Food choices of five-year-old white female preschool children

The prevalence of obesity is increasing in today's preschool children, particularly in white, female children from families of the upper and upper-middle socioeconomic classes. Because obesity in preschool children in often difficult to identify, the prevalence of obesity in preschool children has gone unnoticed and thus, very few indepth studies have been conducted in this age group. This study was designed to investigate the contribution of visual food cues and the calorie content of food choices toward the development of childhood obesity. These factors were investigated in 102 five-year-old, white females from families of upper-middle and upper socioeconomic classes. The girls were taken from their preschool class, one at a time and asked to visually inspect the selection of food replicas and then to choose those "foods" they would like to eat for dinner. Height, weight, and tricep and subscapular skinfold thickness was obtained from each child following the food choices. The results of correlating weight and mean skinfold thickness measurements to the variables, total number foods chosen, the total calories contained in the food chosen, and the number of foods chosen from the foods classified into high and low calorie food grouping were significant. When body weigh, height, and income variables were controlled, the correlation between the numbers of foods, chosen, the total calories contained in the foods chosen, and the mean skinfold thickness measurements remained significant (? .01 significance level). The statistical analysis indicated large standard deviation about the means which signified a high variability of the dta. Despite the limitations produced by the data, the results indicated a notable significance probability of less than .01. On the basis of these findings, it became evident that five-year-old, white, female children with greater body weights and particularly greater body fat-as signified by mean skinfold thickness measurement-can be expected to choose a greater number of food items with higher calorie content than those five-year-old, white, female children of lesser body weight and body fat (? .01 significance level). This investigation offered several avenues for expansion as well as possibilities for research concerning factors related to childhood obesity. Results indicate the potential significant contribution of nurses involved in the care or obese children and their families when the effect of visual food cues and the calories contained in available food are incorporated into patient and family teaching health care planning, and weight control.

FOOD CHOICES OF FIVE-YEAR-OLD WHITE FEMALE PRESCHOOL CHILDREN by Nancy C. Young A thesis submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Master of Science College of Nursing The University of Utah June 1979 Copyright © Nancy C. Young 1979 All Rights Reserved THE UNIVERSITY OF UTAH GRADUATE SCHOOL SUPERVISORY COMMITTEE APPROVAL of a thesis submitted by Nancy C. Young I have read this thesis and have found it to be of satisfactory quality for a master's degree. April 18, 1979 Dale / A?!d. C:-" L. Sue Huether / ku:: zXi;o / Chairman. Supervisory Commitlee I have read t his thesis and have found i Ito be of satisfactory �)ity for a master's degree. April 18, 1979 Date / .I J ___ . � / ����������Marvin L. Rallison Member. Supervisory Committee I have read this thesis and have found it to be of satisfactory quality for a master's degree. April Date 18, 1979 Celia Woodcock Member. Superviso�y Committee THE UNIVERSITY OF UTAH GRADUATE SCHOOL FINAL READING APPROVAL To the Graduate Council of The University of Utah: Nancy C. Young I have read the thesis of in its final form and have found that (I) its format, citations. and bibliographic style are (2) its illustrative materials including figures. tables. and (3) the final manuscript is satisfactory to the Supervisory consistent and acceptable; charts are in place; and Committee and is ready for submission to the Graduate School. April Dale 26, 1979 Sue Huether Member, Supervisory Commillce Approved for the Major Departme.nt / Approved for the Graduate CLluncil ABSTRACT The prevalence of obesity is increasing in today's preschool children, particularly in white, female children from families of the upper and upper-middle socioeconomic classes. Because obesity in preschool children is often difficult to identify, the prevalence of obesity in preschool children has gone unnoticed and thus, very few indepth studies have been conducted in this age group. This study was designed to investigate the contribution of visual food cues and the calorie content of food choices toward the development of childhood obesity. 