittal v Introduction 1 Methods of assessing physical fitness 3 The clinical examination 3 Biochemical tests for specific nutritional deficiencies 4 Functional tests 4 Dietary inquiry 5 Socioeconomic inquiry 5 Anthropometric measures of body build 5 Indices of body build as a method of assessment of physical fitness 7 The four indices of body build included in the study 10 Description of each index of body build 10 The Baldwin-Wood Tables 10 The ACH f Arm-Chest-Hip) Index 11 The Nutritional Status Indices 13 The Pryor Width-Weight Tables 17 Limitations of the four indices of body build 18 Previous studies 19 Material and methods 25 A description of the boys and girls included in the study 27 Age 27 Nationality 28 Homes 28 Location 28 Type of dwelling 29 Presence of parents in the home 30 Number of persons in household 30 Number of persons per room 31 Number of additional persons sleeping in the child's bedroom . 32 Economic status of their families 32 Assistance 32 Principal source of income 33 Employment of mother - 33 Anthropometric measurements 34 Height 34 Weight 34 Diets 35 Health 36 Pediatrician's assessment of general nutritional status 36 Diagnoses made at annual physical examinations 37 Number and duration of reportable school absences 38 Medical and dental care 39 Summary ^ The observations made of these children - 41 Techniques employed ^^ Clinical assessment '*2 Anthropometric measurements.- " UI IV CONTENTS Material and methods — Continued. The observations made of these children^ — Continued. Accuracy of the techniques employed 45 Clinical assessment 45 Anthropometric measurements 45 Socioeconomic and related data 48 Summary of material and methods 49 Results 50 Indices of body build 50 The problem of testing the indices 50 Criteria for testing the indices 50 Observations used in deriving criteria 50 Description of the criteria 51 Number of children selected by the criteria 53 Evaluation of the criteria 54 Testing the indices 55 The Baldwin- Wood Tables 55 The ACH Index 61 The Nutritional Status Indices 63 The Pryor Width- Weight Tables 74 Summary 76 Clinical j udgment 76 The experimental data 76 Review of the literature 78 Variability of clinical judgment in the present stud}; 82 Stability of clinical judgment in the present study 85 Opinions and recommendations of others 87 The need for improving clinical judgment 92 Summary 94 Appendix I 97 Statistical methods 97 Supplementary tables 100 Appendix II 115 List of references 115 ILLUSTRATIONS Map. — Location of the homes of the children 29 Graph. — Weight of Italian boy, A. R., observed at frequent intervals from 72.7 to 92.8 months of age 99 Letter of Transmittal United States Department of Labor, Children's Bureau, Washington, August 15, 1940. Madam: There is transmitted herewith Methods of Assessing the Physical Fitness of Children, a study based on observations made in New Haven, Conn. The problem of finding an efficient, economical, and simple method of assessing physical fitness has occupied the attention of workers in the field of child health and growth for many years. Recently a nmnber of methods have been recommended, especially those based on anthropometric measurements of the child. Some of these pro- cedures have been in more or less widespread use but few attempts have been made to evaluate the different methods as applied to the same group of children and to compare the relative efiectiveness of these methods in identifying boys and girls who are physically unfit. Several years ago it was decided to make such an evaluation of several methods of assessment, including (1) indices of body build and (2) clinical judgment of general nutritional status. The study began in 1934. It was a cooperative undertaking of Yale University through the Institute of Hinnan Relations and the department of pediatrics of the School of Medicine and of the Chil- dren's Bureau of the United States Department of Labor, in coopera- tion with the Department of Health and the Board of Education, New Haven, Conn. The study was proposed by Frank K. Shuttleworth, Pli. D., of the Institute of Human Kclalions, who participated inoutliningtheoriginal schedules and who made preliminary statistical analyses of the data. It was can-ied out imder the direction of the following members of the respective staffs: Mark A. May, Ph. D., Director of the Institute; Grover F. Powers, M. D., Professor of Pediatrics, the Yale University School of Medicine; and Martha M. Eliot, M. D., Assistant Chief of the Children's Bureau. The general supervision of the field work and the physical examina- tions were carried out and a preliminary report of tlu* clinicid aspects of the study was prepared by Susan V. Souther. M. D., now of the Connecticut State Department of Health and formerly on the staff of the Children's Bureau; the anthropometric measurements were taken by Mary E. Parker, R. N. Assistance in some of the clinical and anthropometric aspects of the study was given by Ethel C. Dunham, M. D., and Clara E. Hayes, M. D., of the Children's Bureau, and V VI LETTER OF TRANSMITTAL Sander E. Lachinan, M. D., then on the staff of the Institute of Human Relations, Yale University. After the observations were collected, the analysis was directed and the final report prepared by Rachel M. Jenss, Sc. D., statistician in charge of studies of child health and growth, Division of Statistical Research, Children's Bureau. The study has been confined to the problem of evaluating several methods of assessing physical fitness which have been in more or less widespread use in this country during recent years. It has not at- tempted to ascertain why certain methods fail nor has it undertaken the problem of developing new methods of assessment. It is hoped that the report will resolve some of the difficulties and confusion which have existed concerning the more generally accepted methods of assessment of physical fitness, and that later a more constructive approach can be made to the problem of assessing the physical fitness of school children. Acknowledgment is made to Robert J. Myers, Ph.D., Director of the Division of Statistical Research, and to the following members of the section on child growth and development. Division of Statistical Research, for assistance given in the analysis and preparation of the report: Marie G. Fullam, Mollie Orshansky, Helen R. Robinson, and Lois F. Smith. Respectfully submitted. Katharine F. Lenroot, Chief. Hon. Frances Perkins, Secretary of Labor. METHODS OF ASSESSING THE PHYSICAL FITNESS OF CHILDREN Introduction The importance of safeguarding the child's health and physical fitness, not only for his own sake but also for the welfare of society, is axiomatic. It follows directly that if his health is to be protected, the child must be observed at regular intervals in order to ascertain his physical condition and to determine whether he needs medical attention or nutritional advice and assistance. How is such an as- sessment to be made? Various methods are available. Some of them are elaborate, others are more simple; some are based on clinical examination, others depend on exacting laboratory tecluiiques or are derived from anthropometric measurements of the child. It is important to know the relative value of these procedures. \Vliich of the various methods of assessment in use in this countr}' at the present time may be both easy and inexpensive to apph^, and produc- tive of results? This question forms the basis for a study of the physical fitness of 7-year-old white boys and girls ^ living in New Haven, Conn., from September 1934 through May 1936.^ The investigation was under- taken by the Children's Bureau of the United States Department of Labor, the Institute of Human Relations of Yale University, and the Department of Pediatrics of the Yale University School of Medicine, in cooperation with the Board of Education and the Department of Health of New Haven. Before describing the materials and methods of this study, it may be well to define the term "physical fitness" and to outline each of the methods of assessment. Physical fitness is a comprehensive term which is broader than either health or nutrition. It includes the child's general nutritional status and the presence or absence of organic defects, both considered in relation to his general physical condition evaluated in terms of his own previous growth :ind develop- ment. Unfortunatelv, it is ratlipr elusive of definition and difRciilt ' Age is defined in completed years on the last birthday. ' The children were examined for the first time when they were 6 years old, but the report \s based on the observations made when the boys and girls were 7 years of age, amplified and interjireled in relation to the earlier findings. 1 2 ASSESSING THE PHYSICAL EITNESS OF CHILDREN to detormine because of the fact that it evaluates the child's present condition as a functional state which is partly the result of previous growth and which, in turn, will affect his future health and well-being. The child's physical fitness is related to a large number of factors, both endogenous and exogenous. Perhaps most intimately asso- ciated with his physical well-being are the child's present health and nutrition as well as his psychological state, each judged in relation to his heredity, his general disease history, and his previous growth and development. But equally and sometimes evt^n more important is the child's socioeconomic background, because it is safe to assume that if the familj^'s income is below a certain level, the child runs the risk of being physically unfit as a result of unsatisfactory living con- ditions ^ and of a suboptimal intake of the proper dietary constituents. Although it is difficult, if not impossible, to point out all the factors ^ which may play a role in affecting the child's physical fitness, it may be said in summary that the child's well-being is dependent not only on his present condition but also on his previous history (familial, antenatal, and postnatal). Together they determine his ability to compensate for or overcome his present defects and handicaps, and his future incapacities as well. It is obviously impossible at present to appraise correctly a child's physical fitness. Not all the factors that affect a child's well-being nor their interrelationships are known; no satisfactory methods of measuring some of these factors are available; and many procedures which are satisfactory for judging specific aspects of a child's condition are too elaborate for widespread application, or the cost involved pro- hibits their inclusion in a school health program or a community survey. 3 The term "unsatisfactory living conditions" is used to include such factors as overcrowding; short hours of sleep or lack of conditions for sound, restful sleep; poor ventilation; and lack of sunlight. < Home hygiene, as well as discipline and control both at home and in the school, must also be considered. Methods of Assessing Physical Fitness Several methods of assessing physical fitness are in use at the present time. They include the clinical examination, specific biochemical tests for particular nutritional deficiencies, tests of functional well- being, dietary and socioeconomic surveys, and anthi'opometric measures of body build. The Clinical Exaiiiinalion The child's "apparent" physical condition can be assessed by a physician. This examination will obviously reveal such conditions as thinness, faulty posture, flabbiness of muscles, and certain marked nutritional or organic defects, but at best it gives only an appi-oxi- mate evaluation. There is an increasing realization of the difficulties of assessing a child's condition by this method alone. ^ In an article on the incidence and assessment of malnutrition Harris discusses four sources of eiTor.- They are summarized as follows : 1. Inadequate clinical methods are used. Examinations are usually carried out under circumstances which require the inspection of large numbers of children in a relatively short time. Consequently labor- atory techniques or the more refined methods necessary for detecting the presence of early or slight mahuitrition are usually omitted. 2. Debased standards are used. It is commonly known that such figures as those for the average weights and heights of school children of a given age and sex have undergone a steady and marked rise dining recent years. Furthermore, the average weights and heights of groups of children who have been fed on approvetl dietaries have increased beyond the weights and heights of comparable groups of children who were given neither appioi)riate noi- supplemental diets. It is obvious, therefore, that standards which rely on past averages instead of on more recent and comparable data are dependent on the use of a debased norm, and hence, always tend to be underestimates. 3. No standards are known for assessing mahuitrition; consequently no satisfactory way of measuring it exists. ' See Appendix U: (51) Jones, (CO) Lucas et al., (liO) Wilkins. For a more extended discussion, see f/f/> Wilkins. The italicized numbers in parentheses refer to the numbered list of references to be found in Appendix U, pp. 115-121. ' See Appendix II: (W Harris, pp. 225-226. 3 4 ASSESSING THE PHYSICAL FITNESS OF CHILDREN 4. The effects of malnutrition are delayed. Of the failure of a superficial clinical examination to reveal abnormal conditions which may have a delayed effect, Harris says: It is obvious that the grosser results of mahiutrition can generally be detected readily enough by the clinical observer * * * . It is rather the earlier effects, the influence of -partial deficiencies, which may pass unobserved if reliance be placed on cursory clinical examination — and yet these "milder," less readily detected, types of deprivation may none the less have a profound influence, delayed it may be, on the health and well-being of the subject.^ Biochemical Tests for Specific Nutritional Deficiencies Biochemical tests can be made for particular deficiencies including the vitamins, the inorganic elements such as iron, calcium, phosphate, and other important dietary constituents. Many of these tests are complicated and expensive; others, particularly those for the vitamins, lack specificity or sensitivity; all of them have limited value in assess- ing physical fitness because they measure only particular nutritional deficiencies. Functional Tests Under the general heading of functional tests may be included, among others, tests of lumbar pull (on a dynamometer), vital capacity, and basal metabolism.' Many of these teclmiques are still in a devel- opmental stage and require further study and application before satis- factory norms are available. But even when such standards have been developed, giving the tests will require special traming or equip- ment and will, therefore, be beyond the reach of the average public- health officer or school official. Furthermore, they give only a partial answer to the question "Is the child physically fit?" 3 See Appendix II: (44) Harris, p. 225. * (87) Pryor and Smith give a brief review of the history of strength tests, including the use of the ergo- graph and Kellogg's universal dynamometer, and Sargent's chinning and dipping tests to judge endurance. Gallon's tests of reaction time to auditory and visual signals are also mentioned. For illustration of the use of tests of vital capacity and tests with the spring dynamometer, see Appen- dix II: {19) Cheesman, {118) Warrington [England], reference to Magee in (44) Harris, p. 227, {69) Milligan, and {128) Woolham and Sparrow. See {118) Warrington [England] for application of the Romberg test for determining a child's power of equilibration and coordination and {1S7) Woolham and Honeyburne, for pulse and respiratory tests. For attempts to use standard athletic performances such as the 60-meter race, ball throwing, and jumping as indications of physical fitness, see {li9) Wroczynski, p. 673. Standards based on combinations of various tests of physical performance have also been derived. For example, a physical-fitness index has been develoi)ed by {9g) Rogers. See also {128) Woolham and Sparrow for a physiological formula developed by Flack and Woolham. METHODS OF ASSESSMENT 5 Dietary Inquiry A detailed inquiry may bo made into the dietary habits and en- vironment of the child in order to estimate his nutritional status.* Wilkins, pointing out the fimdamental importance of this aspect of the probk^m, writes: Workers in animal nutrition know only too well that a diet must be tested, not only through the whole life of the individual, but through at least three or four generations before it can be passed as fulh^ adequate. Yet it is customary to pass children, the most valuable animals of all, as normal in their nutrition, and, presumably, therefore to pronounce their diets fully adequate in each and every constituent, without any inquiry into the details of their diets, and with- out even testing the efficiency of a single bodily function. ^ Making detailed dietary inquiries is, however, difficult. They are also expensive and time-consuming, for they require the services of well-trained observers and the whole-hearted cooperation of the child and his family. Socioeconomic Inquiry An estimate of the net family income per capita gives a fairly reliable basis for estimating whether a satisfactory diet is purchasable, because, as Wilkins has pointed out, "Enough work has been done on family budgets and minimum costs of living to enable us to make a rough estimate of the nutritional possibilities of any family." ^ On the other hand, although it is inevitable that if the money avail- able for food is inadequate the child's diet is unsatisfactory and his health suffers, it should be pointed out, as is stated in an editorial which appeared in the Journal of the American Medical Association, that "malnutrition is no longer considered to be exclusively an out- come of poverty or bad environment." ^ Antliroponielric Measures of Body Build Because of the dilliciillics involved in the applicaiiiui and the appraisal of the techniques that have been discussed, attempts were begun to derive indices of body build, based on tlic intcrri'hitions of » See Appendix H: (//) Boudreau and Kruse, (H) Harris, (89) Roberts, and (Itl) Wilkins. For discussion of what constitutes an adequate diet, see .\ppendix U: (S6) Friend, (it) Great Britain Min- istry of Health Advisory Committee on N'utrition, (Gl) McCarrison. (P^i Rose. (Id) Sherman, and (;09) Stiebeling and Clark. • See Appendix H: (HI) Wilkins, p. 145. ' Sec Apixuidix II: (Itl) Wilkins, p. 146. Sec also (U) Harris and reference to British Medical Association Committee on Nutrition in (44) Harris, and (89) Roberts. « See Appendi.\ II: (47) Indexes of Nutrition, p. 1286. 6 ASSESSING THE PHYSICAL FITNESS OF CHILDREN certain anthropometric measurements.^ These indices are used to predict a measure of the child's well-being, such as his weight or arm girth in terms of his physique or body build. They assess physical fitness only insofar as it is related to abnormalities in such measures. Four of these indices, which have received widespread application in the United States, have been applied to the children included in this study: ^° (1) The Baldwin-Wood Tables published in 1923. (2) The ACH (arm-chest-hip) Index developed and published by the American Child Health Association in 1934. (3) The Nutritional Status Indices of Franzen published by the American Child Health Association in 1935. (4) The Pryor Width-Weight Tables published in 1936. These indices are based directly or indirectly on statistical techniques (multiple-regression procedures) which are not always employed in the development and application of other methods of assessment. For this reason it is probably advisable to describe in some detail the fundamental hypotheses on which they have been based, the possible limitations of this type of approach, and the exact methods of applying and evaluating each index, and to review critically other studies which have been made to test the effectiveness of these indices in identifying children who may be in need of medical care or nutritional advice and assistance. 9 This type of index is to bo distinguished from simple ratios, which have been in use for a great many- years, although they arc not generally applied in the United States at the present time. As early as 1829, r Weight Buffonreportedsuchanindex or ratio in the form, -—-T-T-T-j- It was modified by Quetelet in 1836, by Rohrcr in 1908, by Tuxford in 1917, and by Bardeen in 1920. Tuxford's modification was more elaborate than the others. He proposed the following formulas as a measure of the physical development of school children: Forbov<;- Weight (kg.) ^381- age in months Height (cm.) ^ 64 „ , Weight (kg.) 381— age in months ^°^Sirls: Height (cm.) ^ 48 During the World War Pirquet developed a system for assessing nutrition which included a ratio based on weight and stem length. It is known as the "pelidisi" index and depends on the relationship YWci'rht> 'Y' Stem leni htXlO gth Other measurements have also been used in (ieriving this type of index. For example, as early as 1886, Bernhardt used the following formula for predicting weight: ,. . , HeightXThest monsurement Weight= ^^ and in 1901, Oppenheiuier put forth a "nutritional quotient" defineti by tlie ratio Maximum girth of rela.\ed armXlOO Chest circumference at end of expiration Other indices may be mentioned which illustrate the use of combinations of tests of physical performance. One is a physiological formula known as the Flack-Woolham product (1923). The second is an anatomical ratio called the Dreycr product (1919). For a more complete description see Appendix II: (S*) Paton and Findlay, pp. 51-57, on which most of the abf)ve discussion is based. See also (£7) Dreyer, (89) Roberts, pp. 57-83, (Hi) Tuxford, and (ISS) Wool- ham and Sparrow. '0 Sec Appendix II: (6) Baldwin and Wood, (S6) Franzen and Palmer, (76) Nutritional Status Indices, and (86) Pryor. Indices of Body Build as a Method of Assessment of Physical Fitness A number of problems arise in studying these indices of body build. For example, the question is frequently asked: Can external measure- ments of the human body be used to assess a child's physical fitness, especially if there is an early or slight departure from health or if he is suffering from a mild form of nutritional disorder? Can they identify the child who is overweight but flabby or the one who is overdeveloped? It has been pointed out in this connection that such indices attempt only a quantitative evaluation of the child's condition and that the qualitative aspects are ignored. To be more specific: Can an index of body build differentiate between organic and dietary causes of undernutrition? Does it take inherited or constitutional factors into account? May not long-continued mal- nutrition have interfered with a child's growth to the extent that anthropometric measurements cannot be used to evaluate his physical fitness? It is also well to bear in mind that recent studies of weight and weight increments have shown that certain years are "good growing years" and others are not.' Does this phenomenon affect the value of the indices? Can seasonal variations in the measurements from which they have been derived be ignored? Many other questions have been asked. A discussion of their significance has been omitted, however, since this investigation is concerned with the identification of children who may l)e pliysically unfit and not with detailed technical problems in methodology. Nevertlieless it may bo wortii while to indicate the nntiire of some of these questions. For example, a great many statistical problems arise in connection with the development and application of these indices. Do the basic data comply with the hypotheses which the mathematical procedures require; i. e., are the anthropometric measures normally distrihiited and is their relationship linear? Are tlie inallieinatical procechires correctly applied? Is the technique (>qiially applical)le to botii sexes? To children of different ages? Is the definition of age used in deriving the indices too broad? Is an index more satisfactory if it is based on longitudinal instead of cross-sectional tlata? In other words, if 1 See Appendix II: (77) Palnior, p. 14.13, which refers to earlier pajxT by same author (80). 8 ASSESSING THE PHYSICAJ^ EITNESS OF CHILDREN an index for 6-year-old and also for 7-year-old children is derived from successive measurements of the same group of boys and girls, is it more reliable than if it were based on measurements of one group 6 years of age and another group 7 years of age? In applying these indices the child's observed measure is judged in relation to the average or expected measurement of his skeletal peers. For example, the Baldwin-Wood Tables identify as skeletal peers children who are of the same sex, age, and height. These three factors are interrelated in such a way as to determine the child's expected weight, which is in reality the measurement for an average child of the same sex, age, and build (height). Now, such an average is dependent on the kind of children whose measurements have been included. If one assumes that the observations from which such an index has been developed are also representative of the group to which it is applied, the question still remains — how far does the average represent the normal, according to a dictionary definition of the latter term as applied in the medical and biological sciences — "conforming to natural order or law." ^ It should be remembered in this connection that the average anthi'o- pometric measurements of school children of a given sex and age have shown a steady increase during the last 15 or 20 years. ^ Does this evident change in physique necessitate the construction of new indices at certain intervals? Likewise, it has been shown that in experimental trials the average weight and height of groups of children who have been fed on improved dietaries have improved beyond the weight and height of the controls or of previously comparable aver- ages.* Do these findings indicate the use of debased standards? One may also inquire: What are the standards of normal variation? How much may a child's measure deviate from the average and still be considered satisfactory? Is this deviation to be expressed in absolute or relative terms or by means of more elaborate statistical procedures? It is essential also to know the kind of children whose measurements were used for deriving the indices. In other words, what racial, socioeconomic, and geographic groups do these boys and girls repre- sent, and are they healthy children? In this connection it is inter- esting to note that the White House Conference of 1930 pointed out that most of the avaihible standai'ds do not represent (h^sirable combi- nations of heritage, history, and home influence, and that further 2 See Appendix U: (S8) OouM, p. %1. 3 See Appendix U: (IS) Brewer, p. 91, (Stl) Friend, pp. 27-28, 73, (4O) Oreat Britain Board of Kducation, p. 25, (48) Jacob, and (116) Wallsend [KnKland] Kducation ( 'onmuttee, p. U. See also references to Rowland and Stockwell fonlained in (.?9) Great Britain Board of Education, p. 23, and see (tB9) Wroezynski, pp. 59.'>-.59(), (178, for references to Fessard, I,aufer and 1-aujrier, Koch, and Wolff. ' See Ajjpendix II: (44) Harris, p. 22r). He refiTs to studies by Corry Mann, Leighfon and Clark, Orr, and SchartI and Sinnadorai. Sec also (69) Loewenthal, (90) Roberts et al., and (105) Simpson and Wood. INDICES OF BODY BUILD IN ASSESSING NUTRITION 9 biometric work will be necessary to furnish norms for promoting useful application of measurement.^ Finally, in evaluating an index it is important to consider the chil- dren whom it fails to identify as well as those whom it selects as likely to be physically unfit, for although an index may identify some children who are in need of medical attention or nutritional advice and assistance, it may fail to select an even larger number of such children than it identifies. On the other hand, it may select not only all the children who are physically unfit but also a large number of hefilthy boys and girls. In other words, it is important to know just how efficient such an index is. The preceding discussion of the limitations of this type of approach, based on the prediction of one measure of the child's fitness in terms of its relation to his physique or body build, has not attempted to review critically the questions which have been raised, because this study is not concerned with the more technical aspects of evaluating these indices as a method of assessing physical fitness.® * See Appendix 11: (119) White House Conference on Child Health and Protection, p. 323. * It is suggested that the reader who is interested in the more technical aspects of evaluation consult such papers as those of (9) Bigwood, pp. 172-173, (.?4) P'ranzen, (49) Jenkins, (.50) Jones, (65) Marshall, (78) Palmor, (96) and (96) Bosenow, and (117) Warner, listed in Appendix II. The Four Indices of Body Build Included in the Study Before giving a detailed description of the four indices included in this study, it may be well to reemphasize the fact that they evaluate physical fitness only insofar as it is related to abnormalities of certain measures of the child; for example, his weight or arm girth judged in relation to his body build. Description of Each Index of Body Build THE BALDWIN- WOOD TABLES The index of body build based on the Baldwin-Wood Tables estimates a boy's or girl's weight in terms of his or her height (to the nearest inch) used as a criterion of body build at a given age (taken at the nearest birthday). Tables have been derived for ages 5 through 19 for boys and 5 through 18 for girls from which it is possible to de- termine what the child should weigh for his sex, age, and height (used as a criterion of body build). Then the difference between his observed and his expected weight is expressed as a percentage of the expected weight. Baldwin ^ has allowed a deviation of less than 6 percent from the average or expected weight for any height, age (under 10 years) ,^ and sex for individual variations. A larger deviation (6 or more percent) indicates that a child is likely to be in need of medical attention.^ Measurements of an Italian child, A. R., who was included in the study, may be used to illustrate the computation of this index. (See sample schedule of physical measurements.) It may be seen from this sample schedule that the Italian boy, A. R., weighed 46 poimds and was 44 inches tall when he was 7 years of age (7 years, 1 month, 6 days). According to the Baldwin- Wood Tables the average 7-year-old boy who is as tall as A. R. weighs 44 /40-44\ pounds. In other words, A. R. was 4.5 percent overweight ( "TT" ) in terms of the Baldwin- Wood standard. 1 Sec Appoiidix U: (,5) Baldwin, p. 4. 2lf tho child is 10 or more years of age, Baldwin has allowed loss than 8 percent variation. The majority of workers, however, have allowed under 7 or less than 10 percent for children of all ages. See, for example, in Appendix U: (SX) Clark, Sydenstricker, and Collins, (SS) Dublin and Oelihart, and (SO) Emerson and Manny. 8 Accordint; to Maldwin, "childnui who are l.l percent overweight for their heigtit and age may also he in need of medical attention" (see Appendix II: (6) Baldwin, p. 1); but this aspect of tho problem of physical fitness has not been considered in the present report. 10 FOUR INDICES OF BODY BUILD DESCRIBED 11 PHYSICAL MEASUREMENTS Children's Bureau, U. S. Department of Labor Institute of Human Relations and Pediatric Department of Yale University Name A.R. Race Italian Examination Number 277 Address Main Street Boy Girl Born: Year 28 Month 9 dav 4 Date 10/9/34 4/8/35 10/10/35 School Harbor Harbor Harbor Grade I I II Age 6/1/5 6/7/4 7/1/6 Weight (lb.) 41| 43i 46 Height (in.) 42i 42i 44 Arm girth (cm.) Flexed 17.5 18.2 18.4 Relaxed 16.3 16.5 16.6 Total 33.8 34.7 35.0 Subcutaneous tissue * I 11 11 lOi II lOi 11 lOi Total 21i 22 21 Chest breadth (cm.) Inspiration 18.1 18.1 18.5 Expiration 17.2 17.5 17.9 Total 35.3 35.6 36.4 Chest depth (cm.) Inspiration 13.6 13.7 14.5 Expiration 12.6 13.0 13.7 Total 26.2 26.7 28.2 Hips (cm.) Trochanter 19.9 19.9 20.4 Crest 19.4 19.6 20.2 Date 9/25/34 1/23/35 5/27/35 9/18/35 1/23/36 5/27/36 Weight 40i 43i 43i 44 47i 48 * The 2 measurements for subcutaneous tissue were made according to the technique described in Ap- pendix II: (75) Nutritional Status Indices, pp. 8-10, using a special subcutaneous-tissue caliper which has arbitrary units. THE ACH (AKM-CHEST-IIIP) INDEX The ACH Index was developed by the American Child Health Association in 1934 and is "a screenins: technique constructiMJ to nwot the demand for a practical method of selcclinLr. from a lai\ire ^roup, those chikh'en who need a medical e.xaminati(Ui because they have extremely small amounts of musculature and subcutaneous tissue."* More specifically, "wlien the case is selected lie is certain t«> lia\(' an arm-girth deficiency and very likely to be deficient in subcutaneous tissue and wcit^ht as we " 6 » See .Vppondix II: (75) Nutritional Status Indices, p. 1. • See .\ppendi.x II: (55) Franzon auil Palmer, p. 11. 239848° — to- 2 IJ ASSKSSINU rilK IMIVSU'AI- I'lTNKSS OK CHlhOKKN Haslo (jihlos Uiwc lu>(Mi propaivd for chiKlion of oach sox n^xod from 7 tlirouirh \- ymii's (^aii(> at last Mrihilay'* win* \u\\o hip widths rauirint; from hoh>\v 20.0 om. to '2W0 rm. ai\il abovo. Thcst* tahlos irivo tlio minimum ilillVivuro hotwocn ana uirth ami rhost dt>pth ' (^A(i — (^l)") fof a uiviM\ liip witltli (at th(^ tforhanttMs"). If tho ihlVtMiMuu* botwoon a rhiKl's arm irirth ami his rhost ilc^pth is oxaotly iHiual to or loss than iho \ahio ,ii'ivtM\ iti tho tablo, {\\o bov or uirl is itltMititioci as in nooii of furlhtM" o\anun:iii(>n. On iii<> otl\«M- haml. it iho ditforiMU'e is greater lh:in this minimnm, tiie chiKl is not soi(H'ttHl. 'VUc oriii'inal AcMl Inilrx was sot \o solooi "'about 10 poroont of a gn>iip ^,in a brivail. i;on»Mal samphi\ii'\ About 00 piMOiMit of this selec- tion Mr<< oxtronit^-ili^ieet oasi>s ami over SO percent are either extreme iiefeets or thosi> that border on extreme defects. It dt>es miss somo extreme-iU^feet eases and tht^se omissions ari> dt^liberately saeritieed in the interest of speed and simplieity of measurement." ^ The scale has alsi> luHM\ set \o st4(H't l-t, 20. or "Jo percetit of a i;i\nii> of boys or irirls. ' The "Jo-^HMvent sehnM ion contains nearly all vJibout 00 percent') of the extriMue direct eases that can be identified if the more elaborate Nutritional Status Indices (described on pp. 13-U^'> had been applied [o the entire iifoup i>f ehildrtMi in the llrst place, but it also includes an .MppriH'iable nmuber of ehildrtMi who dv> not hav(> extreme do- feols. lt\ other wiuds. it is probably ailvisable to apply the more rehiuHl methiHl, nanu^ly, the Nutritional Status ludici^s, \o the pre- limii\ary AC 'I I "Jo-percet\t selection. Thinmh the index screens out the ehddreu with "'extremely small at\u>unts of soft tissue » * * it does notiuake individual distinc- tions betwvHMi the children. Neither does it discriminate, for a pveu chiKl, between a deliciency in muscula(uro as opposed to a deticiency m subeutai\eous tissue. * * * '\l\^s \('\\ Imh^x was ititentionally so restricted in t>i\ler that tht^ piactical time limits for mensurement in a school or i>ther irrvnip situatioti would ixoi be exceeded.""^ Accorilinj; to I'Van/.en and Palmer; "The use that is to be made of the index depends iipoi\ the objectives sought. If we wish to be surt> to iilentify all extrtMue-ilefect cases, then the screen method [2.">-percent selectionl with the additional measures is indicated. If. on the other hand, this is not feasible and we are content merely with kmnvinvi" that the cases are for the most part severe defects and do warrant review by a physician, then the simpler * * * method |U^-peivent selectionl answers the purpose." The index used in this way *'is intemled both as an aid to the physician at\ii as a n\ass ^< VHliitvt rv>|lar\^s»^nt tlH< sum \>f two iiutisiun'tuonts: Oh<'^( vU^ptU. that is. th«> ttit>asur\nm^nt$ Kvr ((>xim«»«>i» :r u is. llv i«uH«sui\>ut(>nts i>f UhMii^ivr itftn. flfxtKl tuut nji«''Uvli\ U: v'>'' Nutrltivxusil Status liuliwvx. i>. «W. >• S*!*' ViHvtulix U: v?*^ Nutritlw>»l Status ImUcws iv a. FOUR INDICUOS OK IU)1)V lU; 1 l.D l)HS('RIBi:i) \',^ selective measure tor (■oiupnriii'i: *2:r()ii])s."^' If desired, (he index may also be set to identify 14 |)ereeii( or 20 percent of (he childn-n. The calculation of Ihe ACII Index can nlso he ilhisti-atcd for (he Italian boy, A. R. Tlu^ sanipK' schechile of physical nieasnicnients (see p. 11) shows that at 7 years of ast-depth measurem(>nt of 2S.2 cm. (sum of 14.5 cm. inspiralion and 18.7 cm. e\j)iration). The din'erencebe(\veen his arm <;ir(h and his chest (lej)(h is (i.S cm. The \vid(h of his hi])s at (he lioehanteis was 20.4 cm. The tables show that for a boy of 7 years with a hip width between 20.0 cm. and 20.4 cm. (he minimum dilferenee between the arm-o:irth and chest-depth measui'enients is 0.0 em., if 10 perci'iii of (he iii-oup of boys are to be identified ; '-' ().;') em., if 14 |)ereent ; '^ 1.0 cm., if 20 i)ercent;'^ and 1..") cm., if 2.") pereenl '' of (he boys are to be selected for either further nu'asurement or direct reference to a physician. As the diffc'rence Ix^twcen A. R.'s aim <;irth and chest dej)th (6.8 cm.) is larger than any of these minimum dilferences. it follows that he does not fall in the lowest 10, 14, 20, or 25 peicent of the boys in respect to a deficiency in arm <>:iith and probably a d(>- liciency in subcutaneous tissue and weight as well." THK NITHITIONAL STATUS INDICES Tables for the Nuti'iliomil Status lndic(>s were developcHi by Franzen and his coworkei's in the Anieiiean (Miild Health Association in 1935. They may be used to evaluate a child's condition from 7 tlirouf2:h 12 years of a<2:e (age at last i)ii"th(lay) in terms of his weight, muscle size (as indicated by the girth of the upper- arm), and (he amount of subcutaneous tissue (nu^asuiVMl oxov the biceps). I^ach of (hese three measures is compared wi(h the weight, ar"m girth, oi* subcutaneous tissue which the average child of (he same sex and age is expec(ed to have for his skele(:d bmld, judged in (crms of his heigld, (he widlh and (l,'p(h of his chest, and his biti-ochanteric width. These comparisons nnd^e it possible lo determine the child's stand- in«r iu wein and I'ulniiT, p. U. 14 ASSESSING THE PHYSICAL FITNESS OF CHILDREN This procedure was devised not only to reveal differences in the nutritional status of children ^^ but also to discriminate between a deficiency in musculature and a deficiency in weight or subcutaneous tissue, "a distinction that is important if differential diagnosis of nutritional status is to be made." ^^ The indices "are not intended to supplant the physician in the diagnosing of malnutrition. Rather, they furnish the physician with more adequate evidence for his diag- nosis than he has had in the past. Using these indices, in conjunction with other clinical signs, the physician can diagnose the condition and prescribe what shall be done." " The score card for Nutritional Status Indices shows the steps necessary for the calculation of the three Nutritional Status Indices. The computations for the weight index may be found on page 16. The procedures for evaluating the other two — namely, the child's arm girth and subcutaneous tissue — are not described because they in- volve similar calculations. 15 Defective nutrition is defined by Franzen and his coworlcers as "a low amount of arm girth, subcu- taneous tissue and weight, each for skeletal build. Our definition is specific, but it represents a condition which includes many other subjective signs." See Appendix II: (83) Physical Defects; the pathway to correction, p. 63. In 1935 Mitchell, who was also associated with this group, added: "These measurements compared with a random 1,000 children of the same skeletal build are not in themselves an evaluation of nutritional status, but they do give reliable, objective, and valid distinctions in weight, musculature, and adiposity which are significant in such an appraisal. They are indices of physical signs which should be properly evaluated in a composite of signs and symptoms. Of course, deviations from an average should not be interpreted directly as desirable and undesirable signs, but the average provides a convenient reference point which gives definiteness to the measurements. The use of accurate distinctions in these three physical signs releases the clinician from the difficulties of individual judgment and gives him greater freedom to apply all the subtleties of the art of medicine in judging function, growth, and development and all the intricate factors involved in the nutritional process." See Appendix II: (,71) Mitchell, p. 319. 16 See Appendix II: (76) Nutritional Status Indices, p. 2. I' See Appendix II: Ibid., p. 5. FOUR INDICES OF BODY BUILD DESCRIBED 15 Name SCORE CARD FOR NUTRITIONAL STATUS INDICES i A.R. Sex M Age 7 School Harbor Room Grade II Arm Girth MEASUREMENTS Flexed Relaxed Subcutaneous Tissue over the Biceps Weight Height Chest Breadth Chest Depth Width of Hips Expiration Inspiration Expiration Inspiration a 18,4 b d e 16 6 10 5 10 5 i 17.9 i 18.5 1 13.7 n 14 . 5 Date 10/10/35 c 35.0 f 21.0 R 46.0 h 44.0 k 36.4 28.2 o 20.4 COMPUTATION OF INDICES Arm Girth Sub. Tissue Weight + Value Value + Value Value + Value 1 Value Measure 99.1 62.2 118.4 Height 11.5 10.0 1 8.2 Ch. Breadth 11.3 1.8 '12.4 ( Chest Depth Hip Width 14.3 1 4.8 11.7 16.2 18.3 22.4 Adjustments for age 0 1.8 1.5 Sums +110.6 -41.8 +72.2 -26.7 +118.4 -56.2 Indices 68.8 45.5 62.2 No. in 1,000 with smaller indices 964 309 885 1 This form is presented in (75) Nutritional Status Indices, p. 15 (see Appendix II). 16 ASSESSING THE PHYSICAL FITNESS OF CHILDREN The detailed steps in the determination of this weight index for the Itahan boy, A. R., are presented at this point. The exact wording used in the instructions given in the monograph, Nutritional Status Indices, has been followed as closely as possible. a. Look up A. R.'s weight under Weight Reading in Table C for 7- to 9-year-old boys (see p. 24 of the monograph, Nutritional Status Indices) ** and record the corresponding table value on the line opposite "Measure" on the score card (score card for Nutritional Status Indices) in the plus (+) column under "Weight." The table value for A. R.'s weight of 46.0 lb. is 118.4. b. Look up his height under Height Reading in Table D, p. 25. Opposite the reading are three values. Select the one for weight and record this value in the line opposite "Height" in the section under the proper heading on the score card. Since the value is negative it is to be entered in the minus (— ) column. A. R.'s height was 44.0 in., which has the following table value for weight: —8.2. c. Look up his chest-breadth measurement in Table E, p. 26. Record on the score card the table value given for weight and enter this value opposite "Chest Breadth" under the proper index heading in the column with the appropriate sign. A. R.'s chest breadth of 36.4 cm. has a table value for weight of —12.4. Since this is a negative value, enter it in the negative column. d. Similarly, obtain the values for his chest depth. Table F, p. 27, and for hip width, Table G, p. 28. The table values for A. R. are written on his score card (score card for Nutritional Status Indices). e. Look up his age in Table H. Since A. R. is a 7-year-old boy, 7 but not 8, the table value for his age to be used for his weight-index calculation is —1.5. Enter this value on the line opposite "Adjustments for Age" under the proper index heading in the column with the appropriate sign. /. Add the figures in the plus column for the weight index and record the total in the plus column on the line opposite "Sums." Likewise, add the figures in the minus column and record the total in the minus column on the line opposite "Sums." These two sums for A. R.'s weight index are -H 118.4 and —56.2. g. Subtract the sum of the values in the negative column (56.2) from the sum of the values in the positive column (118.4) and record the difference (62.2) in the line opposite "Indices." This score, 62.2, is A. R.'s nutritional index for weight and is strictly comparable with his indices for arm girth and subcutaneous tissue. By referring this score or index to Table X of the monograph, Nutritional Status Indices, p. 65, A. R.'s standing in weight relative to others like him in skeletal build may be obtained. Using Table X, look up 62.2, A. R.'s weight score. The table value opposite this is 885, the number in 1,000 children of the same sex, age, and skeletal build who have smaller weight indices (scores) than A. R. In other words, among a general sampling of 1,000 boys of the same age and skeletal build, 885 in 1,000 weigh less than A. R.'s In order to facilitate comparisons with the other indices included in this stud}^, Table X of the monograph, Nutritional Status Indices, has been modified to show the number of children in 100, instead of in 1,000, having less than a given score in one of the Nutritional Status Indices. When this revised form of Table X is used 88 (88.5) boys in 100 of the same age and skeletal build weigh less than A. R. " See Appendix H: Ibid., p. 24. " This description of the method of eompiiting the Nutritional Status Tnde.x for Weight is based on the illustration given on pp. 14-18 of the monograph, .\utritioual Status Indices. Sec Appcndi.x II: (75). FOUR INDICES OF BODY BUILD DESCRIBED 17 THE PRYOR WIDTH-WEIGHT TABLES The Pryor Tables estimate the average weight of a boy or girl of a given age (at nearest birthday) and body build, judged in terms of his or her height (to the nearest inch) and bi-iliac diameter. The calculation of this index also may be illustrated for the Italian boy, A. R.: According to the sample schedule of physical measure- ments (see p. 11), when A. R. was 7 years old his height was 44 in., his weight was 46 lb., and "the width of his iliac crest" was 20.2 cm.^ As the Pryor Table for T-j^ear-old boys does not give the expected or average weight for a boy who is 44 in. tall and has a crest measure- ment of 20.2 cm., it is necessary to approximate A. R.'s expected weight. The following method is used : The table gives the average weight, 42 lb., for a boy of the same height as A. R., who has a bi-iliac diameter of 19.3 cm. (0.9 cm. less than A. R.'s). It also gives the expected weight, 46 lb., for a boy of the same height who has a diam- eter of 21.5 cm. In other words, an increase of 2.2 cm. (21.5 cm. — 19.3 cm.) in the width of the iliac crest — to use one of Pryor's terms — corresponds to an increase of 4 lb. in the average weight of a boy of this age and height. A. R.'s crest measm-ement is 0.9 cm. larger than the smaller of these two diameters given in the table and corresponds approximately to a weight increase of 1.6 lb., which is added to the average weight, 42 lb., of a boy with a bi-iliac diameter of 19.3 cm., giving an expected weight of 43.6 lb. for a child who was weighed with his clothing removed. As the boys and girls included in this study were weighed wearing theh' clothing, an allowance of 2 lb. for the weight of the clothing must be made for chikhen who are more than 40 in. tall.^^ Therefore, a 7-year-old boy with the same height and crest measurements as A. R. weighs, on the average, 45.6 lb. (43.6 lb. + 2 lb.), compared to A. R., who weighed 46 lb. In other words, according to this index, A. R. was 0.4 lb. overweight (46-45.6), As the Pryor Tables do not define the amount of variation m weight which is within normal limits nor the child's relative standing in weight, a procedure has been developed which is comparable to Table X ^^ in the monograph, Nutritional Status Indices.-^ It gives the number in 100 ohilthen of the same sex. age, and body build who weigh less than any indivichial child. When this method of scoring is used, 54 boys in 100 of the same sex and hiiiid weigh less than A. R. 20 Pryor uses the followinK tonus inttrchanRiably: Width of iliac crest, bi-iliac diameter, width of |)clvic crest, bicristal diameter, and width of crest of ilium. « See Appendix II: (86) Pryor, p. H. » See Appendix II: (75) Nutritional Status Indic<>s. p. 65. " The exact method of derivinp and using this method of scorinp is descrihod in detail in Apiiendix I, pp. 97-98. It was approved by Pryor in a personal communication dated Fob. 3, 1938. 18 ASSESSING THE PHYSICAL FITNESS OF CHILDREN Limitations of the Four Indices of Body Build In outlining the limitations of measuring the child's fitness in terms of his body build, certain questions which refer either directly or in- directly to the four indices included in this study have been omitted intentionally from the discussion. For example, Wilkins in an article entitled "The Assessment of Nutrition in School Children," which appeared in The Medical Officer in 1937, points out a "subtle error" in the standard based on the relation of weight to height; namely, that "the well-nourished child, while both taller and heavier than the poorly nourished, is at the same time lighter in relation to its height than the less well-nourished. In other words, the less well-nourished child is often heavier in relation to its height than is the better- nourished child. This disturbance of the natural relation of weight to height is, of course, the result of faulty nutrition affecting growth over a considerable period." ^'^ Various workers have also raised questions concerning some of the measurements used for the Nutritional Status Indices. For example: Is the amount of subcutaneous tissue which can be picked up asso- ciated with the size of the finger tips? Can one pick up two layers of the skin of a very fat child, with the calipers set at 30 to standardize the size of the bite? ^^ Are the fat and muscle of the upper arm repre- sentative of the subcutaneous tissue and musculature of the entire body? Many other questions arise which one would like to study. For example: To what extent is an index prognostic as well as diagnostic? What happens to an index during an acute illness? Is an index related to a child's gain in weight or to some other measure of his growth and development? 2« See Appendix H: (121) Wilkins, p. 146. "The measurements "are taken with the special subcutaneous-tissue caliper. The examiner stands to the right of the child and measures the bare right upper arm. "Ask the child to extend the bared right arm at right angles to the side of the body, and then to flex the arm and 'make a muscle.' Mark with a pencil the highest point on the biceps muscle. Then have the child relax and lower his arm. (This highest point on the biceps muscle is the same level at which the girth of the upper arm is measured and, in actual practice, is marked at the same time the arm-girth measurements are taken.) Through this 'biceps point' draw a short line across the arm. This line furnishes a reference point for taking the mea,surements. "The size of the 'bite,' or the amount of tissue that is grasped, has considerable influence on the accuracy of this measurement. In order to standardize the amount of tissue grasped, the size of the 'bite' is deter- mined by the caliper itself. Take the caliper in the right hand and open it to a reading of 30. Then, using the 'biceps point' as a center, straddle the blades of the caliper, equally on each side of the 'biceps point,' with the lower edges of the blades touching on the horizontal line which has been drawn on the arm. With the left hand above the caliper, i)lace the index finger and the thumb of the left hand in contact with the outer surface of the caliper blades, remove the caliper, and pinch up the iiiiiouiit of ti.ssue in through 14 years (age at the nearest birthday), including 3,421 7-year-old children, and concluded that underweight and overweight standards should vary with sex and age. See .VpjX'ndix 11: (HI) Faber. " See Aiipendix U: (iS) Clark, Sydenstricker, and Collins, p. 521. 3« The Woodbury Tables for boys and girls from birth to school age were used for the very young children. See Appi'ndix U: (1S5) Woodbury. '• See Appendix II: («8) Dublin and Qebhart, p. 4. 22 ASSESSING THE PHYSICAL TITNESS OF CHILDREN Wood standard as being underweight. For example, there were 77 7-year-old boys who, according to the physician's judgment, were undernourished. The Baldwin-Wood Tables failed to identify 52 (67.5 percent) of these 77 children as underweight. In other words, using a 7-percent limit for percentage underweight, the tables selected only 32.5 percent of the boys whom the physician had diagnosed as undernourished . A Comparison of the Selection Made by the Physician's Diagnosis of Nutrition and by the Use of the Bnldioin-Wood Tables ' BOYS WELL-NOURISHED Age UNDERNOURISHED Doctor's diagnosis Weight table 7 percent limits Percent agreement Doctor's diagnosis Weight table 7 percent limits Percent agreement 137 130 94. 9 77 25 32.5 GIRLS WELL-NOURISHED Age UNDERNOURISHED Doctor's diagnosis Weight table 7 percent limit ■' Percent agreement Doctor's diagnosis Weight table 7 percent imit s Percent agreement 142 134 94. 4 99 25 25.3 ' Adapted from table I, p. 5, in Dublin and Gebhart (,28) (See Appendix II). 2 A child is identified as well-nourished if he is less than 7 percent underweight or if his observed weight is more than his expected weight. 3 A child is selected as undernourished if he is 7 percent or more underweight according to the standard. * Age at nearest birthday. In discussing these results Dublin and Gebhart conclude: "A method which misses three-fourths ^^ of all of the children whom a competent physician, after a thorough examination, would call under- nourished has certainly scant value even as a 'rough index for sorting out the most needy cases.' " They attribute the failure of the index in part to the fact that "these Italian children are heavier for each inch of height than the children used in the Wood-Baldwin-Wood- bury Table> ^s In Mitchell's study (1935) the Baldwm-Wood Tables and the Nutritional Status Indices were applied to only eight children, in- cluding two 7-year-old boys,^^ James and Edward, who had been referred by their teachers to a physician for special examinations. James weighed 43 lb. and was, according to the mvestigator, ob- viously undernourished. He was so unusually low in weight, muscu- lature, and adiposity that he appeared thin and frail and suggested poor nutrition to the casual observer. The clinical examination also showed very little flesh over the chest, well-defined wdnged scapulae, and soft, flabby muscles. According to the Baldwin-Wood Tables " The tables failed to identify three-fourths of the children of all ages (2-10 years, inclusive); two-thirds of the 7-year-old boys and three-fourths of the 7-year-ol(i girls. s« See Appendix U: (m Dublin and Gebhart, p. S. '» No statement is given as to how age was defined, nor an indication of the race, or the location of the homes of these children. FOUR INDICES OF BODY BUILD DESCRIBED 23 he should have weighed 50 lb. and was, therefore, 14 percent under- weight. In terms of the Nutritional Status Indices only 2 in 100 boys with the same skeletal build weigh less than James; 6 have a smaller arm girth and 20 in 100 have less subcutaneous tissue.'"* Edward weighed 47K lb. The investigator described him as of about average height with average chest breadth and a shallow chest and narrow hips. Both arms and legs appeared very thin and his muscles were very soft and flabby. The child was one of nine chil- dren; the family was entirely dependent on public relief; investigation showed a diet of excess starchy foods which may have accounted for his adipose tissue. During 6 months' observation he gained only ji Ib.'^ According to the Baldwin-Wood Tables he was 10.4 percent underweight. In terms of the Nutritional Status Indices, 7 boys in 100 with the same skeletal build have a lower weight; 9, a smaller arm girth; and 50, a smaller amount of subcutaneous tissue. To summarize: The Baldwin-Wood Tables identify both James and Edward as markedly underweight. According to the Nutritional Status Indices also, both these children were underweight; they had exceptionally small arm girths; and James had a deficiency in sub- cutaneous tissue as well. These findings indicate that the Nutritional Status Indices may be more useful than the Baldwin-Wood Tables because they evaluate not only the child's weight but also his muscu- lature and subcutaneous tissue. It should be pointed out, however, that Mitchell tested the indices on only a small number of children, and there is no evidence in this study to indicate that the Nutri- tional Status Indices will identify other children who are poorly nourished. It is difficult to evaluate and compare these three studios for five reasons: (1) The indices were applied to groups of chiklrcn living under varying conditions; (2) clinical judgment of nutrition was made by different physicians; (3) there was no attempt to evaluate the objectivity and stability of clinical judgment nor the accuracy of the anthropometric measurements; (4) in one study (Clark et al., 1924) the indices were tested only on children who were clinically in good or excellent nutritional condition, in another (Mitchell, 1935), on children who had been referred by their teacher to the school physi- cian for special examination, in the third (Dublin and Gebhart, 1924), on boys and girls whom the physician judged to be either well or poorly nourished; and (5) different selection points were used for test- ing the same index. For example, Clark applied the Baldwin-Wood Tables to children in excellent or good nutritional condition and, using 6 percent as a limit for underweight, grouped the children by 5-percent differences *' See Appendix II: (7/) Mitchell, pp. 31&-317. *' See Appendix II: Ibid., pp. 318-319. 