102 five-year~old, These factors were investigated ;n white females from families of upper-middle and upper socioeconomic classes. The girls were taken from their pre- school class, one at a time, and asked to visually inspect the selection of food replicas and then to choose those IIfoods to eat for dinner. ll they would like Height, weight, and tricep and subscapular skin- fold thicknesses were obtained from each child following the food choices. The results of correlating weight and mean skinfold thick- ness measurements to the variables, total number of foods chosen, the total calories contained in the foods chosen, and the number of foods chosen from the foods classified into high and low calorie food groupings were significant. When body weight, height, and income variables were controlled, the correlation between the number of foods chosen, the total calories contained in the foods chosen, and the mean skinfold thickness measurements remained significant (~ .01 significance level). The statistical analysis indicated large standard deviations about the means which signified a high variability of the data. Despite the limitations produced by the data, the results indicated a notable significance probability of less than .01. On the basis of these findings, it became evident that five-year-old, white, female children with greater body weights and particularly greater body fat--as signified by mean skinfold thickness measurements--can be expected to choose a greater number of food items with higher total calorie content than those five-year-old, white, female children of lesser body weight and body fat (~ .01 significance level). This investigation offered several avenues for expansion as well as possibilities for research concerning factors related to childhood obesity. Results indicate the potential significant contri- bution of nurses involved in the care of obese children and their families when the effect of visual food cues and the ca10ries contained in available foods are incorporated into patient and family teaching health care planning, and weight control. v CONTENTS Page ABSTRACT . . . iv LIST OF TABLES ix ACKNOWLEDGr1ENTS x CHAPTER I II III IV INTRODUCTION Purpose . . Problem . . . Objecti ves . . 4 REVIEW OF LITERATURE 5 3 3 Measurement of Obesity . • . . . . External Controls of Food Intake. Social Stimuli . . . . . . . . . Taste Properties . . . . . . . . Sight and Availability of Food .• Calorie Intake and Food Choices 10 CONCEPTUAL FRAMEWORK . 25 Hypotheses . . . . . Hypothes; s 1 . . Hypothesis 2 •. 26 26 Definitions of Terms. Body Wei ght . . Body Fat . . . . Food Items . • . Total Calories. 27 27 27 27 27 ~·1ETHODOLOGY Aim AND RESEARCH DESIGN • • • . • • • • • • • • • • • Population . . . . . . . . . . . Sample Selection and Exclusions Criteria for Sample Selection . 14 15 16 17 20 26 28 28 28 28 29 . ... Design . . . . . Ins truments . . Scale . . . Tape Measure . . . Right Angle Block Sk info 1d Ca 1-i pers . Artificial Food Items f4easurements . . . . . . . Wei ght . . . . . . . . . Height . . . . . . . . Skinfo1d Measurement . . Procedure Sequence . . . Statistical Procedures. V 30 30 30 31 31 31 32 33 33 34 34 36 39 ANALYSIS OF DATA. 42 Hypo theses . . . . Hypothesis 1 . . . . Hypothesis 2 . . • . • . . . Statistical Examination of Data Descriptive Statistics . Simple Correlations Partial Correlations. 42 42 43 43 43 44 48 DISCUSSION OF FINDINGS 52 VII LH4ITATIONS 59 VI I I SUr·1t·1ARY AND VI ..... . RECOM~·'ENDATIONS 61 APPENDICES INFORt'lED CONSENT FOR HUMAN RESEARCH PROJECT, THE UNIVERSITY OF UTAH COLLEGE OF NURSING GRADUATE PROGRAMS . . . • . . . 65 B MEDICAL QUESTIONNAIRE • . . . . . . . . . . 67 C RANDOMLY SELECTED POSITIONS OF FOOD REPLICAS. 69 D WEIGHT 71 E HEIGHT MEASURH1ENT PROCEDURE F TRICEP SKINFOLD G SUBSCAPULAR SKINFOLD MEASURH1ENT PROCEDURE . 78 H DATA COLLECTION FORr·, . . . . . . . . 80 I ARTIFICIAL FOOD LISTING AND CALORIES 82 A r~EASUREr~ENT PROCEDURE . ~'EASUREMENT vii 73 PROCEDURE 75 REFERENCES 84 VITA 94 viii LIST OF TABLES Page Table Descri9tive Statistics . 43 2 Correlation Coefficients 45 3 Significance Levels of Correlation Coefficients 47 4 Partial Correlations 49 5 Significance Levels of Partial Correlations 50 ACKNOWLEDGMENTS I wish to express special thanks to those who have aided me in completing this study: Elizabeth Close, RN, MS; Sue Huether, RN, MS; Celia Woodcock, RN, MS; Marvin L. Rallison, MD. I also want to thank those whose continual support and motivation was greatly appreciated: June Abbey, RN, Ph.D.; Nancy Droubay, RN, MS; Tom Storer, MS; my husband, Alan; and my fellow classmates. Appreciation is extended to the principals and teachers of six elementary schools in the Salt Lake School District for their assistance and interest in this study, and to Irene Griffiths, Nutritionist, and Carol Cembrola, Registered Nurse, in the Salt Lake School District, for their coordinating efforts with the district. CHAPTER I INTRODUCTION The prevalence of obesity ;s increasing in today's preschoo] children (Ginsberg-Fellner & Carmel, 1974; Garn &Clark, 1976; Raugh, Schumsky &Witt, 1967; Ten State Nutrition Survey, 1973), particularly in preschool female children (Johnson, Burkey & ~1ayer, 1956; Ten State Nutrition Survey, 1973). Because obesity in preschool children is often difficult to identify, the incidence of obesity in preschool children has previously gone unnoticed and thus, very few indepth studies have been conducted in this age group. When obesity develops during the preschool years, many children suffer unfortunate associated complications such as flat feet, clumsiness, shortness of breath on exertion, heat discomfort, decreased ventilatory capacity, and/or certain orthopedic problems (Jones, 1972; Mobbs, 1970). In addition, many obese children experience serious psychological difficulties which perhaps produce the most selfdevasting effects (Weil, 1977). These difficulties include: poor self-image, feelings of inferiority and rejection, and teas"ing and ridicule from other children (Brusch, 1975; Knittle, 1972; Lerner & Gelert, 1960; Nathan, 1973; Stunkard & Burt, 1967; Stunkard &Mendleson, 1967). As a substitute for acceptance and activity. the obese child resorts to eating for gratificaiton (Brusch, 1975; Jacobsen, 1954; ~'end1eson, 1964). Mayer (1972) revealed that when obesity 2 continues into adolescence, it leads to attitude changes and forms of neuroses quite different from those observed in persons of normal weight or in persons who become obese during adulthood. These psy- chological problems are often continued into adulthood (Stunkard & Burt, 1967) as a significant number of obese children become obese adults (Abraham &Norsieck, 1960; Charney, Goodman, McBride, Lyon & Pratt, 1976; Knittle, 1972; Lloyd, Wolff & Whelen, 1961; Stunkard & McLaren-Hume, 1959; Wolff, 1955; Wei1, 1977; Young, Moore & Richard, 1955). When childhood obesity continues into adulthood, the obesity becomes much more resistant to treatment (Penick &Stunkard, 1970) and the number and severity of associated complications vastly increase (Hashim, 1977; Woodhouse, 1976), thus compounding the difficulties of the obese individual. Because of the problems frequently experienced by obese children and the compounded difficulties identified when obesity continues into adulthood, it appears urgent that obesity be detected and curtailed early in life. Knittle (1972) supports the necessity for studying preschool obesity by stating that, liThe critical period of development with major consequences regarding one's adult weight occurs between birth and age five" (p. 1049). This statement suggests an excellent time to identify, prevent, or arrest the development of an obesity problem (Wolff, 1955). The factors related to the development of obesity in childhood need to be examined for more adequate management and detection. In order to prevent obesity from developing in the young population, factors contributing to obesity must be identified through scientific investigation. Two factors considered 3 to have significant influence in contributing to the state of childhood obesity are the visual stimulation produced by food cues (Johnson, 1970; Nisbett, 1968, 1972; Nisbett & Kanouse, 1970; Schaeter, 1968, 1971) and the consumption of a calorie surplus of food (American Academy of Pediatrics Committee on Nutrition, 1967; "Management of the fat child," 1966; Gates, Huenemann & Brand, 1975; Hashim, 1977; Hinton, Eppright, Chadderdon &Wollins, 1963; Stunkard, 1975; Weil, 1977) . Purpose This study was designed to examine visual food cues and food calorie content in the food preferences of five-year-old, white, females. The similarities and differences in food choices or prefer- ences of five-year-old, white, females of varying body weight and body fat are reported in this study. Specifically, this investigation re- ported the number of food items chosen and the total calorie content of the chosen foods. Problem Factors precipitating or contributing to obesity in preschool children have not been clearly