24 ASSESSI2^G THE PHYSICAL FITNESS OF CHILDREN in the indox. Dublin and Gohhart, on the other hiind, applied both 7 and 10 percent or more underweisrht as criteria for selection, classi- fied the children in only two groups, those who were identified as underweight and those who were not. Probably a more satisfactory metliod of testing and comparing such indices is to apply them to a group of children belonging to the same nationality,^- of a given sex and age, and living m one connnunity, who have been under observation for a suthcient period to permit evaluation of then- physical well-being and their socioeconomic status, and then to compare the eti'ectiveness of each of the indices in select- ing those children who, in terms of each of several carefully defined criteria, are lil-tely to be in need of medical attention or nutritional advice and assistance. The importance of such a study was pointed out by Jones in his monograph entitled "Physical Indices and Clmical Assessment of the Nutrition of School Children," in which he writes: Hundreds of physical indices of nutrition have been proposed by ^vTiters in many languages, but it is difficult to discover precisely what these indices will perform in practice. Though there is an enormous literature, there is no agree- ment among writers a* to which index is best. General claims concerning the value of an index are met more frequently than precise details of performance, and few attempts have been made to test different indices on the same large group of children. ^^ The observations made of the children included in the present study may be used to test the indices in this way and to evaluate other methods of assessment, particularly the clinical examination. How accurate is the physician's judgment of the child's general nutritional status? Is it stable? Is his assessment of the child's nutrition in agreement AN'ith the judgment of other physicians? To repeat: This investigation of the physical fitness of New Haven children mav be used to measure the relative efficiency of several methods of assessmg the physical fitness of 7-year-old school children. *> This term is used to distinguish broad, ethnic groups, <' See .\ppendis II: (51) Jones, pp. 2-3. Material and Methods The observations were made of 713 7-year-old white children ^ (365 boys and 348 girls) attending the kindergarten, first grade, or second grade of the pjublic or parochial schools of New Haven, Conn., from September 1934 through May 1936. Of these 713 children 661 (92.7 percent) attendc^d 44 of the 49 public grade schools in New Haven ;^ 52 (7.3 percent) of the boys and girls were enrolled in 4 of the 9 parochial schools.* Each of these children may be briefly described as follows: Race White. Sex .- Male or female. Nationality' Italian, American, or other.' Age (in completed years at first physical examination in 1934) Six years (only a child who was a single legitimate birth and was born during the period from July 1 to December 31, 1928, was eligible for inclusion). Residence Living with his family or in a foster home. School attendance Attending a public or parochial school in 1928, inclusive, Xew Haven. Period of observation A minimum period of 12 consecutive months during the school years 1934-35 and 1935- 36. (Most of the children were observed for either 19 or 20 months.)^ ' Included in this section are descriptions of the observations made of these New Haven children. Some of these data will form the basis for additional studies of child growth and development. ' Age is defined as age at last birthday. As is stated on p. 1, the children included in this study were 6 years old when the first medical examination was made. Periodic weighings were ma'le of some of these children before they were 6 years old, the age at the first weighing ranging from 5 years 8 months to 6 years 6 months. This report is based on observations made when the boys and girls were 7 years of age, amplified and interpreted in relation to the earlier findings. Only children who were single, legitimate births, who were bom during the period from July 1 to Decem- ber 31, 1928, inclusive, and who were living with their families or in foster homes were included in the study. ' 5 of the public schools were excluded because they served highly specialized groups of children. That is, one school had a very high percentage of .N'egroes in attendance; the second was conducted for children living in an orphanage; the third was used as a training school for teachers and was omitted for administrative reasons; the fourth had only a small enrollment of 6-year-oId children; the fifth was a special school for mental defectives. * These children were included in order to have the different nationality groujjs in N'ew Haven adequately represented. ' "Nationality" has boen defined arbitrarily according to the birthplace of the grandparents. For ex- ample, a child 3 of whose grandparents were born in Italy is classified as "Italian." It was npa-s.sar>- to expand this definition for "American" children to include boys and girls both of whose parents and 2 of whose grandparents were born in the United States. • A more detailed nationality distribution is given in App>endii I: Supplementary table I, p. 100. ' About 60 percent of the children were observed for 19 months and about 40 percent for 20 months. 25 26 ASSESSING THE PHYSICAL FITNESS OF CHILDREN The following: observations were made of the child during this 19- or 20-month period of observation : Physical examinations: He was given two annual physical examinations by one pediatrician. The first examination was made when the child was 6 years old; the second, 1 year later.* When the first examination was made, most of the child's clothing above the waist was removed; at the second examination, written permission had been obtained from the parents to remove all the clothing above the waist. Anthropometric measurements: He was measured at the same time the annual physical examinations were made by one observer trained in anthropometry.' Periodic weighings: The number of weighings varied: The minimum number was five; the maximum, nine; the average, eight. The child's age at the first weighing ranged from 5 years 8 months to 6 years 6 months; at the last weighing, from 7 years 2 months to 7 years 10 months. His weight was taken at the time each of the annual physical examinations was made; 6 months after the first examination; and at 4-month intervals i° during the school years 1934-35 and 1935-36." Information was also obtained concerning the following items: Socioeconomic status : An economic analyst ^^ visited his home during each of the 2 school years (1934-35 and 1935-36) and obtained information from his mother (or some other responsible adult member of the family if the mother was absent) concerning the child's general disease history, his dietary habits (particularly his consumption of milk and leafy vegetables'), and the economic status of the family. These visits were usually made within a period 2 to 3 weeks after the physical examinations. Schooling : Information concerning school absences — namely, the date, duration, and reason for each absence of 3 or more days' duration — was obtained from the school nurses' files. '^ School progress was also recorded as "passed" or "failed." These 713 white children form a selected group. They were 6 years old at the first physical examination, they were of single, legitimate 8 There was a minimum interval of 11 month.s iti days, and a maximum of 1 year 14 days, between the 2 physical examinations. 9 A few observations were made during the first year of the study by another trained observer who repeatedly checked her observations against those made by the anthropometrist. 10 These weighings were made for the most part during the last weeks of September 1934; January, May, and September, 1935; and January and May 1936. " The weights were secured for the most part within a 3-day interval, and were taken by the anthropom- etrist, the pediatrician, and the office clerk. A. few weighings were also made by the economic analysts who collected the socioeconomic data. n The socioeconomic data were collected from October through March or April of each year of the study. 6 economic analysts were employed during the first year, and 2 new agents collected the data during the second year of the study. ■ IS In the case of absences which occurred prior to the second home visit these data were supplemented by information obtained from the mother at this visit. In appropriate instances additional inquiries were made about absences which occurred subsequently. MATERIAL AND METHODS 27 birth, their nationality was known, thov were Hving in their own or foster homes, they were attending scliool in New Haven, and most of them were under observation for a 19- or 20-month period. Conversely, boys and girls were excluded or discharged from the study for any of the following reasons: Multiple birth; illegitimacy; incorrect age; unknown or incorrectly stated race or nationality; residence in an orphanage or other institution; attendance at a private school (other than parochial) or nonenrollment at school;" omission of certain anthropometric measurements or of either of the physical examinations; interval of more than 1 year and 14 days, or less than 11 months and 16 days between the two annual physical examinations; establishment of residence in another city; death during the period of observation; or age outside limits established for testing the indices.^* It is, of course, difficult to estimate the effect of excluding these children from the study, and no attempt has been made to do so. A Description of the Boys and Girls Included in the Study A description of the age and nationality groups represented, the kind of homes the children came from, and the general physical condi- tion of the 713 boys and girls included in the stud}^ may be helpful in interpreting the results. AGE The ages (in completed years and months) of the children when the second annual physical examinations were made are given in table 1. None of these 713 boys and girls were less than 7 years 0 months of age. Approximately 72 percent were less than 7 years 4 months, and none of them were more than 7 years 7 months of age. In other words, there is relativ^ely very little variation in the ages of the boys and girls included in the study. This somewhat unusual age distribution is probably the result of the following circumstances: Only children bom during the period from July 1 to December 31, 1928, inclusive, were eligible for inclu- sion; the juHiual physical examinations were made from October of one year through March of the next; and no ciiild was retained in the study who was less than 6 years of age at the time of his first medical examination or less than 7 years old at the second e.xaminatiou, which was made approximately 1 year later. '* There were 2 reasons for nonenrollment: (1) Children were not required to attend .school until they wore 7 years of aRc; (2) they were physically incapacitated and unable to attend .school. " At the bcpinning of the study, age was defined as age at nearest birthday. Later, it was changed to age at last birthday in order to permit testing of the Nutritional Status and ACH Indices on all the children included. 239848"— 40 3 28 ASSESSING THE PHYSICAL FITNESS OF CHILDREN Table 1. — Age of the boys and girls at the time of the second physical examination Both sexes Boys Qirls Age" Number Percent Number Percent Number Percent Total -- 713 100.0 365 100.0 348 100.0 7 vfiars 0 months 74 151 150 137 91 70 33 7 10.4 21.2 21.0 19.2 12.8 9.8 4.6 1.0 36 78 67 74 46 39 20 6 9.8 21.4 18.3 20.3 12.6 10.7 5.5 1.4 38 73 83 63 45 31 13 2 10.9 7 vpflrs 1 inonth - 21.0 7 years 2 months . 23.9 7 vpars !^ montlis - 18.1 7 Vfiars 4 months 12.9 7 Vfiars 5 months - 8.9 3.7 .6 1 Age is given in completed years and months. NATIONALITY The nationalities to which the children belong are shown in table 2.^® About 46 percent of the children were classified as "Italian" and approximately 18 percent as "American." The remaining 36 percent, representing various geographic groups, were classified as "Russians," "Polish," "Irish," and "All others." The fact that so large a number of Italian children were included in the study is of particular significance because, on the average, the Italian boys and girls were shorter and tended to weigh less than the other children.'^ Is this difl"erence in their body build likely to aft'ect the efficiency of the indices? Table 2. — Nationality of the boys and girls Both sexes Boys Girls Nationality ' Number Percent Number Percent Number Percent Total - 713 100.0 365 100.0 348 100.0 Italian 330 130 66 35 34 118 46.3 18.2 9.3 4.9 4.8 16.5 166 65 34 19 17 64 45.5 17.8 9.3 5.2 4.7 17.5 164 65 32 16 17 64 47.1 American 18.7 HussiaD - 9.2 Polish - 4.6 Irish - 4.9 All others 15.6 1 Classification was ba.sod on the birthplace of 3 of the child's grandparents. The cljkssifleation "Anu-rican" inchuii's not only children 3 of whose grandparents were born in the United States but also children whose parents and 2 of whose grandparents were born here. HOMES Location. Tlic location of these children's homes is shown on the map on p. 29. i« For the definition of "nationality" used in the study see page 25. 1' See tables 11 and 12, which give the heights an boys and girls were living in one-family dwellings; all the others, approximately 88 percent, lived in flats or apartments. (Table 3.) The fact that so few of the childi-en lived 30 ASSESSING THE PHYSICAL FITNESS OF CHILDREN in onc-faniily houses can be attributed in part to the relatively small number of such dwellings which exist in New Haven.** Table 3. — Type of dwelling occupied by the families of the boys and girls ' Type of dwelling • Both sexes Boys Girls Number Percent Nnmber Percent Nnmber Percent Total -- ---- 5 6S5 100.0 3358 100.0 '337 100.0 One-faniilv dwelling 84 611 12.1 87.9 37 321 10.3 89.7 47 290 13.9 Flat or anartuiont 86.1 > Data based on observations made when the children were 6 years of age. • A 1-family dwelling is a residence adapted to the use of only 1 family. It may be connected with a store or it mav be one of a series of houses with adjoining walls. A flat nr apartment is a 1-family unit in a building adai^ted' to 2 or more families. These definitions are adapted from Appendix II: iS6) Dreis, p. 9. 3 Type of dwelling was unknown for 7 boys and 11 girls. Presence of parents in the home. Most of these children (90.3 percent) were residing with both their parents. Less than 10 percent (9.0 percent) lived with only one parent; and very few boys and girls (0.7 percent) resided in foster homes. (Table 4.) Table 4. — Presence of parents in the homes of th e boys and girls Parents living in the home Both sexes Boys Girls Number Percent Number Percent Ntimber Percent Total- nz 100.0 365 1 100.0 348 100.0 Both parents - 644 64 5 90.3 9.0 .7 339 24 2 92.9 6.6 .S 305 40 3 87.6 One parent 11.5 Neither parent •.. _. .9 I Cases of children living in foster homes. Number of persons in lioiisehold. The number of persons in the household included not only the mem- bers of the child's unmediate family but also any relatives, friends, roomei-s, and servants Hving imder the same roof.*^ Nearly three- fifths (58.6 percent) of the children lived m households mcluding four. five, or six membei"s. A few (4.7 percent) lived in homes containing as many as 11 or more persons. (Table 5.) " The Southern New England Telephone Co. Market Survey of 1931 indicated that less than one-fifth oi all the occupied dwellings in New Haven were l-family residences. See Appendix II: (i6) Dreis, pp. 9, 11. " Boarders were excluded. MATERIAL AND :Mi;TIIOnS 31 Table 5. — Niimher of -persons in the households in which the hoys and girls lived^ Number of persons in household 2 Both sexes Boys Girls Number Percent Number Percent Number Percent Total . ... . 3 709 100.0 3 364 100.0 3345 100 0 2._ l.I. 3 65 140 149 127 77 58 34 23 16 11 3 3 0.4 9.2 19.7 21.0 17.9 10.9 8.2 4.8 3.2 2.3 1.6 .4 .4 0 36 76 74 55 42 29 20 13 10 5 3 1 0.0 9.9 20.9 20.3 15.1 11.5 8.0 5.5 3.6 2.7 1.4 .8 .3 3 29 64 75 72 35 29 14 10 6 6 0 2 0.9 3 8 4 4 18.6 5 21.7 6.. 20.9 7 10 1 8 8 4 9 4.1 10 . . .' 2 9 11 1 7 12... 1 7 13 ../.' . . 0 14... 6 ' Data based on observations made when the children were 6 years of age. 2 Number of persons in household included the child's immediate family, relatives, roomers, friends, and servants living under the same roof. 3 Number of persons in household was unknown for 1 boy and 3 girls. Number of persons per room. If crowding is defined arbitrarily as one and one-half or more per- sons per room,^° it may be seen from table 6 that approximately 32 percent of the children included in this study were living in homes where overcrowding was a serious problem, although in about 43 percent of the households there were from one to less than one and one-half persons per room, and about 25 percent of the children were living in homes with less than one person per room. Table 6. — Number of persons per room in the households in which the boys and girls lived ' Persons per room > Both sexes Boys Girls Number Percent Number Percent Number Percent Total 3 707 100.0 3 363 100.0 3 344 100.0 Less than 3^ .. 2 174 302 145 59 22 2 1 0.3 24.6 42.7 20.5 8.4 3.1 .3 .1 0 94 147 81 0.0 25.9 40.5 22.3 2 0.6 ' 2. less than 1 . 80 23.2 1, less than IJ/^ 1.55 45. 1 IJ 2, less than 2. 64 18. 6 2, less than 24 30 8.3 29 1 8.4 21.2, less than 3 10 0 1 2.7 .0 .3 12 1 3.5 3. less than 34 -^ j.. . 2 ' .6 31^. less than 4 . . 0 .0 1 Data based on observations made when the children were 6 years of age. 2 The number of persons in the hou.sehold included the child's immediate family, relatives, roomers, friends, and servants living under the same roof. The number of rooms excluded bathroom.s and hall- ways. 3 Number of i)ersons per room was unknown for 2 boys and 4 girls. 20 It is recognized that any arbitrary standard of crowding has limitations. In this connection Rollo H. Britten, of the National Institute of Health, has pointed out that "a reasonable index should depcn 3 of the child's grandparents or 2 of his grandparents and both his parents were born iu the United States. 3 These children did not meet the definitions outlined in footnotes 1 and 2. Weight. Tlio weights of th(> children are given in Itible 12.-^ Here, as in iHMglit, tliei'c iip|)(>Mr to he diirerences between (he cliihh-en classified « For the fr<'(itn>ucy 3 of the child's pranditarcut.s were horn in Italy. ' 3 of the child's grandparents or 2 of his grandparents and both his parents were born in the United States. ' These children did not meet the definitions outlined in footnotes 1 and 2. Such nationality diflVTences are important because, as has been mentioned (sec p. 28), they may affect the efficiency of the indices in identifying boys and girls in need of medical attention or nutritional advice or assistance. The indices depend on estimates of body build (based on anthropometric measurements) for judging a child's physical fitness (insofar as physical fitness is related to abnormalities in certain measures such as weight or arm girth) when the exj^ected or average measure for children of the same sex, age, and Iniild is used as a stand- ard. wSuch a standard may vary with the nationality of the children from whose measurements it was derived and may. therefore, be most efficient in judging the fitness of children of the same nationality as the boys and girls whose measurements wcic used in developing the index or standard. DIETS As information concerning the dietary Imbits of tiicse cliiMicn is not very satisfactory, only the data on inili\ eonsuniplion iiave been analyzed.^* (Table 13.) Although 5G.5 percent of the boys and girls 2< The estimates ot milk consumption are obviously inaccurate and in addition are based on a cup which is smaller than the standard measure. 36 ASSESSING THE PHYSICAL FITNESS OF CHILDREN drank an "adequate" amount (2, 3, or 4 cups per day) of milk at 7 years of age,^^ only 28.3 percent consumed an "optimum" amount (5 cups or more). On the other hand, a considerable proportion of these children (13.9 percent at the age of 7) had an inadequate amount (less than 2 cups per day) of milk in their diet, and a very small num- ber of the boys and girls (1.3 percent at the age of 7) did not drink any milk. Table 13. — Adequacy of the amount of milk consumed daily by the boys and girls Adequacy of milk consumed per day No milk: At 7 years At both 6 and 7 years Inadequate: 2 At 7 years At both 6 and 7 years Adequate: ^ At 7 years.- At both 6 and 7 years Optimum: * At 7 years At both 6 and 7 years Both sexes Number 9 0 99 39 402 201 SO Percent > 1.3 .0 13.9 o.i 56.5 i^.S 28.3 7.0 Boys Number 3 0 55 S5 200 H7 106 Percent ' 0.8 .0 15.1 6.9 55.0 40.4 29.1 9./ Girls Number 6 0 44 H 202 165 95 17 Percent ' 1.7 .0 12.7 4.1 58.2 44.7 21. A 4-9 > The percentages in each case were based on the total number of children for whom information at both 6 and 7 years of age was available: For both sexes, 711; for boys, 364; and for girls, 347. (The amount of milk consumed daily was unknown for 1 boy and 1 girl at 7 years of age.) 2 "Inadequate" was defined as less than 2 cups per day. 3 "Adequate" was defined as 2 up to but not including 5 cups per day. The absolute standard of 3 cups a day has not been adopted because the estimates of milk consumption are subject to error. < "Optimum" was defined as 5 or more cups per day. HEALTH Pediatrician^ s assessment of general nutritional status. As has been stated, all the children were examined both years of the study by one pediatrician, who judged their nutritional status to be "excellent," "good," "borderline," "poor," or "very poor." These estimates were based on careful physical examinations which were made according to detailed wi'itten instructions.^^ On the basis of these examinations the percentage of children in each nutritional class was as follows: (1) More than half the boys and girls were judged to be in a border- line condition at 7 years of age. Some of them were probably fairly well nourished; others were in a nutritional condition that ])ordered on poor. Nevertheless, all these children constitute a health problem, 2« The standard of 3 cups per day as adequate and 4 cups as optimum has not been applied. (See Appendix II: UOO) Sherman and Hawley.) Instead, 2 up to but not including 5 cup.* a day (an average of SJ.^ cups) has been considered adequate; less than 2 cups, inadequate; and 5 cups or more, an optimum amount. This grouping has been made in order primarily to identify 2 of the 3 groups of children: (1) Those who had a definitely unsatisfactory amount of milk in their diets as compared with (2) those who had an optimum amount. '« A detailed evaluation of both the objectivity and the stability of the pediatrician's judgment occurs on pp. 82-87. MATERIAL AND METHODS 37 as they are difficult to identify and their condition may grow progi'es- sively worse if they do not receive proper care and treatment. (2) About 34 percent of the boys and girls were judged to be in good or excellent nutritional condition (30.2 percent good and 3.4 percent excellent) at 7 years of age. (3) Necessarily, if about 57 percent of the children were rated as in borderline and about 34 percent in good or excellent nutritional condition, only a small percentage (9.5 percent) could have been judged by the physician to be poorly or very poorly nourished at 7 years. All these children, 28 out of 365 boys and 40 out of 348 giiis, were classified as poorly nourished; none of them as very poorly nom'ished. It is also interesting to note (table 14) that, according to this pediatrician's judgment, about 5 percent of the children were poorly nourished during both years of the study, compared with about 9 percent at 7 years of age. Table 14. — Pediatrician' s assessment of nutritional status of the boys and girls Nutritional status Excellent: At 7 years At both 0 and 7 years. Good: At 7 years At both 6 and 7 years. Bonlcrlinc: At 7 years At both 0 and 7 years. Poor: At 7 years. „ At "^ofh ") and 7 years. Very poor: At 7 years At both 0 and 7 years. Both sexes Number Percent ' 24 H 215 101 406 S9S C8 S9 0 0 3.4 g.O 30.2 11 g 56.9 41.1 0.5 fi.i .0 .0 Boys Number Percent • 11 B 120 47 206 ISl 28 16 0 0 Girls Number 3.0 1.4 32.9 li.9 56.4 41-4 7.7 4.4 .0 .0 13 9 95 S4 200 14S 40 ts 0 0 Percent > 3.7 i.6 71. Z 16.6 57.5 W.S 11.5 6.6 .0 .0 I The percentages in eacii casi were ba.sed on the tolal n'iraber of rhildren for whom information at both 6 and 7 years of age was availah,^: For both sexes. 713: fo' hoys, 365; and for txirls, 348. Diafinoses made at annual pliysicfil exa in i nations.'^' The diagnoses made at the examinations givn when the chikh'cn were 6 and again when thoy were 7 years of age are showTi in table 15. At 7 years of age 53 percent of the l)oys and girls were found to be suffering from various ailments, mostly ol the respiratory tract. Under nonrespiratory infections only one case of rheumatic heart disease was included, and under the heading "OtluM- positive diag- noses," thi"ee congenital heart conditions. With few exceptions, the rest of the diagnoses were of minor importance. This is not sur- prising, for all the children wvvc well enough to attend school and to be under observation for approximately a 2-year period. »" If more than 1 diagnosis was made at a given examination, that which the pediatrician considered most important clinically was recorded. 38 ASSESSING THE PHYSICAL FITNESS OF CHILDREN Table 15. — Diagnoses made at the annual physical examinations of the boys and girls Both sexes Boys Girls Diagnosis At 6 years At 7 years At 6 years At 7 years At 6 years At 7 years Num- ber Per- cent Num- ber Per- cent Num- ber Per- cent Num- ber Per- cent Num- ber Per- cent Num- ber Per- cent Total 713 100.0 713 100.0 365 100.0 365 100.0 348 100.0 348 100.0 No disease 265 448 37.2 62.8 335 378 47.0 53.0 119 246 32.6 67.4 152 213 41.6 58.4 146 202 42.0 58.0 183 165 52.6 Disease • 47.4 Infectious disease 444 62.3 374 52.5 243 66.6 211 57.8 201 57.7 163 46.8 Bespiratory Nonrespiratory. , . Other 2 - 273 171 4 38.3 24.0 .6 232 142 4 32.6 19.9 .5 163 80 3 44.7 21.9 .8 136 75 2 37.3 20.5 .6 110 91 1 31.6 26.1 .3 96 67 2 27.6 19.2 .6 1 If more than 1 diagnosis was made at a given examination, that which the pediatrician considered the most important clinically was recorded. 2 "Other" included asthma and congenital heart conditions. Two other items may be useful in indicating the general physical condition of the boys and girls included in this study: (1) School absences; and (2) the physician's judgment of the child's need for medical or dental care. Number and duration of reportable school absences. Only absences of 3 or more school days' duration are reportable in New Haven. The school nurse's files show the number of such absences, the total number of school days which they represent, and the reasons for the absences. Only absences due to illness have been recorded, and the material has been analyzed in terms of the associa- tion between the duration of reportable absenes cand the number of such absences. (Table 16.) Table 16. — Association between the nnmher of reportable absences during the school year and the duration of these absences ' Duration of reportable absences Number of reportable absences » 0 3-4 5-9 10-14 15-19 20-29 30 or more Total 0 1 186 0 0 0 0 186 0 102 0 0 0 102 0 107 49 1 0 157 0 20 54 22 3 99 0 5 17 20 4 46 0 7 22 19 12 60 0 8 8 7 19 42 186 249 2 -. 1,50 3 69 4 or more 38 Total 3 692 1 The data on school absences have not been presented separately for boys and girls, because analysis failed to reveal any sex difference in either the number or the duration of reportable absences. 2 .\ reportable absence is an absence of 3 or more school days. 3 Information on school absences was lacking for 8 boys and 13 girls. During 1935-36, when the children were 7 years of age, approxi- mately 27 percent of the boys and girls had no reportable school MATERIAL AND METHODS 39 absences. 2^ It should be remembered, however, that these children may have been absent any number of dscys during the year if each absence were of less than 3 school days' duration. It is unfortunate that such boys and girls have not been differentiated from the children who had no absences, for a sicldy child who is able to attend school except for absences of a day or two should be distinguished from a healthy boy or girl who never misses school. On the other hand, none of these children had any protracted illnesses during the school year, although it may well be that some of the boys and girls frequently suffered from minor ailments which undermined their health but did not materially affect then- attendance at school. Most of the other boys and girls (36.0 percent of all the children) had only one reportable absence. This absence usually lasted from 3 to 9 school days (41.0 percent, 3 or 4 days, and 43.0 percent, 5 to 9 days); there were a few boys and girls (10.0 percent) absent from 10 to 19 days, and an even smaller number (6.0 percent) absent 20 days or more. About 22 percent of all the children were absent twice for periods of 3 school days or more. About one-third of these boys and girls were out less than 10 days; a somewhat larger proportion (36.0 percent), 10 to 14 days; and the others (31.3 percent), 15 days or longer. Finally, a considerable number of the children (more than 15 per- cent of the total) were reported as having had three or more absences at 7 years. These boys and girls were usually absent for long periods. Twenty-nine percent of them were not at school for 20 to 29 days, and nearly one-fourth of these children had reportable absences total- ing 30 school days or more.-^ Medical and dental care. When the pediatrician examined the children each year she indi- cated whether they needed medical care or dental care oi- l)Oth (table 17). Analysis of her findings shows that nearly 60 percent of the boys and girls were in need of both medical and dental care at 7 years of age, and that slightly more than 40 p(>reont needed such care at botli 6 and 7 years. In other words, nearly half the children were in need of medical and dental care during both years of the study. Even more significant, perhaps, is the fact that only about 2 percent of tlie boys and girls needed neitlici- medical nor dental care at 7 years of age and that only 1 percent of lli(> childic?! needed neither type of care at 6 or at 7 years of age. ^^ Tlio data on school absences have not been presented separately for boys and Rirls, because analysis failed to reveal any sex di(Terena>s in either the number or the duration of reiiortable absence.s. 2* A larger percentatje of tlie children had reportable absences (ntaliiitr 10 school days or more at 6 than at 7 years, but there appears to be little association between the duration of n-porlablo absenii'S at f> and at 7 years except for the boys and cirls who had absences totaling 15 days or more. Consequently, flndinps at the earlier age have been omitted from the discussion. 40 ASSESSING THE PHYSICAL FITNESS OF CHILDREN In interpreting these findings it should be remembered, however, that need for medical care refers to observation as well as treatment; undoubtedly many of the children needed observation only, because of a reported history of colds, perhaps; or such conditions as dull ear drums; mouth breathing; or moderately enlarged tonsils. Table \7 . -Recommendation of Die pediatrician on the need of the boys and girls for medical and dental care Type of care needed Medical and dental care: At 7 years At both 6 and 7 years. Medical care only: At 7 years At both 6 and 7 years Dental care only: At 7 years At both 6 and 7 years No care: At 7 years At both 6 and 7 years Both sexes Number 422 !196 271 m 13 Percent ' 59.2 41.5 1.0 ■ 4 38.0 17.2 1.8 1.1 Boys Number 215 150 143 66 Percent ' 58.9 41.1 .8 .5 39.2 18.1 1.1 Girls Number 207 146 4 ; 128 57 9 7 Percent > 59.5 4S.0 1.1 .3 36.8 16.4 2.6 i.O 'The percentages in each case were based on the total number of children for whom information at both 6 and 7 years of age was available: For both sexes, 713; for boys, 365; and for girls, 348. SUMMARY Most of the boys and girls were between 7 and 7}^ years of age when the indices were tested. Nearly half of them were Italian, and nearly 20 percent were American children; the others represented various smaller nationality groups. The different sections of New Haven were all fairly well repre- sented. Most of the children lived in flats or apartments with both their parents; and there were generally four, five, or six persons living in their homes. Considerable overcrowding existed; about 32 per- cent of the boys and girls lived in households where there were one and one-half or more persons per room and in about 43 percent of the homes at least two people were occupying the same bedroom with the child who was included in this study. Some of these boys and girls were in extremely poor economic circumstances; about 24 per- cent of their families received assistance (from public, private, or both public and private sources) during the year prior to each of the home visits. Even more significant is the fact that 17 percent of the families were dependent on such assistance for the principal part of their income during the second year of the study. There were marked differences in the body build of these boys and girls. For exam pi (\ the Italian children were shorter and tended to weigh less, on the average, than the American children. The diets, judged in terms of the number of cups of milk consumed per day, were often inadequate. At 7 years of age, only about 28 per- MATERIAL AND METHODS 41 cent of the boys and girls drnnk nn optimum amount (five or more cups) ; about 15 percent of the chikhen had less than two cups per day. If the phj^sician's judgment is used as a criterion of general nutri- tional status, about 34 percent of the boys and girls were in good or excellent nutritional condition at the time the indices were appUed (at 7 years); almost 57 percent were in a borderline condition, and approximately 9 percent were judged to be poorly nourished. Other evidence which gives some indication of the health of these children includes: (1) The specific clinical findings at the physical examinations; (2) absence from school; and (3) the physician's judg- ment of the boys' and girls' need for medical and dental care. (1) At the time the second physical examinations were made, the physician found more than half the boys and girls suffering from various ailments, mostly respiratory-tract infections. (2) Of the boys and girls with reportable school absences (i. e., absences of 3 school days or more) at 7 years of age, 36.0 percent were absent only once, 21.7 percent, twice, and 15.4 percent, three or more times. A large proportion (53.3 percent) of those who had been out three or more times were out of school for as much as 20 school days or longer. (3) According to the physician's judgment, nearly 60 percent of the children were in need of both medical and dental care at 7 years, and slightly more than 40 percent at both 6 and 7 years of age. A child in need of medical attention was almost always in need of dental care, although the reverse was not generally true. All these findings — medical, dental, and socioeconomic — indicate that without doubt there must be boys and girls included in this study who were likely to be undernourished and to be in need of medical care or nutritional .idvice and assistance. The Ohservalions Made of These Children TECHNIQUES EMPLOYED Every attempt was made to have the observations of these boys and girls as accurate as possible. In order to accomplisli this purpose, uniform methods of collecting the material and editmg the schedules were used. No attempt will be made to describe each of the observations.^" On the other hand, the most important items — namely, clinical judg- ment of the child's nutrition and anthroponiotrir measuroinonfs will be described in some detail. 'I- Instructions for making the physical examinations, for takinR the anthropometric measurements, and for obtaininK socioeconomic data, together with the schedules for n'cordinp these observations may be obtained upon request from the Children's Bureau, U. S. Department of Labor. 42 ASSESSING THE PHYSICAL FITNESS OK CHILDREN Clinical assessment. For many years clinical judgment has been the accepted method of assessing a child's physical fitness. With increasing knowledge, improvements have been made in the procedures used in the examina- tion and in the method of arriving at a final judgment of the child's general nutritional status. But the physical examination which forms a part of the usual school health program does not always include these techniques, because they may require specially trained personnel or new equipment or both, and because some of them consume con- siderable time in the examining room or in the laboratory. In other words, such tests are costly, and most school budgets have not ex- panded sufficiently to support so expensive a program. These more elaborate procedures ^^ were excluded, therefore, from the present study in order to have the physical examinations comparable to the examinations included in the ordinary school health program. There were two other reasons which made it advisable to exclude these specialized tests. First, they present many administrative and technical problems. Thus specific permission must be obtained from parents or guardians of the children before some of the procedures may be applied. Second, some of the tests have not been standardized sufficiently to permit accurate interpretation. But even if these considerations could have been eliminated, the first reason still remained; namely, the advantage of having the physi- cal examinations of the boys and girls in this study comparable to examinations which it is administratively practical to employ in a school health program today. If the physical examinations are made in this way, the four indices included in the study can then be com- pared with the type of examination which they would ordinarily be used to supplement or replace. Furthermore, a study of the objec- tivity of clinical judgment of general nutritional status based on this type of examination might furnish some clues as to how to improve the physical examination without the addition of elaborate and ex- pensive tests. In short, it was considered more important to use carefully a method which the average school physician can and must employ than to set up an elaborate and costly procedure which it is not at present practical to adopt in examining large groups of school children. In order to make the examination as objective and accurate as possible, detailed written instructions were prepared for the pediatri- cian's use concerning the number of items to be included in the exami- nation and the method of evaluatmg each item. In addition, a 5- point scale was devised for judging the child's general initritional status (see chart entitled "Grading of General Nutritional Status" p. 43). 31 Kxccpt for li('ni()(;lohiii ami rod-blood-cpll liclcriiiiiial ions iiiado on the first day of the clicck-up exaiiiiiia- tions on IXi cliildrcii (70 boys and (V.i girls) who wore included in tlio study. MATERIAL AND METHODS 43 ^4* q5 o o t/3 « &, ^ cw (h* S ^ >* OS.:; > 1^^ lO CMti,> OJ ^' q; CJ 6 o o 3 .a to . 3 O 2 0.5-C o^ S o3 X-C oti.2 8|5 «8b cS X O ^ G e3 4^ G :: 055 (u G >i 0 3 G ■3.5x; OfeH Ofe> PhSH Ph;i^> Pnt^H p^'^> f^^H rS' OJ o . +3 *- G "3 3 . 0 t? O O tH O X 11 O ■/: oS.S o.iix: O X - "o =:s ^ G O .^ ^ •<-■ X r\ CJ _o p "o " ^• o^ 3 I-.' cj X 3 r^ 2 0 OW> OfeH Ot^> Ofoc» Ofefe PhWCh PnfiHOQ d T3 . O — o . •/; o3 O ;-.' S t"«5 u ;3 **-t O O U2 O " G (N « C c3 O >, ID O X^ 0.5: 3 OWH Ofefe OfeW aj 4J . S 9) C r— 1 "o K S -f^' ..•2^ 11 to 3 u-a 13 1 8 X ^ -u +^ O G C W ^ C_) O G +J O G > - Gt; 0 x 0 w O QJ QJ TJ "q; "aJ o G oj "H g'o " G § X ^ 2 ' o « o O o-rs o o.h o o o O - o 2 c -S O X X cB^cl O X c3 O X X o.S: X 2-- d OWW OWfe caw OtaW OfaM 2;<9848° — 40- 44 ASSESSING THE PHYSICAL FITNESS OF CHILDREN This scale was based on (1) color of mucous membranes, ^•^ (2) quality of muscle, ^^ and (3) amount of fat,^* and was used as a guide in evaluating nutrition. Other items, inchiding posture, condition of the skin and hair, and general development, were given consideration in arriving at a final clinical judgment of the child's general nutritional status.^^ Anthropometric measurements. The anthropometric measurements necessary for determining the Baldwin-Wood, Nutritional Status, ACH, and Pryor Indices were made according to the methods recommended by the investigators ^^ with the following exceptions: Weight. — After his shoes, sweater, and coat were removed, the child was weighed on a balance scale located in the school and his weight was recorded to the nearest quarter of a pound. This is the pro- cedure used for the Baldwin-Wood, the ACH, and the Nutritional Status Indices. In calculating the Pryor Index, there is a correction to allow for clothing if weight is not taken in the nude. This adjust- ment has been made for each boy and girl included in the present study. Iliac crests. — When the children were 7 years old the crest measure- ments were taken next to the skin, according to the prescribed tech- niques, with a sliding wooden caliper. However, during the first 3 months of the study (October through December 1934), the bi-iliac diameter was taken over the clothing with an obstetric metal caliper. This difference in technique will not affect the indices, because the incorrect technique was used only in the first 3 of the 5 months during which the examinations of the 6-year-old children were made, whereas the indices were derived from measurements made when these children were 7 years of age. The instruments used in taking the measurements were checked frequently; that is, the steel tape equipped with a Gulick spring handle, used in making the Franzen and ACH measurements, and the subcutaneous-tissue calipers for the Franzen Indices, were compared at weekly intervals with standardized instruments. About every 2 months they were returned to the factory for replacement of springs or for calibration. In addition, the scales located in each of the schools were balanced at frequent intervals. 32 Color of mucous monibranes was assessed as "good" or "definitoly iwle." 33 Cliiiifiil jinisiiiiciit of iruisclc which was graded as "excellent," "firm," or "flabby," was based on a combinalion of muscle pull and nuiscle tone. '< Clinical judgment of fat which was graded as "excessive," "excellent," ".satisfactory," "fair," "thin," or "very thin," was based on an average of the amount of arm and abdomen fat. " Diagno.ses based on i)hysieal examinations were not used in arriving at a judgment of the child's general nutritional status exempt insofar as they affected th« elinical items entering into the assessment. Neverthe- less, such diagnoses may have influcnaul the pediatrician unconsciously, although every elTort was made to base the assessment wholly on the items specified in the instructions. 3« For a detailed description of the methods of f iiking th(- anthropometric measurements ncci>ssary for com- puting each index, sec the following references given in Api)endix II: Baldwin-Wood Tables— (5) Baldwin, p. 1; ACII Index— (.W) Franzen and Palmer, pp. 9-11; Nulrilional Status Indices— (75) Nutritional Status Indices, pp. 7-12; and Pryor Width-Weight Tables— (88) Pryor and Stolz, p. 3, and (86') Pryor, p. 8. MATERIAL Als^t) METHODS 45 ACCURACY OF THE TECHJNIQLES EMPLOYED Clinical assessment. It will be remembered that all the physical examinations were made during both years of the study by one pediatrician, who used detailed written instructions as an aid in arriving at her assessment of the chikhcn. Since her judgment of general nutritional status has been used both as a method of assessment and as a criterion for evaluating the four indices of body build, it is important to determine both its objectivity and its stability. In order to study these two problems, check-up examinations were made of some of the children after the physical examinations were completed in March 1936. They con- sisted of two parts: (1) An initial and repeat examination of some of the children (51 boys and 52 girls) by the same pediatrician who made all the physical examinations during both years of the stud}^, in order to check the stability of her judgment; and (2) examinations of these same children and of 105 others (56 boys and 49 girls) by the same physician and by two additional pediatricians, in order to de- termine the extent of agreement in judgment among the three physi- cians. The findings arc presented in a later section of tliis report. (See pp. 82-87.) Anthropometric measurements. At the same time that the check-up examinations were made, the accuracy of the anthropometric measures was studied in this manner: " One hundred children (50 boys and 50 girls) were measured twice by anthropometrist D, who made all the measurements during both years of the study, and between D's initial and repeat observations were measured by another anthropometrist, E. In order to study the variability of the anthropometric data, the initial measurements of each anthropometric characteristic made by anthropometrist D have been compared with the measurements of the other observer, E. This comparison has been made in terms of the means and standard deviations of the two distributions; that is, if D and E measured the children with an equal degree of accuracy, the means as well as the standard deviations of the dislributions should coincide. Conversely, if the two observers' measurements were not made with the same precision, the means or the standard deviations or both will not be identical. Usually, however, the means coincide, and under these circumstances the anthrojiometrist whose measure- ments have the smaller standard deviation is the more accurate of the two. 3" For pvidenoe conocminj; the variability and consistency of anthrnpomolric mciusiin'monts sec Appendix H: (/«) Boyd, (45) Hfjinian and Uatt, (56) Lincoln, (66) Marshall, and (67) Meredith. 46 ASSESSING THE PHYSICAL FITNESS OF CHILDREN To illustrate this method the reader is reminded that D and E each measured the bitrochanteric width of 50 7-year-old boys. Analy- sis shows that the average or mean bitrochanteric width for D's observations is 21.9 cm. (table 18); and their dispersion, measured in terms of the standard deviation, is 1.36 cm. If this value is added to and subtracted from the mean, the limits obtained, 20.5 and 23.3 cm., define an interval that includes slightly more than two-thirds of D's measurements. Table 18. — Means and standard deviations of anthropometric measurements made by observers D and E on 50 boys and SO girls Anthropometric measurement Boys Mean Arm girth ' (cm.) Bi-iliac crests (width) (cm.). Bitroehanteric width (cm.)- Chest breadth " (cm.) Chest depth i (cm.) Height (in.) Subcutaneous tissue ' Weight (lb.) Observer — D 35.4 20.0 21.9 38.9 28.3 47.3 23.5 52.9 35.1 19.6 21.8 39.3 28.8 47.3 24.3 53.0 Standard deviation Observer — D 3.45 1.20 1.36 1.86 1.63 2.19 3.39 8.73 E 3.37 1.22 1.42 2.03 1.94 2.18 3.32 8.68 Girls Mean Observer- D 35.4 19.9 22.1 38.4 27.8 47.6 26.0 52.9 E 35.1 19.8 22.1 38.9 28.5 47.7 27.7 53.0 Standard deviation Observer- D 3.61 1.02 1.20 2.21 1.91 2.00 4.41 8.12 E 3.42 1.12 1.22 2.28 1.91 1.98 4.84 8.04 1 This anthropometric characteristic is the sum of 2 measurements. See Appendix II: (75) Nutritional Status Indices, pp. 7-12, for a description of this measure. A similar analysis made of E's observations shows that the mean is 21.8 cm., and the standard deviation, 1.42 cm., compared with 21.9 and 1.36 cm. for D's measurements. (Table 18.) Now, if this stand- ard deviation, 1.42 cm., is added to and subtracted from the mean, 21.8 cm., the limits obtained, 20.4 and 23.2 cm., are about the same as those for D's observations (20.5 and 23.3). These fundings indi- cate that there is excellent agreement between the two anthropome- trists' measurements of the bitrochanteric width of 7-year-old boys. The data for the 50 girls included in these anthropometric examina- tions also show close agreement between the two observers. (Table 18.) Similar analyses of the other anthropometric characteristics of these boys and girls are shown in table 18. It may be seen upon examination of this table that the largest errors for the boys were made in measuring chest depth, chest breadth, and arm girth, in the order named, and for girls, in subcutaneous tissue, arm girth, and width of the iliac crests. As a whole, however, the observations of D and E were in very close agreement. ^^ 58 For any given anthropometric characteristic the error of the measurement has been assumed to be directly proportional to the ditToreiice between the iiuniber of scale units within the interval defineendi\ \\: (r.i^ Xufriiional Status Indices, pp. 7-12, for a description of this measure. '« For a description of the method of estimating the standard deviation of the error of measurement, sec Appendix \l: (70) Palmer, pp. 227-228. 48 ASSESSING THE PHYSICAL FITNESS OF CHILDREN This analysis indicates that D's measurements are unusually stable. This finding, together with the fact that her measurements agree well with those of the other observer, E, furnishes convincing evidence that D is an extremely careful and accurate worker and that her measure- ments are a satisfactory basis for deriving the indices included in this study. Socioeconomic and related data. The socioeconomic data were checked and verified by different methods, depending on the type of material and the sources available. For example, statements about illnesses were checked against hospital and clinic reports, the files of the Visiting Nurses' Association, and the records of the Divisions of Tuberculosis and Venereal Diseases of the New Haven Department of Health. Data on income and assistance could be verified only for families who reported that they had received assistance during the year preceding each home visit or who were known to public or private agencies in New Haven giving direct or work relief. During the course of the study all the families with an income under $1,000, those reporting relief assistance, families with members suffering from severe illness, and those whose income estimates appeared to be grossly inaccurate were cleared through the Social Service Exchange of New Haven. This means that the source and amount of income of these families, the number of persons living on the income, and the type and amount of assistance were checked agninst the records of the social agencies, both public and private, admmistering any form of assistance in New Haven. School absences and progress were veiified from school records. The office clerk also checked relevant data against one another and against the files of the health and welfare agencies of the city. For example, wherever possible the birth date of the child as well as his legitimacy and the age of his parents were compared with the records of the Bureau of Vital Statistics of the City Health Depart- ment; other facts about the parents, with the files of the school nurses and agencies granting assistance. Data obtained at the first home visit which did not tally with information derived from other sources were checked by the eco- nomic analysts at the second homo visit. In general, the socioeconomic data are fairly reliable and accurate. Such inaccuracies as occur are due for the most part to the failure of the mother to remember the child's history or to the fact that, as the child had previously lived in another city, records for checldng were not available. Some inaccuracies have also resulted from the em- ployment of several economic analysts for collecting the data, Inso- MATERIAL AND METHODS 49 far as possible, however, the observations were checked against one another and against the records of pubHc and private agencies in New Haven. Suiiiinary of Material and Methods In the preceding pages an attempt has been made to outhne the purpose of the study and the method of collecting the observations as well as the procedures used in checking the material and editing the schedules. Before proceeding to an analysis of these observations it may be well to review the plan and objectives of the investigation: The plan was to observe a group of 713 children living in New Haven, Conn., from September 1934 through May 1936, who were 7 years of age at the time of the second annual physical examination and were attending the public or parochial schools of the city. The purpose was to study the efficiency of each of several methods of assessing physical fitness (clinical judgment and foin- indices of body build) in identifying children wlio at 7 years of age were likely to be in poor physical condition. The evaluation of these methods of assessing the child's well-being has been made in terms of the findings when the children were 7 years old for two reasons. First, the indices of body build have been computed from measurements of the boys and girls taken at the age of 7 years, as the Nutritional Status Indices and the ACH Index apply to 7-year-old children and not to boys and girls 6 years of age.^^ Second, if the various methods of assessment are tested on the boys and gij-ls at 7 years of age, a whole series of clinical, anthro- pometric, and socioeconomic data, based on observations made over a period of 20 months, are available for evaluating the physicnl fitness of the children. "> In other words, these 2 methods are prohably most elficient in testing children of the same age as the boys and girls from whose measurements tlie indices have been derived. Results Indices of Body Build THE PROBLEIVI OF TESTING THE INDICES Two serious difficulties stand in the way of any attempt to test the four indices of body build included in this study. The first, discussed in detail in the early part of the report, is based on the fact that none of these indices attempts to identify every child who is in poor physical condition. In fact, they measure the child's physical fitness only insofar as it is related to such characteristics as his weight, subcuta- neous tissue, or arm girth. Consequently, if the indices are to be given a fair but rigorous test, it is necessary to apply them to a large group of children. The success or failure of the indices can then be measured by comparing the children they identify with the boys and girls who are likel,y to be in very poor physical condition. The selection of such a group constitutes the second difficulty in studying these indices. It is apparent that some standard is needed for identifying children on whom the indices are to be tested, but unfortunately, as has been indicated, such a standai-d of physical fitness is lacking. In fact, it is this very need for a reliable measure which meets both the statistical requirements of objectivity and reliability, and the practical requirements of unprohibitive cost, ease of application, and expediency that the indices now being studied were designetl to meet. In the true sense, then, the solution to the problem of the efficiency of these indices is indeterminate, since no entirely satisfactory criterion of reference is at hand. CRITERIA FOR TESTING THE INDICES Nevertheless, many of the factors which go to make up this com- plex state are known and can be measured. They can be used for deriving approximations to the true standard of physical fitness which may serve as criteria for testing the indices. Such a proced- m'e— namely, setting up several arlntrary but well-defined criteria for identifying ehikh-en who are likely to be physically unfit — has been adopted in this investigation. Ohserratiotis used in deriving criteria. Before defining the specific criteria which have been used in this study, it may be well to restate the material and methods and to (h^seribe some of tlie estimates of growth and development which 50 RESULTS 51 have been used in deriving these criteria. It will be remembered (1) that during a 19- or 20-month period of observation beginning Sep- tember 1934 one pediatrician examined 713 school children at 6 and again at 7 years of age; (2) that the physician described each child's general nutritional status as "excellent," "good," borderline," "poor," or "very poor"; (3) that at each physical examination the pediatrician indicated whether the child was in need of medical care, dental care, or both; (4) that an anthropometrist took eight measure- ments when the two annual physical examinations were made; (5) that each child was weighed at frequent intervals during the 19- or 20-month period of observation. The anthropometric measurements have been used to derive the following growth estimates: 1. The absolute increase or decrease in each anthropometric characteristif. For example, if a boy's arm girth was 33.2 cm.' at the ago of 6 and 34.3 cm. at the age of 7, his arm girth increased approxiinatolv 1.1 cm. fluring the 12-mnnth period. 2. The relative percentage increase or decrease in each of these anthropometric characteristics. Thus, the arm girth of this same child increased about 3.3 percent in a year (1.1 cm./33.2 cm.).- 3. A more refined estimate of gain in weiglit. In order to determine each child's average percentage gain in weight per month,^ an equation has been fitted to his successive weigln'ngs made at frequent intervals during the 19- or 20-month jjcriod of observation.* This equation measures the child's relative gain in weight much more accurately than an estimate derived from the two weighings made at the annual physical examinations because it is based oji a larger number of meas- urements (five to nine) made at more frequent intervals. Some of these data, both the clinical observations and the growth estimates, have been used to derive provisional criteria of physical fitness which will make it possible to appraise the eU'ectiveness of the foiH' iiulires of })ody build used in this study nam(>ly, the Baldwin- Wood Ileight-Weight-Age Tables, the A(ll Judex, the Nutritional Status Indices, and the Pryor Width-Weight Tables. Description of the criteria. Five empirical criteria of y)hysical fitness havc^ been establislied. Criterion I is based on dinic'd judgment of th(> child's general nu- tritional status. Any boy or girl found by the examining pliysician to be in poor or very poor general nut tilion.-il conilition ;it 7 years of age was "selected" by this criterion. Only the i)oys niid girls who were poorly or very poorly nourisheil as distinguished IVom those who I The .sum of 2 iiuasurt'iucnls made accordint: to the Franzpn U'cliiii(iuu. Se€ Appciulix II: (75) Nutri- tional Statu.s Indifi's, \i\u 7-K. ' For a distribution of pprccntuKC chatiRc in arm girth per year, see .supplementary table IV, App(>ndix I. p. 101. 3 For a di.slril)Ution of average ijercentage gain in weight per inonlli, .see .supplementary table III, ApjM'n dix I, p. 101. « See Appendix I, pp. 08-09, for a de.seription of the method used in fitting the exjionential ocjuation to the consecutive weighings of each eliild. 52 ASSESSING THE PHYSICAL FITNESS OF CHILDREN were in a borderline state of nutrition have been used for testing the indices, on the assumption that if an index fails to select obvious cases — that is, children who are in poor physical condition — it will also fail to identify children who are in borderline physical condition. Unfortunately, it is these very ''borderline" children who constitute the major problem in assessing physical fitness, but until the ade- quacy of an index in selecting poorly nourished children is established there is no reason for testing it on a child who is in borderline nutri- tional condition. "Selected," as used here and as it will be used in describing the results of testing the indices, means that on the basis of a given method of evaluating physical fitness^ — in this case, general nutritional status at 7 years of age — a child is found to be physically unfit and is, therefore, in need of medical attention or nutritional advice and assistance. Criterion II is a refinement of Criterion I. Any child who was found by the examining physician to be in a state of poor or very poor nutrition at 6 and at 7 years of age was selected by Criterion II. Criteria III and IV are based on the child's average percentage weight gain and percentage change in arm girth. Although estimates were made of the increase or decrease in each of seven antliropometric characteristics of these boys and girls, only weight and arm girth have been used in deriving criteria of physical fitness. Weight has been used because it is a measurement which is easy to make accur- ately and because it has been carefully studied, although it has the limitation of being a three-dimensional or volumetric characteristic, which may be more closely related to the child's skeletal development than to his physical fitness. Arm girth has been selected not only because it is easily measured and the error of measurement is relatively small (table 19) but also because it reflects the child's increase in musculature and in subcutaneous tissue as well as his skeletal develop- ment. It has another advantage in its large relative variation. Next to weight and subcutaneous tissue, arm girth has the largest coefficient of variation of any of the eight anthropometric measurements included in this study.^ In other words, there is considerable variation in the 5 The coefficients of variation for the measurements made when the boys and girls were 7 years of age are as follows: Measurement CoeflBcient Boys Qirls Ann girth _ 8.39 5.39 5.30 4.77 5.70 4.73 13. 30 14.02 10.23 I5i-ili;ic crests 5.59 Bitrochantcric width 6.20 Chest breadth . 5.43 Chest depth 6.27 Height 4.16 Subcutaneous tissue 14.81 Weight-.. 16.85 RESULTS 53 arm girth of the boys and girls w^iich may have biological as well as clinical significance. According to Criterion ///children have been selected who exhibited an "unsatisfactory" gain in weight. If a child's average percentage increase in weight per month is exceeded by at least 90 percent of the boys or the girls included in this study/' his weight gain has been arbitrarily defined as unsatisfactory.' In other words, only about 10 percent of the children exhibited an avei'age percentage increase in weight as small as or smaller than that of the child in question. This definition is empirical and open to criticism, but, on the other hand, it seems reasonable to assume that a child of this age who exhibits so small a gain in weight is not in good physical condition.** Criterion IV is based on percentage change in arm girth. Any child whose percentage change in arm girth between 6 and 7 years of age was in the lowest 10 percent ® of the group of boys or girls included in the study is selected by Criterion IV. All the children identified by this criterion exhibited a percentage decrease in arm girth. Criterion V is the most restricted of the five criteria. It is a modi- fication of two of the others and involves both clinical judgment and growth estimates. Any child who was found by the pediatrician to be in poor or very poor nutritional condition (Criterion I) and in need of medical and dental care when examined at the age of 7 years, and who exhibited an imsatisfactory percentage change in arm girth, as in Criterion IV, is selected by Criterion V. Number of children selected by the criteria. The numbers of boys and of girls identified by these five criteria vary considerably. (Table 20.) Criterion III (average relative monthly gain in weight) selects the largest number of boys, 37 (10.1 percent), and girls, 37 (10. G percent); Criterion V, based on clinical judgment of general nutritional status, need for medical and dental care at 7 years, and percentage change in arm girth, selects the smallest numlx^r of children, 4 boys (1.1 percent) and 5 girls (1.4 percent) . « Theoretically, the lowest 10 percent of the children of each sex were to be selected. This would mean identifying 37 boys and 35 girls. However, it was necessary to select 37 instead of 35 girls because of the limitations resulting from grouping the material. All the boys who were selected showed an average per- centage gain in weight per month of O.r.57 or less; the girls, of 0.r)72 or less. ' This estimate of i)ercentage weight gain is a more refined measure than the one u.sed in the preliminary report of this study. See Appendix U: (lOG) Souther, Eliot, and Jenss, p. 437. 8 Although it may well be that an exceptionally rai)id percentage gain in weight is as significant as an exceptionally small gain, a child whose weight gain is exceeded by approximately 90 jwrcent of the boys or girls included in the study is not likely to be in very good health. « Theoretically the lowest 10 iM-reent of the children of each sex were to be selected. This would mean identifying 37 boys and 35 girls. However, it was neces,sary to .select .32 instead of 3.') girls N-cauiie of the limitations resulting from grouping the data. All the children who were selected showed a percentage decrease in arm girth: 1.3 or more for boys, 0.7 or more for girls. 54 ASSESSING THE PHYSICAL FITNESS OF CHILDREN Table 20. — Number and percent of the 365 hoys and 348 girls selected by each of five criteria of physical fitness Criterion I. Clinical estimafe of poor or very poor nutritional status at 7 years of age --_ -. II. Clinical estimate of poor or very poor nutritional status at both 6 and 7 years of age III. "Unsatisfactory" ' average percentage gain in weight per month IV. "Unsatisfactory" 2 percentage change in arm girth per year. V. Clinical estimate of poor or very poor nutritional status at 7 years of age, need of both medical and dental care at 7 years of age, and "unsatisfactory" 2 percentage change in arm girth per year Boys Number 28 18 37 37 Percent 7.7 4.9 10.1 10.1 1.1 Girls Number 40 23 37 32 Percent 11., 5 6.6 10.6 9.2 1.4 1 The lowest 10 percent of the group of boys and girls included in the study have been considered selected by this criterion. This would mean identifying 37 boys and 35 girls. However, it was necessary to select 37 instead of 35 girls because of the limitations resulting from grouping the data. All the boys who were selected showed an average percentage gain in weight per month of 0.657 or less; the girls, of 0.072 or less. 2 The lowest 10 percent of the group of boys and girls included in the study have been considered selected by this criterion. This would mean identifying 37 boys and 35 girls. However, it was necessary to select 32 instead of 35 girls because of the limitations resulting from grouping the material. All the children who were selected showed a percentage decrease in arm girth: 1.3 or more for boys, 0.7 or more for girls. Evaluation of the criteria. None of these criteria is a very satisfactory measure of the child's physical fitness, although each has certain relative advantages and dis- advantages. For example, the clinical examination might identify a child with flabby muscles who probably would not be selected by a growth estimate such as percentage gain in weight. On the other hand, clinical judgment does not measure dynamic aspects of the child's growth and development so accurately as growth measures based on seriatim observations of the child. As none of these criteria is ideal and there is no valid measure to be used as a standard, the reader is left to choose for himself the criterion or criteria which he is willing to accept as a more or less satisfactory measure of physical fitness. He may, of course, find himself imwilling to accept unreservedly any one of these criteria. In that case, how- ever, he will certainly not reject all five as without some value, since the factors on which they are based are generally known to be closely related to physical fitness. In other words, although the failure of an index to agree with any particidar criterion may not prove that the index is inefficient, the reader will probably concede that its failure to agree to a considerable extent with all five methods of assessing a child's well-being justifies the conclusion that the index is not efficient in identifying children who may be in need of medical attention or mitritional advice and assistance. RESULTS 55 TESTING THE INDICES Because the several indices measure different aspects of physical fitness and use different methods of identifying children who are in poor physical condition (pp. 10-17), it will be necessary to describe the association between each index and the five criteria separately. The Baldu'in-Wood Tables. It will be remembered that the Baldwin- Wood Tables give the child's expected weight in terms of his height for his age and sex; that the child's actual weight is compared with his expected weight and the difference between them is expressed as a percentage of the expected measurement; '° and that if the observed weight is 6 or more percent less than the expected, the child is selected bj^ this index as being underweight. • Methods of analysis. — Before evaluating the efficiency of this index in selecting children identified by each of the five criteria, it may be well to digress for a moment for the purpose of outlining the methods used in presenting the results. The data were first analyzed in the form of "fourfold" tables (table 21), which arc useful in presenting observa- tions when the frequencies of each of the four possible combinations of two attributes are known in respect to presence or absence. To cite an example based on the Baldwin-Wood Tables: The girls included in this study have been classified into two groups on the basis of clinical judgment of nutrition at 7 years of age — namely, girls who were poorly or very poori\ nourished (clinical judgment) and girls who were not. They have also been classified, according to the Baldwin-Wood Index, into two other inde- pendent groups consisting of girls selected by the index as underweight, and girls who were not. These two classifications (based on clinical judgment of general nutritional status and on the Baldwin-Wood Index) can be further refined and interrelated to determine the nimiber of girls who were in poor or very poor nutri- tional condition who were also selected by the Baldwin-Wood Index; the number of girls who were identified by the criterion (clinical judgment) but not selected bv the index; those identified by the index and not selected by the criterion; and, finally, the girls selected by neither the index nor the criterion. This type of analysis is illustrated in table 21. According to this table the criterion classifies only 40 of the 347 girls " as poorly or very poorly nourished at 7 years of age; the index identifies 2.') (02..") percent) of these 40 children as underweight for their iieight and age, although it selects 52 other girls as underweight whom the physician did not consider poorly or very poorly nourished. In other words, although the index selects nearly twice as many children as the criterion '^ it fails to identify 1.5 (37. rt percent) of the 40 girls whom the physician judged to be poorly or very poorly nourished. '" For a distribution of these relative differences sec .supplcmiTitary tabic XIV, .\ppendix I, p. 111. " The Baldwin-Wood Index could not be applied to 1 of the (IS girls included in the study as her height exceeded the measuremonts given in the Baldwin- Wood Table. " The index selected 77 girls; the criterion, 40. 56 ASSESSING THE PHYSICAL FITNESS OF CHILDREN Table 21. — Association between clinical estimate of nutritional status and the Bald- win-Wood Tables for 347 ^ 7-year-old girls ^ Clinical estimate of nutritional status Baldwin-Wood Tables Selected as under- weight ' Not se- lected Total Poor or very poor Other* Total 25 52 77 15 255 270 40 307 347 1 348 girls were included in the study, but the index could not be tested on 1 girl whose height exceeded the measurements in the table. 2 For the Baldwin-Wood Tables age at nearest birthday was used; at the physical examinations age was defined in completed years. 3 6 percent or more underweight. i "Other" included children whose nutritional status was estimated as excellent, good, or borderline. This method of analyzino; the data (in the form of fourfold tables) may be condensed ^^ for inclusion in the table which gives the results of testing the index (table 22) to show the number of children selected by each of the five criteria, the number identified by the index, and the number selected by both the criterion and the index. This table also gives the children identified by both the criterion and the index as a percentage of those selected by the criterion and shows the percentage of the boys or girls included in the study who must be screened by the index to include all the boys or girls identified by each of the criteria. In interpreting the results of testing this or any other index, it should be remembered that to be effective as a method of assessment, an index must have both specificity and sensitivity. If an index is specific, every child who is selected is in poor physical condition but not every child who is in poor physical condition is necessarily selected. On the other hand, if the index is sensitive, it will identify all the children who are in poor physical condition, but not every child it selects is necessarily physically unfit. To put it in another way, there are two requirements which a satisfactory index must fulfill. When applied to a group of children, it must pick out all or nearly all the children who are in poor physical condition, and in addition it must reject all or nearly all the children who are physi- cally fit. Does the Baldwin-Wood Index meet these requirements? This question will be answered for each of the 5 criteria. 13 It is impoitant to point out that the original fourfold table used in testing the index can be derived from the summary tables. For example, according to table 22 there were 347 girls on whom the index was tested. Criterion I selected 40 girls; the index identified 77; 25 girls were selected by both the criterion and the index, and 15 girls identified by the criterion were not selected by the index. Similarly, 52 girls identified by the index were not selected by the criterion. Now, if the index identified 77 girls and it w^as tested on 347, it necessarily follows that the index did not select 270 girls. Similarly, if the criterion identified 40 girls it failed to select 307; and finally, if 307 were rejected by the criterion of whom 52 were selected by the index, 255 were rejected by both the criterion and the index. These observations form all the nca^ssary data for completing the fourfold table which shows the association between Criterion I and the Baldwin-Wood Index. RESULTS 57 o to GO S -a Si. c 03 to a, 8 to 1^ e s c fa. a H CQ tJ >> IC >o c OS %^ e o- oi t^ tc a o^ g.. t^ ff 0 >o OT ^ 1=1 '^ "C 5 1^ tc «> 1- f g w t- t^ « cc — ' C3 ■2 =^ C^ « m S2 S 52 s V^ ^ tt tf t^ ■* ^ r~ ^ -^ £ u: -^ 3k '^ 0 o _ S r- to 0: 10 scree ide al erion h. M (j; >> ildren 0 inclu he crit o PQ K. tZ '•O IM CJ in ^ r^ »f: "^ 00 1" Oi « cc j3 ■*-i -^ 1 o ■OTS-e.i a ir- 01 cc c 0 entifie ion an percen ident ritcrio es s ^ s 5 en id riteri X as hose by c CC to to d 0 hildr by c inde of t fled o •0 c5 tc 1-- 10 O u: -* 0: oc CO n ej r-t g 'C ■a o a 0) a »c 0 DC tc ^_^ o w »-H 1 X3 1^ o PQ ■o o Q^ tn l~- t^ t-- t-' t>- *^ to r^ 1^ r- t-- t^ a 1^ S 5 n 2 •ts a IC »c »c »c 10 HH CO «5 to to CO CO IS >> u o o W t-l o .Q o M t^ w ■0 OT ■^ > u 0 " p a OS J3 a> u V 0 >* r" a 4-> a o OtD ^ 8 §§l| ^a • 0 = 0 > "* o r- ■^ ,J c - « ..J 0«j - H K ■ nil? •II ^1 Cfl en en 4j) .i£ C I. "^- cc _ £. ■§■§ ■a -a 5a cn*r" ■5.2 0.2 ^1 n.t: e.2 e 3.1: ^ ■*^ ^ -*-■ t) 6i 0 a 0 -^ i-H 1— J H-4 > P^ 1-^ »— < 1 "O "O a> ^ c 0 0 A CO •a 0 n 0 0! 0 ^ j ^ g 0 0 T) > .a u & ■*-a >* H J3 1 Si Oi ^ < S •if^ \ ; < 3 a *k4 J3 0 ■0 CO "O Lh a V S 00 >> X X3 XI <4^ •a ,4^ J3 •s 0 0 0 T) © A u ^ a) •0 u bs ■0 n *-4 0 a 3 3 0 k^ S ■a 1 S 3 XI 1 XJ -4^ 0 0 > X) 03 >> -t^ XI Tl g ■0 3 CO ■0 3 3 tn nX3 0 . 0 "S-==- ~ a « 3 a c _- i 3 o .s c«- « a .S O o*^ CO - to *J .- • ' "al CO ^X2 ■Cto oE 3'ci s; CO 3 CO -t t. o-a " o t ^ 3.-= c,^jr o 11 = ^2 § " -— -- C> S « ~ — c — ~ tCX: ;>.t£ ^o w p" >; •-> c .«~» .fcj *3 r *"^ c3 weight-gain esti- mates are exceptionally accurate, for they arc based on an average of eight weighings made over approximately a 2-year period; (2) about '« Those estimates of average relative pain in weight ix-r month are probably more aeciirate than the estimates of iiercent ape gain in weight per year used in the preliminary report (see .\pi)enf)ix ll:O0(>) Souther, Eliot, and Jenss, p. 137). C'onsenuently the results of testing the indices iu terms of these 2 forms of the criterion are not strictly comparable. 239848°— 40 5 60 ASSESSING THE PHYSICAL FITNESS OF CHILDREN 90 percent of the boys and of the girls included in the study exhibited a larger average percentage increase in weight per month during the 19- or 20-month period of observation than the boys and girls selected by the criterion; and (3) these children were probably in poorer physical condition than the boys and girls whose weight gain was more satisfactory, although not all children who are in poor physical condition necessarily exhibit a small weight gain. As has been pointed out, however, the Baldwin-Wood Tables identified only about 20 to 25 percent of the boys and girls selected by this criterion. If the index is set to identify all the boys and girls who exhibited an unsatis- factory weight gain, it would identify about 96 percent of the boys and about 97 percent of the girls included in the study. In other words, practically all the children would have to be selected by the index if it is to identify the small number whose weight gain was unsatisfactory. Criterion IV. — This criterion, percentage change in arm girth, ^^ has none of the virtues which result from widespread application and study such as the weight-gain estimates have. It has been used in this investigation, however, because it has the probable advantage, in comparison with weight-gain estimates, of not being influenced to the same extent by the child's skeletal development. In addition, the percentage change in the arm girth of these children varied consid- erably. The range for boys was from a decrease of 6.3 percent per year to an increase of 21.6 percent, with a mean increase of 3.12 percent and a standard deviation of 3.58 percent. Corresponding figures for the girls are as follows: Percent Range -9. 1 to +19. 5 Mean 3. 63 Standard deviation 3. 64 In other words, the variation in this growth estimate is large enough to assume biological and perhaps clinical significance, for there are some indications that marked changes in a child's physical well-being are often reflected in his arm girth. It is for these reasons that per- centage change in arm gu'th has been used, as Criterion IV, for iden- tifying boys and girls who are likely to be in need of medical attention or nutritional advice and assistance. For the application of this criterion, the boys and girls were each grouped in order of percentage change in arm girth per year. Then the children who were in approximately the lowest 1 0 percent of each group were arbitrarily selected, as showing an unsatisfactory per- centage change in arm girth (a decrease of 1.3 oi- moi-(> percent for boys and 0.7 percent or more for girls) between 6 and 7 years of age. 17 Tho sum of 2 nioasurements according to the ti'chiiiciiio described in Appendix U: (7S) Nutritional Status Indices, pp. 7-«. RESULTS 61 The Baldwin-Wood Index identified only about 16 percent of these 37 boys and 25 percent of the 32 girls who were selected by the cri- terion. If the index is required to select all these children who showed an unsatisfactory change in arm girth, it must identify about 96 per- cent of the boys and the same percentage of the girls included in the study. Criterion V. — The limitations of each of the four criteria which have been studied are well recognized. Two of these criteria are entirely dependent on clinical judgment; the other two are dependent on growth estimates. It was, therefore, decided to emj)loy a criterion (Criterion V) which involves both clinical judgment and estimates of growth, and, in addition, includes the physician's judgment of the child's need for medical and dental care. In order to make the test very rigorous, the following requirements for selection by the cri- terion were made: (1) The child's general nutritional status was judged by the pediatrician to be poor or very poor at 7 years of age; (2) the child needed both medical and dental care at the age of 7 years when the pediatrician made the physical examinations; and (3) each child was in approximately the lowest 10 percent of the group of boys or girls included in the study with respect to percentage change in arm girth between 6 and 7 years of age. As would be expected, only a small number of cliildren — 4 of the 365 boys and 5 of the 347 girls — on whom the index was tested, were selected by this criterion. Examination of their anthropometric schedules, physical examinations, medical histories, and socioeconomic schedules indicates that these 9 children were in very poor physical condition. Yet the index failed to identify 3 of the 4 boys and 2 of the 5 girls. If it does not select these children who are extreme cases, will it identify other boys and girls who are physically unfit? The ACH Index. The next index to be tested is the ACH Index. It is based on an empirical procedure developed by the American Child Health Associ- ation for identifying children with a small amount of musculature and subcutaneous tissue relative to body build. According to this pro- cedure a child may be selected as falling in the lowest 10. 14. 20, or 25 percent of a group of boys or girls of the same age and hi[) width. Such selection is dependent on the difference between the child's arm girth and his chest depth relative to his hip width. In ovdor to make the results of testing this index roughly comparable to the findings which pertain to the Nutritional Status Indices and lo the Pryor '^I'ables, eacii of which has })e('n arbitrarily set to identify the lowest 15 percent of a large grouj) of boys and girls, the ACH Iruhw lias been set to select the lowest 14 percent of a group of children of the same 62 ASSESSING THE PHYSICAL, FITNESS OF CHILDEEN sex, age, and hip width.^^ At this level only 9 of the 712 children ^^ on whom the index was tested, 4 boys (1.1 percent) and 5 girls (1.4 percent), were identified by the index. According to table 23 the ACH Index does not agree very well with any of the five criteria. When the index was tested on the girls, there was maximum agreement (10.0 percent) with clinical judgment of nutritional status at 7 years (Criterion I). On the other hand, for three of the criteria — namely, unsatisfactory weight gain (Criterion III), unsatisfactory change in arm girth (Criterion IV), and Criterion V, based on clinical judgment and need for medical and dental care at 7 years of age as well as unsatisfactory change in arm girth — none of the girls selected by the criterion were identified by the ACH Index. It was somewhat more efficient in identifying boys, but the maximum agreement was only 25 percent (Criterion V).^ In other words, evidence derived from this study indicates that the index is not suf- ficiently sensitive because it identified so small a number of children. In terms of the five criteria used in this study, it is also not a highly selective procedure. The Nutritional Status Indices. The three Nutritional Status Indices utilize the principle of com- paring (1) weight; (2) muscle size as indicated by the girth of the upper arm; and (3) amount of subcutaneous tissue over the biceps with the child's expected weight, arm girth, or subcutaneous tissue judged in relation to his sex, age, and body build measured in terms of his height, chest depth, chest breadth, and hip width. The method of computing and evaluating the child's relative standing, or "score," in any one of these three anthropometric measures is described in the monograph. Nutritional Status Indices.^^ Table X of that publication gives the child's score and his relative standing in a group of his skeletal peers of the same sex and age,^^ but the authors do not give instructions for interpreting the significance of the child's standing in terms of his physical fitness. Neither is there a definite statement in the monograph concerning the relative value of each of the three indices. In order, therefore, to test the indices in terms of their agreement with the five criteria, it was necessary to make the following " Supplementary table V, Appendix I, p. 102, tests the index set to select 20 instead of It percent of the children. It may be seen upon examination of this table that the results agree in general with those pre- sented in table 23. '» 713 children were included in the study, but the index could not be tested on 1 boy whose bitrochanteric width was unknown. 20 The ACil Index does not permit analysis to determine the percentages of the children screened by the index in order to select all the children identified by a criterion. " See Appendix II: (75) Nutritional Status Indices, pp. 14-18. " The Nutritional Status Indices were constructed to show the number of children in 1,000, but in this moDugrapb the number of children in 100 is used in order to present the results in pera-ntage form. RESULTS 63 n -tt; ■e-o r^ O O C 1 an ecu ific w o 00 ■^ l~- -t^ t-4 ggagg 3 ■c-c^-e-S _ w 03 o) » hildrei: by cri index of thos by crit M o o *"* PQ O ■^ CJ o o o 1 V3 -o (m a OS 5 II S.S m c^ o - rH -*^ m I 'C >. >. o o X: m T3 c* a »o »o lO IT »o a CO a> 3 2 o a 2 2 2 a •* •^ •^ ■^ ■«»« hH w 5 >. u O 4m n o t^ ® J3 o m t^ CN to a iS -* c^ c « us t4 z s O §s 00 t~ t^ ■^ CQ « « o P3 sg 0 a Hfl « x: — 60 *j 0 CS 2 So v>« X! c O t; 03 CO = j: 1.1 .^ y § 11 o >. So t* «r*^ 03 O 03 si V3 s CD o 4.J ° s CO 0. t- X> c E t^ 0 4J •4^ is CO a ca CO 3 C 2 '4-3 •*-> » cofi s -4.3 >- 3 CO 03 03 •4^ s: S CO CO h 0 '2 J3 ^0 03 3 a "3 a "3 a 4^ o .2 .2 e c 1 *C •4^ 3 ■3 s i3 03 o c a bl ° .9 3 oT 1-1 o a a 3 bfi o a 3 ; as 00 & » a; c ^2 r* t* c 1.. bi 0 >■ >. o o u « § a ° 1 as o o o o p. *4M 1 ! o o ^ 00 o a i :: Q^^ a -4^ 03 e ' b OS*.- C3 a CO o a CO c s a a 3 Q 1 s s tr i 1:3 ^ o ; u -: ^ > >■ HH •3 s ^■ O CO 3 >. co=> 03 a, ^§ x; i- 2o3 ^■5 Si-" Z C o o «§ ^a — o; « o o ^^ JS o ^ -4^ ^3 0.3 X3 M •4J O 3^ -^ H 2 =* "3x1 ^v ".- O to 3 o 2J o a x: © ■si .s:5 u o K CO !0 3 „ S °= 2 ^o.a ■^ to*^ Cl 'O ~ G ♦^ Sis - = •3 Sj= ^ 2 O -rt ^ •§ CO >> 2 o — XI -" >, Xi ■o o ■4.J o "3 CO ■3 o 2 *CO 3 3 o o x: > 03 J3 >> •3 3 3 O o w !3 •4.3 >> J3 •3 s> •4iJ o "3 CO » 2 *C0 a S 3 Ol J3 C3 XI >> ■a 3 o .X3 Cl CO a o ^ _t ■053 "> O ^— I- •3000 3 •— 3 — C-- g " - CO fe 3 K^ O Tub U4 ■" coo C .->, . ^ CO o CO §0 So £L=°a ox:£g « C c •; 3 I- S'5 « S 2 t^ 2f'fe"S |2=a O.S-S ■2 3^5 c3 o » a s2Sg .3 0-3 o s«£a rri £ 03 CO 64 ASSESSliifG THE PHYSICAL FITNESS OF CHILDREN arbitrary decisions about l)oth the chil(]'s score and the three Nutri- tional Status Indices: 1. A child has been considered selected by any one of these three indices if his score shows him to be in the lowest 15 percent of a group of his skeletal peers of the same sex and age as shown in Table X of the monograph, Nutritional Status Indices.^^ 2. Each index and every possible combination of the three indices have been used in comparing the Nutritional Status Indices with the criteria, (a) The weight index; (6) the index for arm girth; (c) the index for subcutaneous tissue; (d) the indices for weight and subcuta- neous tissue; (e) the indices for weight and arm gii'th; (/) the indices for arm girth and subcutaneous tissue ; and (g) the indices for weight, arm gu'th, and subcutaneous tissue have been compared with each of the five criteria. The results of testing these indices are given in tables 24 to 30, inclusive. S3 See pp. 14-16 for method of computing score. RESULTS I e s ■" — .2 >^ <» 1:0 to o a CI kt t», 1 0 0 '^ ^^ c 1 -S-c" coi oi 00 f'-* C! S^ 500 00 01 05 00 •9-s M <^ X 0 S 0 1~> ■cc 0 I*C5 05 CI (^ 05 C'" «o 0 -fl" •- H •gm M cc CC £2 B 1 >.2 .Q 0 god oc 00 00 00 o_ UCI Oi a> 02 0 cree call •ion tfl ^ ildren si 0 includ he critei 0 b. !-<. ,^ t^ „ t^ ^.0 »c ■0 >o »c 1 CO M n CO ^ -4^ ■♦-> 0 •e-o-e.i a ■0 0 0 0 0 entifle ion an percen ident ritcrio (N * ■ ■ 5 drenid ■ criteri lex as those d by c 0 d 0 C 0 c 0 m 0 X r-t 0 0 0 0 ■5 rr. g 'u ■0 0 0 0! a ^0 03 0 0 0 0 0 1 'Ui i>> 0 ■4-i 0 T3 0 0 > 0 0 0 m a 0 C5 t- 0 ■0 M •^ M m CO 3 1^ z a 0 1 0 00 00 t~. t- •<»< 0 V3 n (N M n ■ .i, j;^ ; .g a .Ajs-a a k--" C P u W ' S *^ ' *-> 0 is b fl 3 ; a 03 a a oi "3 .a a^_ oTa u t 0 ^1 « a 8 or very poor of age, need 0 at 7 years of ag ntagc change i > >> a 0 a u c tti s twos _o 0 tr 0 b. c 0 0 b I. u as >._ ft •c •o> ^ o| "r o» ot~2r 0 b- ^ •^ ■*-! •0 , a- 5>*. 1^ v| >*fc. 1-^^. 1| ■is 1 £ c 5i . 5 estimi status il and isfactf er yea 11 V5 'a's 3 a ^1 .2 c ' If C3 *- a "5 "a :H '^ j= B.C a.S H ■; a.!: a.Sc =■- •p3 ♦^ tD =* ?" [ Sw a; tt o*" U 0 >-: ;-; l:^ ^ >■ -^ •"• ja ts 0 0 a> w tn 0 •a 2 0 0 1 0 .a »-t . £ fe a 0 CO ^ a P •0 a eg -o 0 •3 0-5 03 tS A a 0 0 s: a ®^ a C3£ ^ 'J mS S fe 0 — £5 a 0 JS 0 . S -4-» m a S.2 0 t- 0 fto w a> «'S JO '4.A a 0 M 1! £1 S s 0 A oS ?r 0 ^J ■3 s: OK > 0 •3 0 •— 0 S J2 0^ 0 _« "S'w 0 a 0 a c 0 a q ■o ,0 0 ftg Q V) 2^ > 0 ce 4-» 0 x: A £ « >> fcj5 T3 0 2>. 3 a -4^ o-S VI ■0 ^ 3 T «-s? m£ 0 0 X a) .So 11 "O c-i £ =^ a r^T3 fc- ea ~. 3.2 Si 3 '■5 3 — 0 i, £i 2 "■- >-.t u 8 = 5: t£ 0 J3 a 3 ill — Si a c> c/i " = i I. c x:-^ ~ ii .^ >.;:! s 'c - ~ c o o ::. t. M .— -. o »- , X _. "a- s^ -a 5ss!'e:a -!S >•- is "^ a *tfi » p- c ? , . c .= .5 — j: - a — -- - - •ax :, ~ o cs i2§ es « -S: ^ a F C'3 o J3 66 ASSESSING THE PHYSICAL FITNESS OF CHILDREN ^N iH t» 1 1 -H w o I^ r- 1 Is u S § "5 CO t3 aS 1 CO (j; :^ "I^ v> 'O'^ 5 I^C^ N t^ (N o i C3'^ •* CO 1— ( <»2 .0(M IN CO CO »-< S5 1 ■M»C m t^ t^ 00 ii lO CO CO S£ O P3 lO OJ 05 CO "WO CO •O CO 55 325 1 -S£ o ^^ V S XJ-C-lf .i fl o t- O •—< O aj a 0 -w o 5 lO 00 co' ^« ^1? ^ o oi ts ^ o c^ CJ o •— t T-H CO -^ S s 5 ^•2 1 to c^ i-H o 1—1 -' o P5 o "« Qi S «J 2 cs N c^ N CI ^ « W o^ ^ § 'C 2 5 0 S « 2 a ■* ■* ■>)< ■* ■<• S HH 1 O) V, o o CO -^ <«-• PQ o ^^90 CO ■* (N CO CO e 3 'C u-^ S g % a .2 5 s^e *c t§. 0^ 00 00 I— 1 CO co •^ 'o>^ o o •S-* n •^ .^ 1 .Jl, 1 H : a V. 1-1 ' u ■ •^^ I o 3 ; 3 1 1=1 i g E fl 1 a<^ a a c8 "=1 V- 0? H ' s i IS. 61) 1 i> : be a 1 S> i O 1 alog: > bi;P*>i» . ^-I tH 1 "S" S^o §. 1 ® c3 t^ tuo . iM CO o:* *v ' s a -g o O C M o "^ C5 a o " £ o ec O t-i 0«D i-i o o OS oa te •s a« > ■ ^5i^a ; "u Q< (5 2« : 0 >i 1- aslli .2 3 in U^ ■*-* tJt. m o O O < '-' t— 1 >■ CO Si tJ •^ CD c> o o o O) x: 1) CO o 73 Of o l-l Ef T" ® fl o ca !* fl •« ■C 2 ■3 (B 0-y 03 ^ 2 q « CO 2 o « o 5 fl ox: iA Fi ^^ ;_! fl 03 ja ^ 3 CO +J r/j O— ' 08 ft 5<) fl O V-i J3 o . CO fl Si ^ ^ S a® CO '% 5fe xa -4^ ■»~* O CO p>> a X! •a Ol D." 3^ bl3T3 ■a 2 >. 05 OJ -hXI 03 4J o ja OJ CO t> ^ ^^ >> 2 o fl ^>. a CUTS en ■c SI 3 -i ■"t?; co'^ 8 ■2fl.^ 4) CM O M a ■"-o ■^t3 11 Si a o=° 3^ CO O ° -" O 3 ^ 0 co-^ a ^ 2 CO "S- I- Tl siS co.ti bib£0 -e CO . 3 s^ a ■a -§ 3 cu-O S x: fl g PI lo a 0) £-1 o q en bo o 0) tj fl fl*^£ ■" CO « g: a-" •• bf-o-« CO *-> -^ fl •^-> ^ ar, CO 2! cc oj cc £3 4; CO 2 o " o S oi o;^ " fc S 5 ":= co'C >< R>< S .5x1 fic 2 oi £ a MCO _ " C.« C." <<;: 5 -l-l feel's OJ O! 5 O) O J? s •« ^ OQ CO RESULTS 67 B e S O S e I- <» »■* 13 B 8 o S u ; a C4 ■■^ 'B K Os B f^ 00 ^* CO 90 «i s-e s B B 00 ■He" <» J- B «) 8 o CO B a CO (N < u t» 1 "" l-H t>. »-H 00 ^^ ■§« « >rf CO tf o a, <-> Ol a> o> CT. CQ (S: -co "u 5 J. CO M IN CO WO ^"1 SO) IN CO Ci t— t — c •3« CO CO CO CO f '^ s £2 1 X2 CJ V^ ft: • t^ ■^ CO CD ^ -*^ 5; CO CO 00 c4 C> c_ ^OS CO en o» 00 V3 fe-s i; >-. ^ ZJ t-i c b-^ ^^ 00 CD CO S ■«' rr lO CO IN 1" CO CO CO CO "^ o*^*" ^ ■OTSt;./. a o M ^ ^ o o a c -. o ^ ■o *«* oo CO • ent ion pen idt rite 5 hildrenid by criteri index as of those fled by o CO o CO fl o S5 PQ O 1 « N •-* CO f-H o -6 tn _o u xs 5 a 1 -1 o c^ ■o rt tfl 1 C >> o •c o o a ■o iT ■o •o ■o v> W cs C^ e» « IN a 'T o •— 2 o a X £: CJ •V3 p ■* ■"J* Tji ■^ ■^ l-H C/] IN IN CM cs N 5 >. o o o o u o CO t~ IN "5 OT ■<»< CI CO CO d U ;? o 5 M >> ^ 00 ?5 CO Tl< o P5 , ^_^ a ; — _ 1 b 03 ' °3 1 CO CB-" 4) a : a . "* , c o a a .2 ' .2 1 .9 ! 01 • .2--= S^ 5 I ^ 1 '3 1 .9 ' •Hi^E *c 1 Ui ' tu 1 4-> ■ 3 ■ 3 ' c^ 1 V ! c5 2a ti O 03 !2 0 ! ^ i BO a . o a • 03 > OX! .03 o . o a ■a : o_ o a O. ' a" S 1 O [ >> 1 h 1 lU 1 s. i 2 ' > ' > a o s| a I kl o o > 1 O 1 iH 1 ° w "^ 2 o o S •5^ r a w> 1 ° £ o I' 1- ^5 15 1- ^ CS b. ^ •s " « a v> CQ Is "55 a 5 a a a id i£ a.h: -^ ^ -^ V £«ab :s w o3.ii ;» o U * ^ t— ( t-H ~ > > HH »-H 1 .-H g;> Is o >. & tl 'yT ■r. k- C b£ i 00 3 •o « • O « 03 9 S3 « O >. D >>o r *.^ «-> 4/ IC a T-i; OJ eN .52- ^ r- *-l g ^ i^ 3 p ■O3S2S .9o2&a :n Si O o t^ >•— So-' •^ ° s- •• •a-S a5 ■3 £5 -9 ■O S 'c "" ~ 25b°S O i, =! « « •^Sai:* 03 o CO a •— I. O k. "■ " - " i *- 'S to- to to 8 o 00 S < T3 ^ O T3 CI « -4^ X a QJ •n-o Cl '^^ a a a ^ O 00 »;oo I" o Oi ri n +-* S S 8 a-: n o S.2 ft C S fe « » " r^ »*> r-i v^ CD j^xj.s Ota o •S >• i-« o a, b bo O 03 a> 03 - 0^ 1^ CO OS « c« 25 5 5 a 2 M O 03 !? « > » ©"O en 4^ _ 03 ♦^ a! (A 03 — m 03 3 3 « ^ to O &J3 03 =^ 9 eg cu; 5^ a 03 .3 a be 03 J3 c.t: ^ Cfl trt ^ .2-3 « - 5S5.i3 3X3- "^ D-o o a , bjD*^ CV '^ oj o S-rr §» > § => O o cJa'O •::> tC' > a) O t« 03 Q C3 k1 ™ ^^ a ►^o « ■^^ilR "il.i o o n o ■o a 08 a o a a a a 03 S3 o o ■V & o to "O CJIO 03 a " 03 ® xS O) o a-s 2 >> si o °2 «5 ©■"J "3 . .2 i- a-" D >> £-^ MT3 tsS o_o a to a| o £1 a o X3 a «■« ■a ® -fj !> a S 25 a o o > tS J3 >. ■O 3 3 x; -a 3 j t«"J3 a»s:^ . 2-0 r^"^ >- to Svh 2 «* f'— o H !:; o o tos a .2 -n a '>' o •^'512 o'^ ."a u-^ ° ■s i^^^s l55go-| V a ■" «J S P •- .t! r o o C. t? bs '^ a-=! 0-" aa □r 3 « O.X) a S ojto G-Soa TO g-OS'S •3 a £ « ^ m .2 . a g o to „ S " g^Ssl«-2§ owaa-^tu-^'o S a ,« - -H 3 -e ^ o. oi i'^ a £ a f? _3 tnx) ot£2 as Im O 2a bico I— ( l-H "^ 9 •« A to '•3 <0' RESULTS 69 a. C3 S e e s u •O S CQ to ^ ^^ Co ''^ e e ^ SI. ■"^ *~ f^ 90 ©"TS v\ S eo c SI e S5 «. ^ r^^ e-^ ^.^ r* «-* •f^ -c 1^ to ^> ^■^ Tl t C to 1. e to 5?) •«' 1 .i- O to ^* CS to "o- iS o O) ■r» >.. C3 a. ,o O 00 (N H hJ pa < H U, ►> t^ t^ IM "O m ^^ «•* ^ ■«! O M VJ feoo 1 t» » Oi 00 o a; .5 - "u 1 5 m 2 1 s j=~ •* "rr § >i.C Xi ^ o 00 o B_ oOs O) Oi a 00 scree ide ai: erion k. « (^ >. dren inclu ecrit o m t°' o> CO •* ■o< ^m m >r CO « s" CO c^ CO yi •~ OJ= J J3*j*j <^ ^ — -« w ■ -- o o c C o O rt C 'C c S 3 S g-g ^ o * g.on— c 5 en id riteri X as hose by c o o c C o hildr by c inde: of tl fled en >* O c> ■ m Ci o o c C o ■e CO B Im •a o B 0! B _o to o o c C o 1 u» >> 1 x> o m "^ G o; S o o c C o 1 '■^ V3 a "u 3 X o a ^ Zj ■o B t 2 CO c^ C-1 cs cs w i Is s>. O o C n iM E o « t^ cs >o CO •<)• (N c: c^ 3 "b z B O 1 5 o S ^ s 00 t^ n ^ .J. ' __^ ; a a •=5-ga krt 'u • ■*J .*^ 1 bd 3^«S 3 ; 3 ! a C9 a : n» '3 _B B^ sj' a h 1 ^S M h°^':; O 1 en 1 0) cm B S to 03 1 CS U fe o o " 0«0 o 2 o°-SS a i lip O-t^ o.,... «- 3 3 ">. 11 11 Illl? ■■So l| ii • ll|i? cn CO W -tai XT. £ - — ^ 'S — |1 "a "3 .Si B B.2 ■w£ -■1 2 c~ £'■£ ■=o2 3.t S-S « -*^ ::) bi t s ■S E : ei o - O i-J »-( - ^ > — "^ 1 ■S -c •o " o o a o *© A OT3 > S a2 A "o 3-" "O s ^-g "3 o "s-S CO O .w IS ■O J= Bo 1" 1 boy w 15 perce idered s C 8 fi ■- d a tedo lowes en CO a> ■" x:-" > a a •" o ,^_ '-^ 03 >. ■^ fl J= >> *^ "O 0*2 — .S o X ■a.B .Ew the inde res show uded in t of 0.672 0 3-2 but SCO incl — o X .isS - .22 r. b£ &£ O % ^-^ ^ Co •i to >s >*w O " M :•/ r >> ■S -^ o- ^^ — — .— -^ L- c ^ "■•^ o ^r ■" - 5 ■C53X 1/-: 3 «^ im CS d 22 Oc- " ^ E.5 O c J= o •Z=-^£ «^ r^ «^ >x: c o *■ n-a . - w O C O f' t: c p o 5. "^ -a 5fo ^ rs. and idere west — a a to - -i = B -l^B-l Ilill E -5? a a ■S§ X S . " !t-r :l tcx — a C -"•§-' a -1 ti B J3 M a n 70 ASSESSING THE PHYSICAL FITNESS OF CHILDREN S e s CO s CO 1^ "■£ ■*~< -^ '^ s CO «:> ^^ t> (20 CO -• a. f» o CO e fa. a u (0 fe ►> 1 r^ I^ t^ »C Tj* 0-2 Is d »C c" r-4 00 o 01 CO EO <2; "C 73-0 5 t-O o IN ■* 05 H seoo QC CO 0 •" a ■go, C^ CO CO £2 S 1 >.2 -^ Cj •^ •<*< •* Oi (N IN m screened :lude all th riterion o is 1 s o> CO FC *- 00 CO O) IN 00 ^3S •^ai ro ■.^ CO IN •O ,—1 CO CO 2— o s 1 43 +^ -^ o 'V'^-^.i. a o o o 0 0 entifie ion an percec ident riterio CO 5 d en id Titeri X as hose by c CD o o t^ 0 >. CO (N > 1 £1 4-9 O B -o o S o o O 0 0 4^ W a 'H 2 5 a X 2 n ej PJ N c^ > o o o n s o eo t- IN •0 C/3 ■* o CO CO 3 12 Z a .o 5 V3 s 00 l-H CO CO ■.»< O m .Jl. • ^Jl^ 1 0 ! a i utri- both and arm *C ' *l-l ' — H ■ 3 : 3 ; S i ca 1 a ! ■3 i be 1 .a M > >> °1 s ; S3 1 D. 1 i : a ; 5 ; or very poo of age, need at 7 years of i ntage change a u o bfi 1 S : Si2£S ; O O M °,^ CO 1 8 : •c Q o a oco h 1 of po 7 yea atal ca " s pei Q,ea «5 to ' <0^^ S-" r 0 •. 1 estimate status at al and dei tisfaetory )er year,. "S " t5« CA 4^ V>!3 03 S 3 2| 1m 03 0 1i .20. si 11 as tx — — 0 « — '3.2 a.S H.b '5.2a3.t ~ w H 5 til. S "^"^ M 4^ t>bt o o CJ M h-i 1—4 HH > > hH M S ^ 0 0 OJ 0 ja QJ w A -C £^ \f oiS tL 0 Ei CO 0^ as. £ ^ a X 0 ^ ^ !S 0 Im a en 0 0 -o af a 03 !* 0 >> 0 0 J3 0 c ■>^ J2 * "' M °s ^ a a 4i ■o ^ C .^ c 0 si J3 u o 0, o J3 o o •o p 3 >> .Q T3 o iM o 2 B o c o > 03 Si 3 Oj3 &*" TO -fcJ a "> P "^ t" ^ >o '55 -^ c *^ o o5 tS — ' jvj o 0/) 43 oCO 3 O "■OO-S"" 8.9 «-^ a m 1: a i=* O :.- >iE'' r, anils' nM 03 ® ^ t^ ^ ■a S wi a!o- 3 o a; , ® 3cD o ti 35-« S^ Mfl - xa i^-SoT .9- Ox: .a ,0-0 w J3 o 0J3 '■O O ^ -- ri O *M O -, r^ ■•^ +j tj ^.^ "^ .;^ ui fl o »-■ C "r* bx) •r- CJ « 'tj e o -•^ lU ■£ 5)5 o ti o '^ -«; a. ^ CO O CO g 't* i~ C P4. o ec PQ u >. 00 00 ri Ci •^ ■S-o c'~ t>; Tp OJ ^ t--^ Ct~ t^ ^ 00 CO o o 1. aS ^ — *j yj •*i y. a u an 5 g J5 « s — a •0>2 X! Ci —1 1' to-^ '^ T—t a> (N S-^ O IC t § § CO Iron scree nclude all criterion o ^ P3 ».(» 00 to ,_4 00 WCJ 2 -^ c>5 M ^I^ •— 1 CO cc s J3-»J*^ ^ o ^2; •o-o-s-- a o o o o o o a a*j o OT — ; " o 5 drcn id ■ criteri iex as those d by c >> o c> o o O o J3X2.5 ©« O n O 0^ o o o O o •o 1^ a !-• ■o 5 a a .2 to o o o O o 1 'C >> i X5 o •o o ^ . h- Tl< O m c< i-H n m 1 1 'si ■| % 3_g<>SoS 1 n ; a '3 G c_ P a 1 y* \ U CJ bi) uo^- ; ^ 1 > a, -I O 6i °5 1 oi be a 8 c y c8 a I- o tXi a Si2£g i o 08 8 •c 01 O iM Q.C3 o O 0! o > « _ 5^5 'C O t: _ JS o I- -S 1 u •^ c C» 1 Vm o^ : c - .5 a .§ = 2^ en 03 03 2 CO 1- c ti O 4- w •- c: i O 1^ "^ •— 'i' CO tf VJ -4- « ^ OS ♦-> 2. sg II — X .^.s > If S =~ a w a. 2 O.S = s c.t c c i Si ^ -^^ ?* ? " 0 s— a: ei. o o*" U MH 1-^ 1^ K* > •— * »-H » to o J3 B o •o a C3 a o a a 3 to OS ^ £1 -•J 2 i _o 'C o a a! J= o o to O >> O a o ■a o o X2 a 3 o u X o to o -3 9 == £ OS 53 » ^ to o « s. ox: »-. ^ Q« CS c ■£§ .2 5 .C to o5 £•>> cC o-t; ■«■:> CO §•2 I-" s al to ^ p a a . •a ; u o d 2 u o J3 a _2 to >. •a ■«^ u o "3 to •a o I-I 1 8 a > ff o to "5 o !*> • ■a g'-^'^ 2£o£2'^a O _ CO '"^ S^ ti ^■af -S^ ai -;„^ a t, a> t^ "> J2 - o « S .t; ia-oii =•£ M t£ i ^ ice c. a r^ a o ^--a a 2 S "-"*■« •" . a 03 O i. rj p 5, CQ2i«--ts| e5toaioOB°S p ^D a""--"" k. k.Or;OuK"C' ^o a a- g.*"^ B£aa|8gfl oi OJ3 a fc. a 3 aj: rf= c ^ £ 2. t/ r^ _, w ~ CO c. -'• a"' a" ^ o p 2 72 ASSESSING THP: PHYSICAL FITNESS OF CHILDREN Examination of these tables permits the following broad generaliza- tions: (1) There is some variation (from zero to 7.2 percent) in the number and percentage of boys and girls who were identified by each of the Nutritional Status Indices or by any combination of these indices: Nutritional Status Index Number of children selected by the index Weight Arm girth Subcutaneous tissue Weight and arm girth Weight and subcutaneous tissue ■_ Arm girth and subcutaneous tissue Weight, arm girth, and subcutaneous tissue. 1 2 25 None None None None ' The 2 children identified by both weight and subcutaneous-tissue indices were not identified by the arm- girth and subcutaneous-tissue indices. (2) Probably the only index that identifies a sufficient number of children to justify a detailed comparison with any of the five criteria is the Nutritional Status Index for Subcutaneous Tissue, which selects 24 boys and 25 girls out of the 364 boys and 347 girls on whom the index was tested. It is surprising that so few children were selected by the other six indices. Although no attempt has been made to determine the reasons why the indices failed to identify more children, there is some evidence that the Nutritional Status Index for Weight and the Index for Arm Girth do not satisfactorily describe New Haven boys and girls. According to Table X of the monograph, Nutritional Status Indices, the scores of any large group of children tested by either of these two indices should be distributed according to the normal probability curve, with a mean or average score of 50.0 and a standard deviation of 10.0. If a frequency distribution is made of the weight or arm-girth scores of the boys and girls included in this study, the distribution is not normal, the mean is not 50.0, and the spread of the scores, as measured in terms of their standard deviation, is not 10.0.^^ In fact, the whole curve is shifted to the right and is skew in the positive direction.^^ The reason for these differences is not known because the Nutri- tional Status monograph does not describe the 7-year-old children whose measurements were used in deriving the indices. It may be 2< The mean and the standard deviation can be translated into .r/V values of 0 and 1. respectively. See supplementary tables XV, XVI, and XVH, Appendix I, pp. 111-112, which give the observed values of the mean and the standard deviation of the distributions of the Nutritional Status Indices for the New Haven children included in this study. " A personal conirnunieation froni Dr. C. K. Tahner, TI. S. Tublic TIealtli ."Service, states that the Nutri- tional Status scores for Ilagerstown school children also fail to approximate the normal error curve. See Appendix II: (81) Palmer. RESULTS 73 said, however, that the New Haven children inchided in this studj^form a selected group -^ and that the majorit}" represent two quite distinct nationality groups. About 46 percent of the children were Italian,^^ and approximately 18 percent were American boys and gu'ls. This large proportion of Italian children may partly explain the failure of the indices to identify 7-year-old New Haven boys and girls, but, whatever the reason, the number of boys and girls selected by the weight or the arm-girth index is so small that the v^alue of the index is immediately subject to question. In table 26, which presents the results of testing the Nutritional Status Index for Subcutaneous Tissue, it is found that although this index identified 24 boys and 25 girls, it selected only a small per- centage of the children identified by any of the five criteria. Indeed, the maximum agreement between this index and the criteria was only 25 percent. \Mien the index is tested in terms of the percentage of the boys or girls who must be screened in order to identify all the children selected by a criterion, the minimum is almost 90 percent. In other words, there can be little doubt concerning the low selectivity and sensitivity of the Nutritional Status Index for Subcutaneous Tissue in assessing the nutrition of these New Haven boys and girls. To summarize: 1. The Nutritional Status Indices, either singh^ or in combination, failed to select most of the children identified by any of the five criteria. 2. The maximum agreement is 25 percent for Criterion V. 3. The maximum agreement for any of the other four criteria is only about 14 percent. 4. In 50 (71.4 percent) of the 70 tests (five criteria for each of the seven Nutritional Status Indices for the two sexes) in which the children identified by the Nutritional Status Indices (tables 24-30, inclusive) were compared with the boys and girls selected by the five criteria, the index failed to identify a single child who was selected by a criterion. 5. If any of these indices is set to select all the children identified by any one of the criteria, the minimum number of childroti which the index must select in order to include all the children identified by any of the criteria is about 31 percent; the maximum, 100 percent; and the average, about 81 percent. In other words, there is con- siderable evidence that the Nutritional Status Indices are neitlicr selective nor sensitive if any one of the five criteria used in this study is accepted as an approximate standard ol j)hysical fitness.** '• Sec pp. 27(1. for description of the children included in this study. " For definition see p. 25. •' These indices have also been tested by usiiip a criterion which selects the lowest 20 instead of the lowo.>!l 15 percent of the boys and of the pirls. See supplementary tables VI-XII, .Vpix'ndix I. pp. lO.Vloo. In general these tables conflrin the flndincs based on the selection of the lowest 15 percent of the cliildrcn. 74 ASSESSING THE PHYSICAL FITNESS OF CHILDREN The Pryor Width-Weight Tables. The Pryor Tables are logically an elaboration of the Baldwin-Wood Tables which estimate a child's weight for his sex, age, and height. The Pryor Tables take one more variable into account in evaluating a child's expected weight; namely, his bi-iliac diameter. In other words, they use two variables (height and width of the Uiac crests) instead of only one (height) in judgmg weight. Although the instructions for using these tables do not prescribe a method of selecting boys or girls who may be underweight or of comparing a child with other boys and girls of the same age, sex, and body build, Pryor has approved the use of a procedure which approximates the one used by Franzen and his coworkers in deriving the Nutritional Status Indices. It utilizes the principle of sub- tracting the child's expected weight from his observed weight and expressing this difference in terms of the standard deviation of the regression equation for weight on height and width of the iliac crests.^^ It has been arbitrarily decided that a child is selected by the Pryor Index if his relative standing or score by this method of scoring shows him to be in the lowest 15 percent of a group of his skeletal peers of the same sex and age.^" With this selection point in use, the Pryor Tables have been com- pared with each of the five criteria. (Table 31.) It may be seen from this table that there was not a close agreement between the index and any of the criteria. The maximum agreement was 32.1 percent; the minimum, 10.8 percent. If the index is set to include all the children selected by the criteria, the minimum percentage identified by the index is 67.7 percent; the maximum, 97.5 percent. One would expect this index to be as satisfactory as the Baldwin- Wood Tables, but it must be remembered that (1) the New Haven children on whom the index was tested may include a larger percentage of Italian boys and girls than the children whose measurements were used in deriving the Pryor Tables; and (2) these tables are an elabora- tion of the Baldwin-Wood standard, in terms of the widths of the iliac crests of a group of California boys and girls who may not be comparable to the children included in the Baldwin-Wood Tables. But whatever the reason or reasons, the Pryor Index is not an entirely satisfactory method of identifying New Haven children of this age who may be physically unfit. " Soo Appendix I, pp. 97-98, for a detailed description of this method. See also supplementary table XVni, Appendix 1, p. 112. for a distribution of the z/ ^ U c o CO CO w ►J < fe >» 00 00 •r m t ^ I -s* c; *■ -i -< t -^ O t^ fe'O «ij OO 00 0> C3> CC o ^ J; cO t ■^ *-fcJ> tn ftn X 0 u< 5 -IS s 00 ^ S5 CO CO CO o s S2 s •c-5 s *^ r^ ^ s5 •^00 00 00 "0 00 TjJ r-^ od o_ ^00 CO Ol 05 5 o-r f^ Vi (^ KT3 S O C35S o o *-• m h'^ — to «> ,_! 5J=: s «o -rr lo »o ■gpo M CO CO IM ^ o-a s X^-t^ -t^ JS o ^ entifled ion and percent identi- ritrrion lO ■* 00 lO o tn t^ o p ci o t- M n e^ 5 •- « S S" '^ O t, XJ2^ '-I 00 05 00 o hildr by c inde of t: fled o C»5 M 00 o (N m O w < t^ M* ..^ - _c "u ■o 5 a a .9. w 0» "5 (^ >K - 1 Ut >. 1 J3 o m ■B Q OJ «c i~4 vH w^ — ^ „^ -♦J OT C*3 CO CO CO CC 0 o s s o 1 a a; Ci 2 w s s s s s !c ^ o o o m Ui o a 5 CO r^ c^ »o V3 (N CO CO 3 Im 2; a o 5 ^■4^ a W3 S 00 CO CO •^ o n 3 •c '■ 3 : a a utri- both and arm a 0 o ~a a a as •g o a "_ « a a a| it) o a e 8 r or very poor s of age, need o B at 7 years of ag rentage change i o §12 > cs 2 s f poo year alcari » pcrc 'C o ' "s g ja O l~ w " o t^ £> " «j _ el .13 §3 CO c3 1^ • as estimate status at 1 and del isfactnry IT year . to 1§ 1.1 .2 & ax: "S "3 ~ Si p. 2 a. 2 2 -«-* *5 M C.2 £ 3-i: O O = U i-i 1— 1 ^-t ^ >' I-* l-H 1 03 c x: o O o T3 CO s> o a o u 2 o .2 S3 >. 0) u o ■a Ui be?' cafe CJ " —. M CO « S o g"o 3-p5 C to 5; ® -■ x: 01 c *^ c.c c ca o p— ♦^ *j o^ o s o;;:; ■= a£ to o o o 2 « .C be GJ is ^ " •- o £ bc^ o ^ p-c pSg .So P ■2-xi O ® 0^ J= fe ■.> « o =» P g >> 3fl.=^ o — ^ <-■ V, ■ <^" o _• ,= "— .cx!^;^ ^ a ::£2c.sg ■^ J3— t-.!:: ki •>:« J2'S t^ ° ■?■£.!= c •-■- aS ".P = = *- CJ fc- *^ to -. '.o to o . . >, to — •- >. CJ pC ° W.P o — ^-^ .- ^ o ^ o =•-£2" a w .2 M o o t, .5 ~ i j: -P o , ".Pti*^co p o o p § > ca J3 >. 3 4 — > to -J : to i^*j .2-Jt }> i; <5.t3 .- « E ^ fc " ^--o-r — ca J P.? S C J^ -J 239848°— 40 6 76 ASSESSING THE PHYSICAL FITNESS OF CHILDREN SUMMARY Although the four indices included in this study — namely, the Baldwin -Wood, the ACH, the Nutritional Status, and the Pry or Indices — differ so markedly in type that it would be manifestly unfair to comj^are their performance, one fact emerges clearly from the study: There is little agreement between these four indices and any of the five criteria. None of these criteria is an entirely satisfactory standard, but the failure of the indices to identify a considerable proportion of the children selected by all five criteria together with the identification by the indices of an even larger number of boys and girls who were rejected by the criteria means that the indices are probably neither a selective nor a sensitive method of assessing the physical fitness of the 7-year-old New Haven children included in this study. Clinical Judgment THE EXPERIMENTAL DATA As has been stated (p. 45), it is important to determine both the objectivity and the stability of the clinical judgment of the pediatrician designated as "A" who made all the physical examinations during both years of the study. In order to investigate this problem, 208 children (107 boys and 101 girls) ^^ were included in a series of check-up examinations which were made in March and April 1936 after the tests and examinations of the children at the age of 7 years were completed. These check-up examinations may be discussed under two headings: (1) Examinations of each of the 208 children by three pediatricians, A, B, and C, in order to study the variability of clinical judgment of general nutritional status; and (2) reexamination of 103 (51 boys and 52 girls) of these 208 children by pediatrician A in order to measure the stabiUty of her judgment. To make these tests comparable with the annual physical examina- tions the following precautions were taken: (1) The distribution of the nationality of the 208 children included in the check-up examinations corresponded roughly to the nationality of the 713 boys and girls on whom the indices were tested (table 32); (2) the three pediatricians (two women, A and B, and one man, C) who made the examinations were exceptionally well qualified for the task, having had similar training and experience, especially in examining New Haven children — a factor of considerable importance in studying the variability of their clinical judgment; (3) each physician was provided with the set of instructions used by pediatrician A in completing the annual " A few of these 208 children, 21 boys and 18 girls, were not ineluded in the group on which the indices were tested. REvSULTS 77 physical examinations; (4) the importance of following the instruc- tions in arriving at a judgment of general nutritional status was stressed, although the phj''sicians did not work together before begin- ning the check-up examinations to test their interpretations of the instructions or to compare their methods of assessment; and finally, (5) no time limit was set for completing an examination. Table 32. — Nationality of 107 boys and 101 girls included in the check-up examinations Nationality ' Both sexes Boys Girls Number Percent Number Percent Number Percent Total 208 100.0 107 100.0 101 100.0 Italian 90 44 18 8 13 35 43.3 21.2 8.7 3.8 6.2 16.8 47 22 9 5 6 18 43.9 20.6 8.4 4.7 5.6 16.8 43 22 9 3 7 17 42.6 American 21.8 Russian 8.9 Polish 3.0 Irish 6.9 All others 16.8 ' Classification was based on the birthplace of 3 of the child's grandparents. The classification "Ameri- can" includes not only children 3 of whose grandparents were born in the United States but also children whose parents and 2 of whose grandparents were born here. The routine adopted for making the observations was as follows: On the first day of observation the child was examined by pediatrician A.^^ On the following day he was to be examined by B, and also by C. Unfortunately, it was administratively impossible to have A's reexami- nation made immediately after the three physical examinations by A, B, and C. An interval of 13 day? elapsed, on the average, before A reexamined the children. Table 33 shows the number and percentage of these children according to the interval between A's original and repeat examinations. Table 33. — Interval between the initial and repeat check-up examinations of 51 boys and 52 girls made by pediatrician A Interval (in days) Both sexes Boys Oirls Number Percent Number Percent Number Percent Total - 103 100.0 51 100.0 52 100.0 7 10 6 0 7 8 16 0 8 36 7 5 9.7 5.8 .0 6.8 7.8 15.5 .0 7.8 34. 9 6.8 4.9 7 3 0 3 2 10 0 4 16 5 1 13.7 5.9 .0 5.9 3.9 19.6 .0 7.8 31.4 9.8 2.0 3 3 0 4 6 6 0 4 20 2 4 5.8 8 5.8 9 .0 10 7.7 n - 11.5 12 11.5 13 .0 14 t. t 16 38.5 16 .T8 17 7.7 )' On the same day some of these children were also measured by anthropomctrists D and E. Sec p. 45. 78 ASSESSING THE PHYSICAL FITNESS OF CHILDREN It would have been more satisfactory, of course, if the repeat examinations had been made earher. Only a few of the children, however, were ill during the interval between examinations, none of them seriously enough to produce a clinically recognizable change in their general physical condition. Although this routine was carefully followed, it was impossible to make all the examinations as planned. Made during school hours, they had to be fitted into the teaching program. Likewise, children who were examined on Friday by physician A could not be examined by B and C until Monday. In addition, one of the pediatricians, C, became ill and was unable to examine 52 boys and 53 girls until A made her reexaminations. The resulting maximum interval of time between A's first examination and the examination made by one of the other two physicians is given in table 34. Table 34. — Maximum interval between initial check-up examination of 107 boys and 101 girls by pediatrician A and examination by pediatrician B or C Interval (in days) Both sexes Boys Girls Number Percent Number Percent Number Percent Total -- 208 100.0 107 100.0 101 100.0 1 . . 79 0 24 0 0 0 11 6 0 7 8 16 0 8 36 7 5 0 0 0 0 1 38.0 .0 11.5 .0 .0 .0 5.3 2.9 .0 3.4 3.8 7.7 .0 3.8 17.3 3.4 2.4 .0 .0 .0 .0 .5 44 0 11 0 0 0 8 3 0 3 2 10 0 4 16 5 1 0 0 0 0 0 41.1 .0 10.3 .0 .0 .0 7.5 2.8 .0 2.8 1.9 9.3 .0 3.7 15.0 4.7 .9 .0 .0 .0 .0 .0 35 0 13 0 0 0 3 3 0 4 6 6 0 4 20 2 4 0 0 0 0 1 34.5 2 .0 3 12.9 4 .0 5 .0 6 - .0 7 - 3.0 8 _ 3.0 9 . .0 10 4.0 11 5.9 12 5.9 13 .0 14 4.0 15 16 19.8 2.0 17 4.0 18 - - .0 19 .0 20 .0 21 .0 22 1.0 '■ From this table it may be seen that, on the average, a period of 7 days elapsed between the physical examinations of A and B, or A and C, although about 50 percent of the children were examined within 3 days and 38 percent during a 1-day period. REVIEW OF THE LITERATURE Before turning to a discussion of the variability of these findings, it may be well to review rather briefly some of the studies b(>aring on this point which have appeared recently in the literature of the subject. One study is described in an article by Mayhew Derryberry {25) entitled "Reliability of Medical Judgments on Malnutrition," pub- RESULTS 79 lished in Public Health Reports (U. S. Public Health Service) for February 18, 1938. The article may be summarized, in part, as follows : Six experienced pediatricians examined, independently, 108 1 1-year-old hoj's who represented the entire 11-year-old population — both resident and nonresident — of an institution in New York City. There was no time limit for making these examinations; the boys were usually stripped to the waist, and their nutrition was graded according to the Dunfermline scale ^^ as "excellent," "good," "fair," or "poor." The physicians who made these examinations differed markedly in the number of children they found to be poorly nourished. The number each physician selected was as follows: 2, 6, 7, 10, 12, and 15. "But even more confusing was the fact," writes Derryberry, "that children classed as malnourished by one physician frequently were not the same children that were rated malnourished by another physician * * * jj^ all there were 25 of the 108 boys rated 'poor' by at least one of the physicians but only 1 who was so rated by the entire group of doctors * * * Two of the cases were given every rating in the scale." ^* Derryberry confirmed these findings ^6 by the analy- sis of a similar series of examinations made by five women physicians on 113 girls ^^ attending an institution in New Jersey. In summarizing the results of this study Derryberry concludes that whether or not a boy or girl is rated as malnourished depends more on the physician who is the examiner than it does on the actual con- dition of the child. Another study, made by R. Huws Jones (51) and reported in the Journal of the Royal Statistical Society, Part I, 1938, is also con- cerned, among other subjects, with the variability of clinical judgment. Pertinent sections of this report may be summarized as follows: The observations were made in three localities, Liverpool, Manchester, and Prescot. In Liverpool 142 white boys " who were attending two schools in a poor district were examined by four experienced male members of the Medical Department of the Liverpool Education Committee. These physicians were asked to assess nutrition as they would in ordinary routine examinations and to place a child in one of eight nutrition grades: 5, 4+ (excellent); 4, 4— (normal); 3+, 3 (subnormal); and 3 — , 2-f (bad). Although they discussed the definitions and points of procedure before making the examinations, their judgment varied considerably. The results of these examinations may be summarized as follows: Although all four physicians agreed on 34 percent of the children, most of them were normal in nutrition; the number of cases of excellent nutrition ranged from 1 to 17 and of subnormal or bad nutrition from 8 to 23; for 12 of the 142 boys there was a differ- ence of three or more grades in the physicians' assessments, and for one child, a difference of five grades. Of 30 boys whose nutrition was graded as subnormal or bad by one or more of the four physicians, only 3 were there by agreement of all four. " Sec Ai)i)endi]c II: (tS) Derryberry, p. 265. "Id. >' There wa.s more uniformity in the proportion of children cliuvsifled as poor than in the provioi» experi- ment, hut the disaRrccment in the ratings of diflcrcnt physicians was even more striking. ^' Age unknown. " Age unknown. 80 ASSESSING THE PHYSICAL FITNESS OF CHILDREN The second series of observations were made at Manchester, where four assistant school medical officers — two men and two women — examined 168 boys,'* the entire population of the senior boys' department of a Manchester school. These physicians classified nutrition in only one of four grades instead of eight as in Liverpool: A, excellent; B, normal; C, slightly subnormal; D, bad. One of them was unable to place certain boys in any but two grades and classified them as either good or poor. Nevertheless, the Manchester inquiry in general confirms the Liverpool findings. The third test was made at a school in Prescot, where four male medical officers, one from the staff of each of the four local authorities cooperating in the study, examined 155 boys, '® using the Board of Education's classification of four nutri- tional grades. It is interesting to note that although these physicians came from four areas, they agreed somewhat better than the physicians in Liverpool and Manchester. This same English study contains also an evaluation of the con- sistency or stability of clinical judgment. This part of Jones' report may be summarized as follows: Five of the members of the school medical staff of the Cheshire County Council — two men and three women — examined twice each of 193 boys attending two schools in Norwich, one in a good district and the other in a poor district. There was an interval of a week between the first and repeat examinations made by each physician. Some of the results of this study are pertinent to the present discussion. In the second examination every one of these physicians found a greater percentage of boys in excellent nutritional condition, and four out of five found a smaller per- centage whose nutrition was subnormal. This difference may be due to the fact that at the initial examinations the children attending the school in the good district were examined first, but when the repeat examinations were made the procedure was reversed. In other words, it is possible that after examining the boys from the poor district the physicians were all the more impressed by the condition of the children in the other school. Thus, the second physician, when he reexamined the same children, decreased by about one-half the number of boys whose nutrition he judged to be subnormal. The analyses of these data also show that one of the five physicians changed his assessment in 20 percent of the cases, and the other four, in 27 to 31 percent. This means that, on the average, they placed one boy in four in a different grade at the second examination. One of the physicians even changed a "slightly sub- normal" diagnosis into a diagnosis of "excellent" and another an "excellent" into a diagnosis of "slightly subnormal." Further analysis shows that for children found to be malnourished by one or more physicians, consistency of judgment varied from 8 percent for the first physician to 70 percent for the fourth. These findings were verified by the results of a second experiment undertaken at Bolton, where two local school medical officers and three others (three women and two men physicians) from Leigh, Southport, and Wigan conducted the ex- aminations of 200 boys *" under conditions which were similar to those at Norwich, with the exception of the fact that the differences between the boys from the good and the poor districts were not so great. The results of this study confirm in general the investigation made at Norwich, although the consistency of clinical judgment at Bolton was greater, owing in part " Arc, race, and socioeconomic status unknown. *• Aro, racp, and socioeconomic status unknown. <" Age unknown. RESULTS 81 to the fact that a larger percentage of the boys in Bolton were normal in nutrition. Thus, three of the physicians were consistent in 50 percent of the cases of sub- normal nutrition, the others in 70 and 76 percent, but one physician changed a child two grades — -from "bad" to "normal." In this article Jones summarizes his findings on the subject of cHnical judgment, in these words: Perhaps the most important part of the work reported in this paper is that concerned with the reliability of clinical assessments. It has been shown that the distribution of nutrition found in a given population, and the lumiber and identity of the boys assessed subnormal, are largely dependent upon the particular doctor who makes the assessments. The doctors compared in this inquiry were, with one exception, persons of long experience, and all were urged to take as much time as they wanted over the assessment of each boy. Nevertheless, these doctors show important disagreements not only with each other but also with their own assess- ments of the same population after a short time interval. The present criticism is directed against the method, not against the doctors concerned; in fact, the care these doctors would take, knowing the purpose of the inquiry, leads one to fear that the results set out in the previous pages may show the position in an unduly favorable light. * * * As a result of this inquiry, one may venture to claim that the method of assess- ing nutrition at present followed by school medical officers, on the direction of the Board of Education, is unreliable. The results obtained by that method are, to say the least, of doubtful value.*' Investigations by Betenson (8) and Herd (4^) confirm these find- ings. Bctenson writes: The experiment of mine which he [Harris *'^] was good enough to mention was, I think, the first of this nature to demonstrate what faulty conclusions can be arrived at by various doctors using a classification which had no scientific basis at all. I expect you know that our Board of Education in London about 3 years ago wished us to classify all our school children into four categories called (A) excellent, (B) normal, (C) slightly subnormal, and (D^ bad, and it was on pur- pose to find out how a definite area in South Wales was responding to this classi- fication that I suggested that three medical men, one from each county, and the same of medical women should meet together in one of my schools. The meet- ing accordingly took place, the children were .selected unknown to any of us by a school nurse, and these children passed in turn before all six of us, who were spaced quite a considerable distance apart from one another in a fairly large room, the object being that we should not have any discussion at all on any of the children or see one another's classifications until all the 100 had been seen. When these examinations were completed, the average agreement between any pair of observers was found to l)e as low as two out of five.« The other investigation reported by Herd (46) showed clinical assessment by medical ofilccrs to be "absolutely fallacious." Thus, in one test iu which .SG children were graded as excellent, only 8 were « See Appendix II: (61) Jone.s, p. 33. " See Appi>n nutritional condition, but did not identify the same number as likely to be poorly nourished. <' See Appendix II; (.91) Roberts. Stone, and Bowler. See also references aiven on pp. Sflfl. 86 ASSESSING THE PHYSICAL FITNESS OF CHILDREN The table also shows that A's judgment was probably less consistent for girls than for boys.*^ Table 40. — Estimates made by pediatrician A of the nutritional status of 51 boys and 52 girls at the initial and repeat check-up examinations Boys Girls Nutritional status Initial examination Repeat examination Initial examination Repeat examination Number Percent Number Percent Number Percent Number Percent Total . - 51 100.0 51 100.0 52 100.0 52 100.0 Excellent 0 13 34 4 0 0.0 25.5 66.7 7.8 .0 0 14 35 2 0 0.0 27.5 68.6 3.9 .0 2 16 28 6 0 3.8 30.8 53.9 11.5 .0 2 14 34 2 0 3.8 Good 26.9 Borderline .__-_- 65.5 Poor _. _ 3.8 Very poor .0 The association between the nutritional ratings made of each child at the examinations designed to check the stability of the pediatrician's judgment is shown in tables 41 and 42. These tables indicate that A's judgment was most stable for boys and girls in a borderline nutri- tional condition and least stable for those who were likely to be poorly nourished. If these tables are examined in terms of the number of boys or girls who were placed in the same nutritional grade at both initial and repeat physical examinations, expressed as a percentage of the number who were placed in this grade at the initial examination, it may be seen that at the repeat examination A gave a borderline nutritional rating to about 88 percent of the boys whom she placed in this same grade at the initial examination; a good rating to about 77 percent and a poor nutritional rating to only 50 percent of the boys classified in this grade at the initial examination. Corresponding figures for the girls were about 89 percent, 69 percent, and 33 percent, respectively. In other words, there is considerable evidence that A's judgment was least stable for the children whom it was most important to identify; namely, those who were likely to be in poor nutritional condition. Table 41. — Association between the estimates made by pediatrician A of the nutri- tional status of 51 boys at the initial and repeat check-up examinations Estimates at initial examination Estimates at repeat examination Total Excellent Good Borderline Poor Very poor Excellent 0 0 0 0 0 0 0 10 4 0 0 14 0 3 30 2 0 35 0 0 0 2 0 2 0 0 0 0 0 0 0 Good - 13 Borderline.. 34 Poor .. 4 Very poor 0 Total... fil *> The number of boys and girls classified as poorly nourished is small and the results must be interpreted cautiously. RESULTS 87 Table 42. — Association between the estimates made by pediatrician A of the nutri- tional status of 52 girls at the initial and repeat check-up examinations Estimates at initial examination Estimates at repeat examination Total Excellent Good Borderline Poor Very poor Kxcpllent 2 0 0 0 0 2 0 11 3 0 0 14 0 5 25 4 0 34 0 0 0 2 0 2 0 0 0 0 0 0 2 Ooofl 16 Horderline .. 28 Poor . 6 Very poor 0 Total 52 In interpreting these findings it is well to bear in mind that the pediatrician who made all the examinations in this study was excep- tionally well-trained and had had considerable experience in examin- ing school children, particularly boys and girls living in New Haven. Certainly her clinical judgment is as satisfactory as that of many other physicians. It is also well to point out that even if her judg- ment was not always consistent and objective, it is probably a more satisfactory criterion than the individual judgments of several phy- sicians would have been. A real advantage exists, therefore, in having had the clinical data for this study collected by one observer. Nevertheless, the evidence indicates that A's judgment was lial^le to considerable error and should be followed with reservations as a criterion for evaluating indices of physical fitness or nutrition. OPINIONS AND RECOMMENDATIONS OF OTHERS To sum up this discussion of clinical judgment, it is illuminating to survey the opinions and recommendations of other investigators. Various writers have pointed out the necessity for improving the clinical examination. Lishman states it this way : We have tended in the past to rely too much when making our assessment upon "general impressions," and the more striking skeletal defects arising from insufficient food, as distinct from imperfect functioning of the many processes involved in nutrition, to the exclusion of specific signs of prolmble nutritional deficiency revealed after questioning the parent or teacher and examining the child.=« Brewer is more specific in his criticism. He points out that the average physician has been taught in medical school to diagnose pathological conditions and has had limited experience in examining healthy children. Brewer continues: " * * * though we have* Inindreds of speciaHsts in children's diseases, we have few in cliihh-en's health and these few are not counted as specialists unless Lhcy {i])j)it)ach their function through pathology." " 50 Sec Appendix 11: iHS) Lishman, p. 341. ' 1 See Appendix II: (IS) Brewer, p. 93. 88 ASSESSING THE PHYSICAL FITNESS OF CHILDREN The White House Conference of 1930 has pointed out the necessity for making seriatim examinations and appraising the individual child in terms of the progress made between examinations. Improvement in his condition, the Conference reported, is often of far greater signifi- cance than status at the moment. ^^ The Lancet takes a less conservative attitude in an editorial on the variability of clinical judgment of nutrition. This editorial interprets the problem as an "indictment of a system, not of the medical assessor, of whom the impossible, it seems, is being asked." ^^ Stuart, in a more detailed discussion, writes : The clinician is constantly comparing one child with a composite picture of children of similar age. This picture includes a wide range of variations which he has come to expect on the basis of experience. The way any one physician will interpret a child will depend both upon the accuracj^ of his observations and upon the extent and type of his personal experience. ^^ Eliot illustrates this aspect of the discussion in writing of the phys- ical examinations made of Puerto Rican children by physicians ac- customed to judging the nutritional status of New Haven, Conn., school children : That the usual standard of gauging the physical condition of the children was not adhered to (because of the preponderance of poorly nourished children), but that a standard based on the range within the group itself was unintentionally sub- stituted, will be shown later. * * * It was without question the intention of the physicians who made the examina- tions in Puerto Rico to use the same standards for estimating subcutaneous fat as they had used in similar studies in New Haven,»andso to have comparable data from the two places. However, in the face of the preponderance of poorly nour- ished children and the scarcity of really well-nourished ones, the judgment of the physicians with regard to estimating amounts of subcutaneous fat rapidly became warped, and unintentionally there occurred, in conformity with the variations within the group, a definite readjustment in their whole scale of values, as has been pointed out. Children who in New Haven would have been considered to have a "fair" amount of subcutaneous fat, were, because of this unconscious readjustment of standards, reported as having a "good" amount, and those who in New Haven would have been considered to have a "poor" amount were reported as having a "fair" amount. There is little doubt that the ratings of the fat of these Puerto Rican children are high as compared with the ratings given in New Haven by the same physicians. ^^ In discussing the situation in England, Herd expressed the same point of view: All children inspected in the routine age groups are assessed by the medical officers in regard to their state of nutrition * * *_ This assessment is an attempt to gauge the general physical condition of the child, as apart from specific »2Sec -VpiiciKlix H: f//.9j White House Confercnw on ChikI Ucaltli and Protection, I'art IV, pp.296, 300-301. «.Sec Appendix il: (107) State of nutrition, p. 12r)8. 5« See .\ppendi\ M: (///) Stuart, p. 105. « See .Vppciidix II: (115) U. S. Children's Bureau, Publication No. 217, p. 25. See also Qrcou (4S) for discussion of this same subject. RESULTS 89 defects. Some defects would count little or not at all in such an assessment, e. g., defects of the senses; on the other hand, the presence of a generalized defect like anemia would naturally place a child in a low grade nutritionally. Assessment is made partly by static qualities — stature, bulk, skin, color, muscle tone, etc. — partly by dynamic qualities — alertness, general liveliness, and activity. These qualities are not measurable, except stature and bulk, and it is even questionable whether such measurable quantities should be given much weight in the decision, especially stature, which is so eminently a hereditary quality. Assessment there- fore has to be decided by individual judgment, based upon past experience. There is room therefore for considerable difference in the assessments made by individual medical officers and of this a good deal of evidence has been found.^^ In other words, according to Cathcart: There is no reliable objective measure of the state of nutrition. The physical measurements of the child do not give much help and the other generally accepted signs are in the main subjective, the gloss of the hair, the bloom of the skin, the brightness of the eye, the alertness of response, and so on. Each doctor forms his own mental standard and judges the children by this subjective measure. All subjective measures are liable to great distortion. They seem, no doubt, to the individual to be fixed and sure, but are indeed fluid. His judgment is warped by his immediately preceding stimulus. If he has examined a group of children who are fit and well and the next group is less satisfactory, he ranks the second group lower than he would have done had the first group been only very moderate. Until some objective standard can be devised it is quite impossible to exjject any uni- formity in assessment of nutrition in a wide area. Too often, as Bacon has said, the eye of the examiner "is bedewed with human passion." •'■" An interesting discussion of this aspect of the j3ioblem is also con- tained in a book entitk'd ''National Fitness," edited by F. Le Gros Clark. In this report the author discusses the clinical significance of the term ''normal nutrition" in these words: We arranged for a letter to be sent to a number of school medical officers, asking them to explain to us how they and their assistants interpreted the term "normal" in their reports. Did the term imply that the children so cla.ssified reached a fair average standard for the district considered, or did it imply in their minds that the children approximated to a certain ideal standard? * * * Fourteen of the officers * * * seemed * * * to mean by normal no more than a fair average for their area. One from a niidlaiid liorougli says: "Normal nutrition implies a fair average for the child poi)ulati()n of the elementary schools of the area; it does not imply that the child so classified reaches any ideal standard of fitness." His colleague in a northern borough says much the same: "My own impression is that, in the absence of any accepted standard for the assessment of nutritional conditions, one is bound to be influenced by the general average standards of the children examined." A third from a southern borough gives his opinion that "the word normal as used in the service clearly means average or that usually seen." But even those who suggest that normal means to them the approximation to some ideal, frequently qualify their statements. Thus a doctor from a southern county remarks: "Normal nutrition * * * implies that the medical officer reviewing the child is satisfied that the child's condition is .satLs- factory in regard to nourishment, taking all the factors into consideration." »6 Sep ApiM\ EIGHT The procedure used in determining the child's standing or score for the Pryor Width- Weight Index corresponds approximately to the procedure employed by Franzen in deriving the Nutritional Status Indices scores except that the standard deviation of the multiple-regression equation for weight on height and width of the bi-iliac crests is unknown. It has, therefore, been approximated by deter- mining the differences between the observed weights and the theoretical or ex- pected weights of all the children of each sex included in this Children's Bureau study, squaring these differences, and dividing by the number of boys or girls. ^ It would have been preferable to derive this statistical constant from the differ- ences between the observed weights and the expected weights of the children from whose measurements the Pryor Index was derived or, better still, from the formula for the standard deviation of the regression surface, had the necessary data been available. Although this procedure necessarily introduces a certain amount of error, it furnishes a method for classifj'ing the children according to the Pryor Index. ^ In applying this method, it was necessary to calculate four such standard deviations, since the Pryor Tables define age at the nearest birthday, instead of age at the last birthday as in this study. The computed standard deviations and the number of children on whose meas- urements they have been based are as follows: Sex Males Females. Age (at nearest birthday) 7 years 8 years 7 years 8 years Standard deviation 4.20 G. 13 5.19 5.36 Number of children whose observations were useii in com- puting the stand- ard deviation ' 360 2fi »349 10 1 Observations of 39 children (21 boys and 18 girls) who wore included in the check-up physiodand anthro- pometric examinations but who were excluded from the group of 713 chililrrn on whom the t indices were tested, were included in making these estimates. This fact explains the dilTcrence between the number of children on whom the indices were tested and the number who.se nip;i,'^urpments were u.sod in csilculaling the standard deviation. 2 1 girl whoso height exceeded the measurements given in the table wag excluded. ' The number of degrees of freedom was taken into account in calculating the standard deviation. = ?/(„+/)/?/„. In this problem the parameter e^ may be considered an arithmetic estimate of the child's weight at any given age expressed as a pro- portion of his weight the preceding month. If the child exhibits a continuous weight gain from month to month, e'' is greater than 1.0; the larger e*, the more rapidly the child is gaining. If e'> equals 1.0, the child's weight remains about the same throughout the period of observation. If e^ is less than 1.0, the child is actually losing weight. Since e is a mathematical constant (2.718), the value of e* is determined by the value of the exponent, b, which is the relative instantaneous velocity (-T^/y)' In other words, it is an estimate of the child's average percentage gain or loss in weight per month during the period of observation. For example, if b equals 0.008 (0.0077) or eight-tenths of 1 percent, the child's weight increased at a rate of 0.8 percent per month. In other words, if he weighed 50 pounds at 6 years 6 months of age, he probably weighed about 50.4 pounds at 6 years 7 months. The parameters, b and a, have been determined under a least-squares criterion, as is illustrated in the following sample calculation and graph based on the ob- servations for A. R., the Italian boy whose measurements have also been used to illustrate other parts of the text: 3 WoiRlits were taken at 4-month intervals, at the time both annual physical examinations were made, and about 6 months after the first physical examination. * Number of weighings, period of observation, and dates are average figures. For more detailed descrip- tion of these observations see pp. 25 and 26. APPENDIX I 99 Age in months Weight in lb. Log y (x') = x' log y (X) {X')* (2/) 72. 70 12. 70 40. 75 1. 6101 161. 2900 20. 4483 73. 17 13. 17 41. 75 1. 6207 173. 4489 21. 3446 76. 63 16. 63 43. 50 1. 6385 276. 5569 27. 2483 79. 13 19. 13 43. 25 1. 6360 365. 9569 31. 2967 80. 77 20. 77 43. 50 1. 6385 431. 3929 34. 0316 84. 47 24. 47 44. 00 1. 6435 598. 7809 40. 2164 85. 20 25. 20 46. 00 1. 6628 635. 0400 41. 9026 88. 63 28. 63 47. 50 1. 6767 819. 6769 48. 0039 92. 77 32. 77 48. 00 1. 6812 1073. 8729 55. 0929 S 193. 47 14. 8080 4536. 0163 319. 5853 *An arbitrary x scale (x') has been used, with 60 months equal to 0. Normal equations: 14.8080= 9 log a+ 193.47 b log e 319.5853=193.47 log a + 4536.0163 6 log e Solving: b= 0.007711 a= 37.44 and y= 37.44 eO-oo77iix' Therefore, during the 20-nionth period of observation from 72.7 to 92.8 months of age, A. R.'s weight increased on the average 0.77 percent per month. WEIGHT IN POUNDS SO HI 1 1 1 1 1 1 1 1 1 1 AS 46 — THEORETICAL WEIGHT ^t/ - 37 44 6°^"'"'') O OBSERVED WEIGHT ,u, AGE IN MONTHS Weight of Italian Boy, A. R., Observed at Frequent Intervals From 72.7 to 92.8 Months of Age Supplementary Tables MATERIAL AND METHODS Supplementary table I. — Nationality of the boys and girls Both sexes Boys Oirls Nationality i Number Percent Number Percent Number Percent Total . -- 713 100.0 365 100.0 348 100.0 Italian - - - 330 130 66 59 35 34 24 17 9 9 46.3 18.2 9.3 8.3 4.9 4.8 3.4 2.4 1.2 1.2 166 65 34 34 19 17 12 6 5 7 45.5 17.8 9.3 9.3 5.2 4.7 3.3 1.6 1.4 1.9 164 65 32 25 16 17 12 11 4 2 47.1 American 18.7 Russian . - - - 9.2 Mixed European 2 . .._.. _.__ 7.2 Polish --- 4.6 Irish - -- 4.9 Mixed North European 3 _. 3.4 North European ■• 3.2 Central European 5 _ 1.1 South European 6 .. __ .6 1 Classification was based on the birthplace of 3 of the child's grandparents. The classification "Ameri- can" includes not only children 3 of whose grandparents were born in the United States but also children whose parents and 2 of whose grandparents were born here. 2 Did not have 3 grandparents from any 1 country. 3 3 grandparents from various North European countries. * 3 grandparents from 1 North European country. 5 3 grandparents from 1 Central European country. 6 3 grandparent.s from 1 South European country. Supplementary table II. — Bitrochanteric widths of the boys and girls Bitrochanteric width Boys Girls (in centimeters) Total Italian i Americana Other 3 Total Italian i Americans Others Total 4 364 166 64 134 348 164 65 119 18.0-18.4 0 1 3 11 31 31 51 78 62 40 24 15 10 3 3 1 0 0 0 0 0 1 2 5 15 19 24 41 27 13 10 6 1 1 1 0 0 0 0 0 0 0 1 3 7 5 10 12 10 3 5 3 3 0 1 1 0 0 0 0 0 0 0 3 9 7 17 25 25 24 9 6 6 2 1 0 0 0 0 0 2 0 1 17 25 53 51 56 47 33 27 11 14 3 2 3 1 0 1 1 1 0 1 10 14 32 28 26 17 10 11 6 5 1 1 1 0 0 0 0 0 0 0 2 8 11 11 8 9 7 5 2 1 0 0 0 0 0 0 1 1 18.5-18.9 0 19.0-19.4 0 19.5-19.9 . - 5 20.0-20.4 3 20.5-20.9 _-. 10 21.0-21.4 . 12 21.5-21.9 22 22.0-22.4 21 22.5-22.9 -- 16 23.0-23.4 11 23.5-23.9 3 24.0-24.4 8 24.5-24.9 2 25.0-25.4 1 25.5-25.9 2 26.0-26.4 1 26 5-26.9 0 27.0-27.4 1 27 5-27.9 0 Mean 21.9 1.16 21.7 1.08 21.9 1.35 22.1 1.11 21.9 1.36 21.6 1.25 21.7 1.30 22.3 Standard deviation ...^ 1.42 " 3 of tlie child's grandparents were born in Italy. 2 3 of the child's grandparents or 2 of his grandparents and both his parents were born in the United States. 3 These children did not meet the definitions outlined in footnotes 1 and 2. « Bitrochanteric width was unknown for 1 boy. 100 APPENDIX I 101 RESULTS Growth estimates used in deriving Criteria III, IV, and V. Supplementary table III.^ — Average percentage gain in weight per month of the boys and girls Average percentage gain Boys Girls Total-. 365 348 0.20-O.39 . . 1 17 106 144 63 20 9 4 0 1 0 9 0.40-0.59-.. 13 90 127 68 22 1R 0.60-0.79 0.80-0.99 1.00-1.19 1.20-1.39... 1.40-1.59 1.60-1.79 6 1.80-1.99 3 0 1 2.00-2.19. 2.20-2.39 Supplementary table IV. -Percentage change in arm girth per year of the boys and girls Percentage change Boys Girls Total . - -- - - . 365 348 -08.1 -10.0 to - 0 2 5 17 30 73 115 63 35 14 4 3 1 1 1 1 1 -08.0 to ■ -06.1 I -06.0 to -04.1 2 -04.0 to -02.1.- 13 —02.0 to -00.1 28 00.0 to 01.9 64 02.0 to 03.9 91 04.0 to 05.9-. 74 06.0 to 07.9 35 08.0 to 09.9 24 10.0 to 11.9 10 12.0 to 13.9 2 14.0 to 15.9 -. 1 16 0 to 17.9 0 18 0 to 19.9 2 20.0 to 21.9 0 102 0 ■8 ■to so I e 1 <» a. CO *-§ o so e 55 > m H y, S &< ASSESSING THE PHYSICAL FITNESS OF CHILDREN .S C S « c O aj n - •a ■>: -o .2 — g cg._ t^ .^j C3 ^ © St: ^ .ti £ o X o C U o o pq I a 3 ca Q O o a o O .i3 s B o Cue 3 g3 03 a o a a o o C g SI bo a a C8 a o a (U >-' a S M J= a ■a u 3 a Ui o o a > .Sf '3 03 .!3 ■z a tat be a as JS o >>> ■wis CO 2 •—I Q §=== o S» g03 o rt O O 03 -*J c3 a a o .5 S' >> S *5 :a u &i( u 5 ►- 6 o a >> cxi o. 3 o 03 .2« O CO o XI >> c X2 x: o X] a a o L. a ■ii >. O C3 ■M (U > ^,^■0 X! -S« C8 ^1 x: >, 3 « OJ 3 •4-3 CO -4> V •fl 3x: o. be BJO 5 ® a a oT3i2 o o~' g 3 cu K ■;:: -i: a ^a x:'3 4^x: 3-fi 0) p x; 0) .at! 00 r3 ."S « >i CO , o ©■' - " " k L-x: °'^ O ■♦-» CO o =" iT^co- &5 3-- 3002 £0 &)=> ti o w a ® -x: I- x!5-i-^ o ■" a"p •• a ^ £5 o-x: "a CO a « coiS >>o.2 ■OXl CO "t? o -^ a - c a «•" g £ " •Sic £ o '^ s |.al 03 C3 o Six; SSaO •S fe-S S C o t. C a5P&a ■ -^^^^ ■ s •« 03 CO APPENDIX I 103 i 54. e CO ^ I- O -to S ^ e s I I. Si ~ Si Is" CO tM >. o o O ^ o ■6-° ■goj oi 00 r^' 05 "-"O u oc 00 o> o oo o S § a« — ^ >j CO Q, X 3 ■D'O "m 5 •~c» 05 « t^ o '^"Z j;° o ■^ CO o — • O •2" CO CO CO CO 52 S 1 >>3 -aS s" — " " -^ c;^ Sac 00 00 oo' 00 So o 05 C-. Ol scrco deal erion h. ^ c2.^ O O *-• O ffl fcf^ t^ t~ (^ r~ S u- >o »C »c lO 1" m CO CO CO ^ OJ3 ^ -*^ -*-* o :?; •n-v-^.j. a iO o •^ o o Oy ?; C-J O M entifl ion ai perce iden riterii 5 !< *^ CO P ^^ ,~^ .— . ^H *"* CO •— ■ ^^ '-* •— ' ""• 3 t>. u O «— m o X3 O §^ t^ N »c a 3 CO ■^ eo CO 'Z a 5 X 00 r^ ,, ■^ c^ CO CO *1m ■ Im ' .9 i a i 3i!= ; -*J 1 *^ ' '3 1 is ' ^ I 3 ' tti [ C3 ' a ! c o .9 ■ b^cTb : . to © • . o ex— ' S i O I Q. ; fe : . til a? >>• St. ^ i S ; g i ® '. B 1 o 1 ^ : a P <" ^ "- ^ ; a I- o * ' a I gS£8 : assi; 1 2 « l2 o 2 o 1 s ; a I o 1^ ^ .» o C " 0*j .§2 : E I- fc C9 r- 1^ CO o « ■5>> C3 ^ wt^ fi''?:? C!> •— ' %^ Qi ^ « "3 •— Cy en to tXi bi :ci c. Srt - 11 11 11 is S 2 — 'g J= a.° C.2 ^5 B E.!S — » b: be ^ *-» ^ ■** t3~ D " o O O h-i l;^' w >■ >■ ■- 1— ( o o c o •o b a % o B .if! c C8 J3 O x: o x; E o o X3 o B 3 >i 3 •3 o Si a — 2 2 CO O Si S < ■^< a o go CO C3 U ""' >. ■=•3 1^ C CO c3 kM B S o O Xi _2 "3 B O B § XI •3 3 U CJ B to > O (3 5 >• o2 Bog? B oco^£ 2l2.E£ £ s ° « § ?.3 cot: " to .t " 1. 1^ o tc o o ' . 8 ° tt >> CO- CO to (« O >, .y; >. l^ X! O X - fi 3 .SS-- 3— o =o u3 oc'to^c S;^. .E--5Sx:t:=' =j B ^t: — _ CO o >5o - ^S!:: St; CO ♦J o o o =^ t: ig'SgcfS C o ca CO E oj c c— o _?^ t: cj ^ ci j: S I *— fA. W) E CJ. C o ■a 104 ASSESSING THE PHYSICAL FITNESS OF CHILDREN Si. a, e !~ ^ i- O '-^ Ki -( CO «c — ^ ^ H 51 C ^ s W s Cl ■^ a^^ •c* -O V. S? a Cl ^ '^ "« ^ •*-j Oi -w ■**» s: S n 5i> "^ c? <30 c 'e !~. s 1~ 'o' 'T3 o O -^ e 1 n. S^ o =^ O ?> < < PM p u> !>. ,— < f-< 0 I- t~ »£i I "^ -T TT ^■ »o' ^ §^ 1 ■4J 00 00 d CO 1 Ji ^ gjC *^ *- • ♦J m c. ''"'■ s (^ a- 0 ■c-o c5 »^ oi (M t^ c^ 0 C i; o> 01 •^ CO f-( — a •g(N IN CO CO r-* =i2 -^ l-=I !^ ^ 0 -w ic »C I^ t^ 00 gu:- 10 0 to to cZ. u»o >o c» 05 CO creo e al! ion w ildren si 0 includ he criter >> 0 m t^oi c^ c^ c^ Tj* ^ 0 >r5 10 CO 1 (N « CO X3 -*^ -^ ^ -o-B-e.i 0 tC c 0 ^ © 5 (N c^ CO s en id riteri X as hose by c 0: IN 0 (^ 0 hildr by c inde of t fled >> 0 m r-l IN IN fj id 0 "5 m 0 1— ( rH « T3 i-i a C3 X 0 II w^ •V 0 ^ ,-H M 1 *tM >i >> 0 0 T3 2 t^ t~ t^ t- t^ **j V3 c > 0 0 « 0 t^ s £ 0 CO I^ IN •0 B 3 CA •^ IN m CO •z p ,0 0 5 ."t^ 00 00 t^ r~ ■«< 1^ ^ (M 1— t cc CO 0 0 m Ui ' u ' a ; •3 ' a ; +^ 0 jd b +J ' -M 1 o3 I b ' 3 ; D ' M ; o3 1 a : Oaj ■3 i a > 05 ' '^•" ® 0 CO 0— "^ IC ^ 0 o5 fl CD 0 IB ID ^ i 0 ai a u 0 CO ■ a ' »« to o) 9? 0 0 M t! ! ? ' 0 lu S « 0 U "0^ OcO ■00 0 ' 0 t- ,c ■* 0 £« : 6 ' ■5 " hi- 0 a; •§2 .S 3 S a 0 „ eft « ■n to J2a -'- a43 a 11=3 i5 ?? a! 11 |5 0.2 a. 2 ?•- 0 0 0 ^ H- » 1— 1 >■ o X! c o a o a 4ii a a c CS o o o a o o a X o _a J3 a o x: w iJ o CO -^ o ^ • l- r-l w O i^ ° A M 9 u, O S S ^— *n ca o OT >> K^a .^ o o *^ o 0.0 So C wl "r a) 03 b «« «^ ■ss o >. 5s •" m . _ I'J a o CO _, . — J= =i.5 w aotc^.S a 3^ a ti ■a o o J3 a o XI ■a o a> "53 w "O a> s 'w a o a a> 01 > 03 XJ 3 X3 55 « ■3 c 2.2.fc!'S o QJ W C/3 Jo" O f*» ^-S&S§£ •S_§-03«£o 35 a go bj.g a. a-: fe =^ a w 2 z; fci 5 S o £ c- fc M S S o £ a. o ^ «, a 0) C-x; _ a -' » "= bis fe cob-oi a w »>>cia'^.gi:: ^ Or- O c. Oj O O _^t3 Or- biXi « ^« ca-o^^-o ■w S « ? >> 5f >■ Si, 4)co 2 e.Z?«.^60 r: -•SX! o -e o 03 BJr CO to 03 o c3 fl «; m oJ c o a o 05 c;j3 a c a s caS3— c^ci a- ct! a OS o 43 n APPENDIX I 105 Ui t>. 1^ *H t^ i— 00 •S^' "S « « tei M 0 o-S ^ Oi V 0 s> CJ o> >" 03 a. s| •c ■a2 5 *- n CO « e*5 m 0 ": V M (N CO cs — a ■gcc « CO cc CO ® £ s 5-0 •w :? t~ -T CO « 2*2 CO 00 c4 05 (3„ ?^c: 0-. » OS 00 §■3 9 "»•§ CO a, w-O fe !» ej cj *-■ 0 n U t-H ,~l 00 CO CD 3> •« TT 10 0 •§ m CO CO CO CO .a -*^ -*-^ ^ 0 ^ ■a-o-^.i a « ■«■ 00 •<»• 0 0 B a*j c t-^ r^ 0 05 0 en ident riterion X as per bose idi by critc 5 t^ to Tt* OJ 0 hildr by c inde of t: fled 0 2 ■0 ■0 oo' S ^ t^ ■* ■* CO . 0 0 P3 ■0 a> ta a 5! N c^ CM '■3 cn •^ •* ■»< •««< a 0 TZ S K 5 a a ■a •0 n CO CO CO CO HH to ■* ->> ■^ ■^ ■^ '2 S 0 0 ,•-■ m 0 i-i s JS s CO t^ M >o 1 O) c^ CO CO ■z a 0 1 •c 0 3 0 ^ s 00 CO CO ■^ 0 PQ 1 1 . ; .s i a is utri- both and arm ■5 1 3 : 3 1 0 i 03 a 1 a d ■5 1 .a a_ © a • U 1 em *-• CO te 1 fOSf- : © 1 be <. ca 1 a ; § i 0 1 © be a ca a u © be 4-> or very poo of age, need at 7 years of 1 ntage change a I- 0 be 1 a ^ Vi il f, \ 0 0 en 0 ca ' s 0 c li © ' 0 i2 0 = ^ I 0 f po yea al ca » per i Q,e3 © 1 > • a c 0 a or^ i3. 0 t^ .0 * 4^ c: . (U .„ 0*i : a •. estimate status at al and dei tisfactory per year. . ♦2 IS 2^ .13 ^1 It © ^ .S2 C 11 11 «£ ical nal 'die nsa thi fl.2 a 0 a £ a ^ c.c £ 3.i: '.IS ^ V^ -t^ ? ::j'5 c S — a; be 0 u 0 >-4 ^ >' > I—* t-H © © s •« B o a C3 3 o Si o a o o © x> 3 O © •o c © Si 3 o o t3 o 2 fe © © be k « "^ © >* ©ja ©s J=<1 a o "E — 'E -fl ca u =« >> .2x1 ._ © °'S a © 3"© O VI eaCi <*^ '^ a a © o u © © a a© © ca ^ >> ^3 ■aS ♦^ « a © \ c ■ XI a o a 0 > 03 >> •a 3 te e ©12 3 3 •^ -^ *— .n: i-i w o o o g— Mi: a ^-2.h:u M.t © be o c n "--a 9 o *J !5»~ w i»>tO « >» c/5 x: o ■" ''^ °^^X2 a-T- ,_ ._ b. ^ b« — £ — "5 o ^ ^ ^©^sS og S-aS o5fe2 g C "•■ "c ^i u © a ♦js ■^. CO z" i£ i; © fc © t: 3 r <=■—■'' a k 3 _;£t; ©£ bex: © ©•G a a- ©-""O •?'^aj->. ">.© g-^^-a.5 t^.E S S-Sf :ro3 n w-!*-!! ^•<;? -!;r:g ^ '• ■' r "' © it o X3 106 ASSESSING THE PHYSICAL FITNESS OF CHILDREN s e C33 ^ CO CO '^ &2 ^ CO S *^ CO ■= CO i2 fO Q ^1 no to >o a> OS 00 § iC irf «• «> a3 a *^ ^ iC •o 03 03 CO cree eall ion m ildren si 0 includ he criter O n 1~ (N CJ 05 o> •Tf 1 s ^ s CO I— i •G •*-^ *^ O ■OT3*;.- 0 o o o o o a a P -t^ o m o en identifi riterion ai X as perce hose iden by criteri 1—1 ^H o o o hildr by c inde of t fled r>^ 1-H Q o o o o o w ■a t^ S (3 ^ 5 a » s- N (N o o o -M to 1 *E t» t>> O o pq ^ •o ® ca o O o o o '■5 CO 'u 2 5 a g Q^ p J— 1 1- 2 w •* ■^ ■><< ■«»< T}< 2 P» o O u^ PQ o fe £ X! ■§ CO t^ N kO S en ~ r- ^ O o PQ N l-H m eo _ 1 1 'u* .a i a i •cS^g i 3 i 3 I s i s i ^ocafe ; a 1 P a) en b« ^- ^^ ^ ca be .^ ' a IX) < s i g ; a ; a i CO ■ or very poo of age, need at 7 years of i ntage change a u, O bjo > a ' I- to o 9i o O to o « ca ' R 1 o il. ti g O u 0o , _ C3^ ' 0) .^ a^ : 0 -. 1 timate atus at and del factory year.. t5« as 1a« .§3 ■St; to 03 1! O OJ ^ Cl' t" ii to t.* n to-z:-- o *^ tn to w w .— O. "' ct! *j o. Sg §g en™ aXJ to -tJ linical tional medic "unsa girth ] !s° '3.2 ?^ o.ta '^ -•-> tD M o"" o - O l-H >■ > ■ 1— ( t-H o o a © x: 3 -a 3 T1 ■o tli o ^ O Oj x: o » A r. & '5: C J3 — "So & o 03 Qj a 2 ■a <0 "^ ■n n CS oXI c Si* o fe ^ W o to >, £! o o ^ 3 od S to , 71 fe aj"*!. O 03 P ^ !p ^ o£ a >> 3XJ bJD o C8 O — iJ O o ^ w ana a o 3 o 3 I- — >> OT •^ 3 S 2-t^S l> ■a 3 o bfl l^"3S to.3't;.3 i- K 00 w O >> JD fl<_. O ^ .O 03- §.Sfl, c - a ■o 2 2 3S ii o .— ' -»^ H «— . *^ is bcx! *j B a XI •• to FT-5 a L, q; *j """~- -So hD 3 lia2^--l. X!« gj3 iui£ o ^ 8 .3J3 2gc.gS "^.nSgic. g 3 5 APPENDIX I 107 S s e S o S e •,S> tj ." Si ^ 1—.* cs •*-» s? K o S •*o >^ •f-a ^1 55- s :^ ^ u ^— -c cfl "« oi «: i~. e cc t>. v^ - > ^J ^ r-^ C o <;> 'o' S ;»> S t^ •r» ^- o c a. w ^ 5ii n •p^ o n. R. c X <» a s 00 Pi <; U H ►J A. 0. P ir-t^ o O c ta X a a> 'O'O a "^ a 05 gcag MX) S TJ — " ^ •- CJ3 x: -tJ w •ot3 ■«.:;. a © a o*j o S.2 o. >^ o o 00 o o o m X5 ■a 0 03 D a ■§ o o PQ o o o c o ■o c OS a a s o c5 o m o n a o 3 n u o o Q. >> o " o « §1 > a o >> •- a a s k- a a 03 a & - c - ., o ^ C3 ^a '3 " 22 "2 2 U 9 a a. B.h t3 at 2*3 M fe •2-5 a „ *r" Jii a *^ B^-v a ^ ^ a . S ° ""- 1- MO " 2 o >.o ° i2 ? c a CS X3 o s: o X> 3 ■O 3 CO o ■a o a> •a . ® a a o S 5^ :: V. *^ *-^ — • o ^- — u o "3 to o & o a ^ => to .2 ".t; -a .2 o -a CO 05 a" O > o ■3 a ' to a 8 a a> U3 03 3 a •3£^-S« «13 fea "' to u — tmg ~ o to ^ >» to 35^ go be a .a-o_g££ o „ a^£t:^co ? Gf - E cij «-i£og tfe^ .1: g o £i c t ee ta S3-£t a 2 5 'i S — to a w >• ^ s s > > •^ s l-^ 22 o to^ ei2 U . a g o S - ?' Q a to = w o c^ 2 xjt: o^ tcj= -) o-o a a-- o—'S -^■■.5x;o22^ iljligig 03" 03 5 Q ^ O ^ 09 tn 239848°— 40- 108 ASSESSING THE PHYSICAL FITNESS OF CHILDREN s S e to 1? f. t3 o K •^ • «c V -C 5S ;3i •-i CO fi a o •*o "<-a C3 W CO <» f/l c s- •40 o s to ^ (V) C^ £^ >> Si -^^ 2 "^^ -«e 1^ VJ n 0 ^ s 0 0 CO "e •ri to !-. •rJ 1 0 0 to 3 CO CO 1— 1 CO [A S < 0 to e H (- CJ « <; g«( ^ w a M hJ PL< Ph p w •Sa gcag o a)-— co-O fe » o h ~ OJ3 O o rt a -kJ o 2 ;>,'0 V. ja-i^." Ota O c s seoo •2 '^* s 1^ 00 ■2 "-H 55 CO CO t^ CO O —( CO o « s a -a o s 3 2; a O o m a o O -M o .55 o m a o 3 a t4 o o a >. u > O en 3 fl o® I- >.o > 0! ^^ °:: on aa ?CD a a ; fl S] 1 03 .9 ' (O ttl ts, ] 1^ a ' a 03 ' •s i a U bJO > 03 ' Ml a ' ^ \ Ut > a : -«■ ■*- - C •. 1 --. c "b^ tJ g'3 2 ol a o'3 0, 5|S5 atSx) a t- o a g §X103S ao®B >, O CJ o 03 -*'aj o w * a So3 iX) w Cfl ^-> 11 5 5 ^ CO 11 a ?; u 0 l-H l-H "^ O ■a o f? aj4-» » ■01 5 a g£ > ■3 a So ^ a-f 2 0 a a 3 °5 ^ S ^ «5 3 J3 C3 . CJ •0 ^0 -c fc fp •2 " *^ a o2 -^ 03 XI 0 3>. 2 t. OX! 0 2 ^2 S 0 =J -s S3 ^ c £ » S'O 2 >. 0 « OT ^ feS a a "0 9 0 mo S> "3 ? « -^ |g 0, <0 ^03 ^ art "3 0 a •3 0 3 2 i a«"« TJ ^-So-S-a 0 _, c-l -c . Oi •Sl^-g-2 Sl23£ J;'o C"~ 0 5 ^•S "f-S B « c«.:2.t: Sg >> •a f-a^fe-. 3 g W M 0 f-. .2 >. m jD 0 O; a a; ■3 g^^Xl .H-c ^0 I- i2 — ■- 0 0 ■0 3 C g-^ go Sa is > i2 XI 3 S.^^ .. '5Jb 00 CO ■3 . 0 w 03 a « □ a P s 0 a ■sas^a 2^.Hg| >> 0 r 0 c CO Oi S , i-c Ox: '"'X! OP y>o n C ■!-» 0 ■^'3 ■2 tc =" >■ 5f >. 4> a^co ■^■2aS5 w w; CO Cj 03 c £^a£fB£ i2gftgS So CO " a « & " , >^^. J3 t> -^ © ^ •« Tf ,_H o» M g5 s en g eo cree e all ion n dren si Includ e criter O P3 W05 2 to ■* CO ?5 S cc CO •^ ~ o^ J3 -tJ-*-> 1 Q x) -o -t: -i a o o o o o Qi a a »^ o W entifl ion ai perce iden riteri 5 o en id riteri X as hose by c o o o o o T3 . -r) tS CO d S >>-=•-- » o 5x3.:: Ota ffl O o o o o o CO -o 1-1 a OS ^ 5 o o o O o -*-:• CQ 1 (>H t>. (>. o o X2 W o o o o o '■IS a cn 4) "C 2 5 a ll 2 a 1— ( ,— 1 — l-H ,-t — ' M a >. o o •m n o Ui Ol XI a J3 ? a s; 14 ,T3^ a a o i u >• S, 0! 3.:: © c -p c ■3| 1 h o! 'If a.h :_) «> >■ , Clinical estimate of poor or v status at 7 years of age, n and dental care at 7 years isfnctory" ' percentage cl HH a h-4 k>4 > S3 O S3 a o •o a OS a o a a 3 3! a es SI u o o >> O XI a o © xa *j o a 2 3 O .s: O CO to oS-O —t © « " CO © © o to O ©■? x: & o^ WX! O o " O S3 © •3 © © © © © o x: a © XI u © SI a o x: © _-S .2 o'g •?. CC J2 a © © o e© © " Is © a £S a'^ © > Xi CO .X3 a o u a © © x> X] >> ■3 3 B-^ .5 3 XI ~ © ■3 a © X3 3 X5 (^ ■O 3 CO o x: © .S3 o 5 — -2 « =" a c ~ ■- M 3^ c 1; w y c— o 2.5 «^ a - ^; 2 o^ ■'^ i: 2 g 3 =3 o S 1; =~ Ci ; C) I, — c -3 x: •£ -• CO tc 7- i 9^ ^ o ■-* •SEiife3-a S 3 -c — .© ^ Q 73 S £; t; * ? — . c © © © c .^ .V ? ." © c^ E- £: ^ssagr i © a.--"-^ ai — Qa!/-. wcsteJC S©Si 110 ASSESSING THE PHYSICAL FITNESS OF CHILDREN lH >r >> 00 c» -r CO t^ ■o^ ■g rt -J t^ CO (^ »- C» CO ^ o o o as w a: • (-1 3 I-* Tt* -"^ CO -^ tC •2 a |S S ?? S? ?5 o ■Ona ^ ^ ■to S3 •wOO 00 00 »0 00 § CC 00 M* t^ 00* 1 S*^ ?^ o_ o 00 cs CD £■3 a ». > O P5 fin 1^ CD 1-H CD CO »H ^0" T3— . a; S o «o Tt< lo "O •g c«5 0 ■* . OS 00 o oi Id a +j o s § § g-g 3 ?j ^ 2 2 S oi.o a fc- i" O . ?>li^ CO CO CD 05 O O CO >> o ^ ^ d 2 s PQ -< i-H !>. t» O r-l a X 5 ^^ ^ H t^ 00 t^ .-1 '^^-i 1 ■*-» l^ CO •-H "tf «o t>> o O e X! n U ■" •^ s ■a hys '.rce S CD CO CD CO CO '-M -^ "^ "^ -^ Tjl ^a a "jH "o-o S 3 ^ ®* p e^w •^ ^ M o a 2 t^ t> b* b- r>- S^ CO t>. t>. r* t* !>• 1 -ki CO 2 >> o O m " Os o <» a 3 CO s s 0- CO ^-^ ;? a 3 -< s o of eac oys a i o CO >> o §5 2 ■* s-* ffl • 2.« t4 o » Oi o S > « ^ C3 X >s a PQ 1 O b, 2 > o CO p u- GJ Oi «, < El 1 . as to J2 > 0! p, o o ->.a si o '^ a t- is a. a 3 ^ o « It Oco^g « "3 CB t. '2 1- t; - CO S- W S ►J 0-1 •3°' a 3 _" 03 P >> N •72 ^ t-< OT p HH 1—4 t— t I-H >■ HH H-( > 1 X! O ja CO •2 -2 5 g 2 » Xi CO o P 9i o S3 8 03 -- 03 J3-> 03 ti.2-^ C3 _03 03 °0 CO 2 s^ ■n 03 ^ 5 to ^ ^ a o C3 O.l: CO a a a CO '03 03 1-5 -C .*j S-a^' X 03 •'^ ''■Q a a 03 t> Xi X) 3 03 a , -a a c •s^^^sf^ 03 pJrio =2 a £ a =. =0 I ■^ 03 tC 03 O ^ 3 t- i_ r; o -*^ O o +S CO ^ to O ° „ >> CO o CO CO if o >, P ,-_ U. O Ui 3a £pt>«a Sco^osi: •as^^5° ■" a aj3 P !_, 03 -*-a 03 5 03 Ui M^ S ■" O. a ^ 03 C.J3 a co'cS-s'-' « (S CO., & a b- ♦i O P o 03 CJ O tjX; ^^ bJC-C C i; a CO ts 2 CO ^ RTJ-^-- te-5 O CO '^■-- Ui 03 fc- Ci a, " -P C. M a JO 03 w APPENDIX I IJl Supplementary table XIV. — Baldwin-Wood Index for the boys and girls Percent underweight or overweight Boys Girls Total 365 '347 -20.0 to -15.1 .. 1 19 56 86 94 60 27 11 0 0 1 0 0 4 27 50 83 78 -15.0 to -10.1.... .. -10.0 to- 5.1... - 5.0 to- 0.1 0.0 to 4.9 5.0 to 9.9 .. lO.Oto 14.9 24 10 I5.0t0 19.9 20.0 to 24.9.. g 25.0 to 29.9... .. g 30.0 to 34.9 0 4 40.0 to 44.9..-. 1 45.0 to 49.9.. .. 1 50.0 to 54.9 .. 1 0 60.0 to 64.9 1 Mean _ .. .. 1.77 8.70 2 21 11.2 ' 348 girls were included in the study, but the index could not be tested on 1 sirl whose height exceeded the measurements given in the table. Supplementary table XV.- - Nutritional Status Index for Weight of the boys and girls x/a- ' Boys Girls Total »364 »347 -1.6 .. —2.0 to 1 3 18 43 94 95 62 33 10 3 2 0 -1.5 to -1.1 .. 1 -1.0 to -0.6 14 —0.5 to -0.1... .. .. .. -. 1. 38 0.0 to 0.4 91 0.5 to 0.9. .. 95 1.0 to 1.4 56 1.5 to 1.9... ..:::: 35 2.0 to 2.4 . . 10 2.5 to 2.9 0 3.0 to 3.4.. 5 Mean .. . .. .. 0. 652 0.773 0.715 Standard deviation. 0.772 • See Appendix II: (76) Nutritional Status Indices, p. 65. Table X gives the child's index or score according to an arbitrary scale which may be expressed in terms of the abscissa of the normni curve. Table X score = 10j/o-|-50. 2 305 boys and 348 girls were included in the study, but the index could not be tested on I boy whose bitrochanteric width was unknown and on 1 girl whose chest breadth was less than the measurements given in the table. Supplementary table XVI. -Nutritional Status Index for Subcutaneous Tissue of the boys and girls z/o-i Boys Girls Total »364 »347 -1.6 —2.0 to 6 18 80 113 82 45 14 5 1 0 0 2 — 1.5 to -1.1 23 — 1.0 to -0.6 59 —0.5 to -0.1... 104 0.0 to 0.4 81 0.5 to 0.9 54 1 0 to 1.4 15 1.5 to 1.9 7 2.0 to 2.4 0 2 5 to 29 1 3.0 to 3.4 1 -0.106 0 678 -0.0137 Standard deviation... 0. TX) • See footnote 1, supplementary table XV. ' 365 boys and 348 girls were included in the study, but t he iniie\ could not bo felted on 1 boy whose bitrochanteric width was unknown and on 1 girl whose chest breadth was less than the measurements given in the table. 112 ASSESSING THE PHYSICAL FITNESS OF CHILDREN Supplementary table XVII. — Nutritional Status Index for Arm Girth of the boys and girls xlody type in young children, Am. J. Phys. Anthropol. 13 : 287-307, July-Sept. '29 (46) HERD, H., Discussion of paper, Assessment of nutrition, by J. G. Woolham, Pub. Health 50 : 258-260, May '37 (47) [Editorial] Lidexes of nutrition. J. A. M. A. 81 : 1286-1287, Oct. 13, '23. (48) JACOB, R., Heights and weights in a girls' public school. (Letter to the editor) Nature 142: 436-437, Sept. 3, '38 (49) JENKINS, R. L., On a distortion embodied in the Pryor-Stolz width-weight tables, M. Rec. 146: 127-128, Aug. 4, '37 (50) JONES, R. H., The Nutrition of Liverpool Schoolboys in Relation to Certain Physical Measurements. Pp. 172. Submitted for the degree of M. A. in the University of Liverpool, May 1937 (51) Physical indices and clinical assessment of the nutrition of school children, J. Roy. Statist. Soc. 101 : 1-52, Part L '38 (52) KOCH, E. W. : Sec reference contained in (129) Wroczynski (5^a;KRUSE, H. D., PALMER, C. E., SCHMIDT, W., and WIEHL, D. G., Medical evaluation of nutritional status; methods used in a survey of high school students, Milbank Memorial Fund Quart. 18: 257-29S, July '40. (53) LEIGHTON, G., and CLARK, M. L. : See reference contained in (44) Harris, p. 262, Dec. 18, '37 (54) LEIPOLDT, C. L., A. C. H. Index as applied to South African boys: See reference contained in (99) Scantlebury (55) LE RICHE, H., k. C. H. Index as applied to boys and girls in selected Pretoria schools. South African J. M. Sc. 3: 60-62, April '38 (56) LINCOLN, E. A., Reliability of anthropometric measurements. Pedagogical Seminary and J. Genetic Psychol. 38: 445-450, Dec. '30 (57) LISHMAN. F. J. G., A scheme of clinical assessment of nutrition and a survey of the nutritional state of elementary school children in Gloucestershire, 1938, in Annual Report of the Gloucestershire [England] School Medical Officer (Dr. K. Cowan) for 1938. Pp. 22. London and Southend: W. TL Houldershaw, Ltd. (58) How do we assess nutrition? Correspondence to the editor, Pub. Health 51: 341, Aug. '38 (59) LOEWENTHAL, L. J. A.. Effect of addiiioii of milk to diet of school boys in Huganda, East African M. J. 15: 35 If), May '3S (60) LUCAS. W. P., PRYOR, H. B., BOST, C, POPE, S. T.. and HENDERSON. M. C. Growth problems and their relation to zones of normal physiologic reactions, J. Pediat. 1 : 572-592, Nov. '32 118 ASSESSING THE PHYSICAL FITNESS OF CHILDREN (61) McCARRISON, R., Nutrition and national health, J. Roy. Soc. Arts 84: 1047-1107, Aug. 28, '36 (62) McPHERSON, C. A., The A. C. H. Index and other assessments of the nutri- tional state contained w Leicester [England] EHucation Committee: Annual Report of the School Medical Officer for 1938. Pp. 9 (63) MAGEE. H. E. : See reference contained in (44) Harris, pp. 225-229, Nov. 27, '37 (64) MANCHESTER [ENGLAND] EDUCATION COMMITTEE, Annual Report of the School Medical Officer (Dr. Henry Herd) for 1936. Pp. 96. Man- chester: Education Offices, 1937 (65) MARSHALL, E. L., Comparison of four current methods of estimating physi- cal status, Child Development 8 : 89-92, Mar. '37 (66) Objectivity of anthropometric measurements taken on eight- and nine-year-old white males, Child Development 8: 249-256, Sept. '37 (67) MEREDITH, H. V., Reliability of anthropometric measurements taken on eight- and nine-year-old white males. Child Development 7: 262-272, Dec. '36 (68) Stature of Massachusetts children of North European and Italian ancestry. Am. J. Phys. Anthropol. 24: 301-346, Jan.-Mar. '39 (69) MILLIGAN, E. H. M., Chief Medical Officer, Borough of Glossop, England, Personal communication, dated Jan. 27, '38 (70) MITCHELL, H. H., Nutrition, Section II in Porto Rico; inquiry into the health nutritional and social conditions in Porto Rico as they may affect children, by Committee of American Child Health Assn. Staff. Pp. 170. New York: American Child Health Assn., 1930 (71) Physical measurement and nutritional status, J. Pediat. 6:316-321, Mar. '35 (72) Study of factors associated with growth and nutrition of Porto Rican children, Human Biol. 4: 469-508, Dec. '32 (73) [Author unlisted] Nutrition of school children, Committee Against Malnu- trition, Bulletin No. 9 (July 1935), p. 28 (74) [Editorial] Nutrition of school children. M. Officer 59: 64-65, Feb. 12, '38 (76) [Author unlisted] Nutritional Status Indices; method of obtaining measures of musculature, su})cutaneous tissue, and weight with allowance for skeletal build. (Boys and Girls, 7 to 12 years of age.) Pp. 66. New York: American Child Health Assn., 1935 (76) ORR, J. B., Influence of amount of milk consumption on rate of growth of school children: See reference contained in (44) Harris, p. 262, Dec. 18, '37 (76a) Physiological and economic bases of nutrition; assessment of the state of nutrition, J. Roy. Inst. Pub. Health and Hyg. 3: 9-24, Jan. '40 (77) PALMER, C. E., Further studies on growth and the economic depression; comparison of weight and weight increments of elementary school children in 1921-27 and in 1933-34, Pub. Health Rep. 49: 1453-1469, Dec. 7, '34 (78) Nutrition as a public health program. The Registered Nurse, pp. 3-7, issue of July-Aug. '38 (79) Relation between size of frontal nasal sinuses and attacks of upper respiratory disease (common cold) and certain nose and throat conditions, Am. J. Hyg. 16: 224-232, July '32 (80) Seasonal variation of average growth in weight of elementary school children. Pub. Health Rep. 48: 211-233, Mar. 3, '33 (81) Passed Asst. Surgeon, National Institute of Health, U. S. Public Health Service, Personal communication, dated Oct. 7, '39 APPENDIX II 119 (82) PATON, D. N., FINDLAY, L., and others, Poverty, Nutrition and Growth; studies of child life in cities and rural districts of Scotland. Medical Re- search Council, Special Report Series, No. 101. Pp. 333. London: His Majesty's Stat. Off., 1926 (83) [Author unlisted] Physical Defects; the pathway to correction. Pjj. xvii + 171. New York: American Child Health Assn., 1934 (84) PINCKNEY, J. M., Director, Office of Nutrition and Health Education Bu- reau, Univ. of Te.xas, Austin, Texas, Personal cominunication, dated Mar. 3, '38 (85) PIRQUET, C, An outline of the Pirquet System of Nutrition: See reference contained in {89) Roberts (^6') PRYOR, H. B., Width- Weiglit Tables for Boys and Girls from 1 to 16 years; for men and women from 17 to 24 years. Pp. 15. Stanford University, Calif.: Stanford Univ. Press, 1936 {87) and SMITH, R. T., Physical strength of adolescent girls, J. Pediat. 14: 610-617, May '39 {88) and STOLZ, H. R., Determining appropriate weight for bodybuild, J. Pediat. 3: 608-622, Oct. '33 (89) ROBERTS, L. J., Nutrition Work with Children, Second edition. Pp. xix-1- 639. Chicago: Univ. of Chicago Press, 1935 {DO) — —BLAIR, R., LENNING, B., and SCOTT, M., Effect of milk supplement on physical status of institutional children; growth in height and in weight, Am. J. Dis. Child. 56: 287-300, Aug. '38 {91) — STONE, D. M., and BOWLER, V. P., Investigation into nutrition of children of school age, M. Officer 60: 209-212, Nov. 12, '38; 219-220, Nov. 19, '38 {92) ROGERS, F. R., Physical Capacity Tests in the -Administration of Physical Education. Dissertation submitted in partial fulfillment of the require- ments for doctor of philosophy at Columbia Univ., 1925 {93) ROHRER, F., Eine neue Formel zur Bestimmung der Korperfiille: See reference contained in {89) Roberts {94) ROSE, M. S., The Foundation of Nutrition. Third edition. Pp. xi + 625. New York: Macmillan, 1938 {95) ROSENOW, C, Weight and skeletal build. Am. J. Orthopsychiat. 3: 55-64, Jan. '33 {96) Weight and skeletal build; a rejoinder. Am. J. Orthopsychiat. 4: 258-261, April '34 {97) ROWLAND, S. R. : See reference contained in (55) Great Britain Board of Education, p. 23 {98) ST. PAUL [MINN. I DEPARTMENT OF EDUCATION. Bureau of Schools, Report of the Director of Hygiene (Dr. G. W. Snyder) [to the Commis- sioner of Education and Superintendent of Schools], 1937-38. Pp. 150. 1938 {99) SCANTLEBURY, V., Nutrition of preschool child, M. .1. Australia 2: 1 IS 155, July 30, '38 (iOO)SCHARFF, J. W., and SINNADORAI, S., Milk for the school child: See refer- ence containefl in {44) Hurri.s, pp. 261-265, Dec. 18, '37 (iO/)SCHLUTZ, F. W., Dept. of Pediatrics, Univ. of Chicago, Physical Fitness in the Child as Reflected by Cardio-Vascular and Blood Serum Changes (luring Exercise. Personal communication, dated May 12, '39 (iO^)SCHMIDT, W., Newer medical methods of appraisal of nutritional status. Am. J. Pub. Health 30: 165-168, Feb. '40 120 ASSESSING THE PHYSICAL FITXESS OF CHILDREN (103) SHERMAN, H. C, Chemistry of Food and Nutrition. Fifth edition. Pp. vii + 640. New York: Macmillan, 1937 (104) an<^ HAWLEY, E., Calcium and phosphorus metabolism in childhood, J. Biol. Chcm. 53: 375-399, Aug. '22 (105) SIMPSON, T. W., and WOOD, E. C, Value of certain supplements to diet of children, M. Officer 53: 125-127, Mar. 30, '35; 135-138, April 6, '35 (106) SOUTHER, S. P., ELIOT, M. M., and JENSS, R. M., Comparison of indices used in judging physical fitness of school children, Am. J. Pub. Health 29 : 434-438, May '39 (107) [Editorial] State of nutrition. Lancet 2: 1257-1258, Nov. 27, '37 (108) STEGGERDA, M., and DENSEN, P., Height, weight, and age tables for homogeneous groups— with particular reference to Navaho Indians and Dutch whites. Child Development 7: 115-120, June '36 (109) STIEBELING, H. K., and CLARK, F., Planning for good nutrition in U. S. Dept. of Agriculture, Yeairbook of Agriculture, 1939; food and life. Pp. 321-340. Washington: Government Printing Office, 1939 (110) STOCKWELL, G. E. S. : See reference contained in (39) Great Britain Board of Education, p. 23 (111) STUART, H. C, Standards of physical development for reference in clinical appraisement; suggestions for their presentation and use, J. Pediat. 5: 194-209, Aug. '34 (112) TURNER, C. E., Precision and reliability of underweight measurement, Am. J. Pub. Health 19: 969-979, Sept. '29 (113) Weighing and measuring school children, J. Outdoor Life 28: 377- 381, June '31 (114) TUXFORD, A. W., Estimation of physique and nutrition in children, J. Hyg. 39: 203-216, Mar. '39 (115) U. S. CHILDREN'S BUREAU, Publication No. 217, The Effect of Tropical Sunlight on the Development of Bones of Children in Puerto Rico, by M. M. Eliot. Pp. v + 122. Washington: Government Printing Office, 1933 (116) WALLSEND [ENGLAND] EDUCATION COMMITTEE, Annual Report of the School Medical Officer (Dr. R. Rutherford) for 1937. Wallsend: 1938 (117) WARNER, E. F., Use of nutritional status indices developed by the Ameri- can Child Health Association in Principles and Practices in School Health Education, Health Education Conference at Iowa City, June 1935. Pp. 16-30. New York: American Child Health Assn., 1935 (118) WARRINGTON [ENGLAND] COUNTY BOROUGH OF WARRINGTON, Annual Report to the Education Authority on School Hygiene for the Year 1935 made by G. W. N. Joseph. Pp. 84. Warrington: Mackie & Co., Ltd., The Guardian Press, 1936 (119) WHITE HOUSE CONFERENCE ON CHILD HEALTH AND PROTECTION, Growth and Development of the Child; Part IV, Appraisement of the Child. Pp. xix + 344. New York: The Century Co., 1932 (120) WILKINS, E. H., Assessing nutrition of school children, Lancet 2: 1265- 1267, Nov. 27, '37 (121) Assessment of nutrition in school children, M. Officer 58: 145- 148, Oct. 2, '37 (122) WILSON, H. E. C, AHMAD. B., and MITRA. D. D., Nutritional survey of schoolboys in Calcutta and Punjab by means of clinical observations, A. C. H. Index of nutrition and other measurements, Indian J. M. Research 24: 817-837, Jan. '37 (123) and MITRA, D. D., A diet and physique survey in Assam, rural Bengal, and Calcutta, Indian J. M. Research 26: 131-154, July '38 APPENDIX II 121 (124) WOLFF, Die Nachwirkung der Kriegshungerperiode auf das Schulkinder- wachstuni: See reference contained in {129) Wroczyn?ki {125) WOODBURY, R. M., Tables for infancy and early childhood. Mother and Child 4: supplement to July '23 issue {12fi) WOOLHAM, J. G., Assessment of nutrition, Pub. Health 50: 258-260, May '37 {127) and HONEYBURNE, W. R., Fitness of Schoolboys; correlations between the results from pulse and respiratory tests, in Manchester [England] Education Committee: Animal Report of the School Medical Officer (Dr. A. B. Ritchie) for 1928. Pp. 15 {128) and SPARROW, L. W., Physical Fitness of Schoolboys in Man- chester [England] Education Committee; Annual Report of the School Medical Officer (Dr. Henry Herd) for 1935. Pp. 55-68 {129) WROCZYNSKI, C, Physique and health, Bull. Health Organ. League of Nations 6: 551-682, Aug. '37 {ISO) ZAYAZ, S. L., MACK, P. B., SPRAGUE, P. K., and BAUMAN, A. W.. Nutri- tional status of school children in small industrial city, Child Develop- ment 11: 1-25, March '40 o BOSTON PUBLIC LIBRARY 3 9999 05708 5852