defined. Even though the visual stimu- lation produced by the food cues and food calorie content have been identified as two contributing elements of obesity, there are no studies which involve examination of possible effects these variables may have on the selection of food and the total calories contained in these selections by preschool children. Furthermore, previous studies have not demonstrated differences in the number of food items and the total calorie content of foods chosen by five-year-old, white, females 4 of greater body weight and body fat from those of lesser body weight and body fat. If the caloric content of food preferences and the effect of visual food stimuli could be shown to contribute to obesity in preschool children, counseling of parents might lead to a decrease in the frequency of preschool obesity and curtail life long obesity patterns. Objectives The specific objectives of this study include the following: (1) To describe characteristics of five-year-old, white, females in terms of weight, height, and mean tricep and mean subscapular skinfold thickness measurement; (2) to obtain data in order to assist others in developing clinical guidelines for the identification of childhood obesity; (3) to identify the effects that visual food cues alone have on the number of foods chosen by five-year-01d females of varying weight and skinfold thickness; (4) to discover the effect that visual food cues alone have on the calorie content of foods chosen by fiveyear-old females of varying weight and skinfold thickness; (5) to demonstrate possible correlations between the total calorie content of food choices and weight, height, and mean skinfold thickness measurements of five-year-old females. ) CHAPTER II REVI EW OF LITERATURE Unrecognized prior to 1968 as a significant problem in the American population, childhood obesity has recently achieved acknowledgment as a prevalent problem which necessitates investigation and study. Previous emphasis was placed on the high incidence and treat- ment of adult obesity. Presently, the focus of research involves pre- vention and recognition of obesity in the early phases of childhood development. Several studies have been conducted which verify the prevalence of obesity in the United States preschool population. Results frool New York studies based on a weight for age definition of obesity (~ 95 percentile) revealed an incidence of obesity greater than 15 percent among 737 youngsters ages three to five (Ginsberg-Fellner & Carmel, 1974). Other investigations have produced similar results, although a variety of definitions have been utilized in designating obesity in children (Cannings & Mayer, 1966; Garn & Clark, 1976; Johnson et al., 1956; Lloyd et al., 1961; Neumann, 1977; Ten State Nutrition Survey, 1973). The increased incidence of preschool obesity in the United States among the upper-middle and upper-socioeconomic classes re- 'l J mains a questionable issue. Although it has been demonstrated that the quantity and quality of food ;s directly related to family income 6 whereby more children of the affluent become obese (Abraham &Carroll, 1977; American Academy of Pediatrics, 1973; Garn, 1976; GinsbergFellner &Carmel, 1974; National Center for Health Statistics, 1966; Owen, Kram, Garry, Lowe &Lubin, 1974; Walker & Richardson, 1971), many investigators have demonstrated socioeconomic status to be inversely proportional to obesity in children {Goldbloom, 1975; Mayer, 1972; Stunkard, d ' Aqui1i, Fox & Filion, 1972}. Based on the Ten State Nutrition Survey, Garn and Clark (1976) reported that until adolescence, children at all ages from a higher socioeconomic status were consistently fatter than those from lower income levels and that most of the prepubertal fatness occurred early in children of higher income levels. In addition, Weil (1977) discovered that the higher the socioeconomic status of the chi1d--both male and female--the greater was the amount of adiposity. Weil (1977}--also in agreement with findings of the Ten State Nutrition Survey (1973}--concluded that, for comparable income groups, white children were significantly fatter than black children. Among white children of affluent backgrounds, more females than males exhibit obesity (Abraham, Collins & Norsieck, 1971; Canning &Mayer, 1966; Hewitt, 1974; Johnson et al., 1956; Mayer, 1963; Weil, 1977; Wetzel, 1941). Results from the Leicester Nutri- tion Clinic Study in 1965 reported by Mobbs (1970) supported the higher prevalence of obesity in females. Mobbs' sample population of obese, school-age children reportedly contained twice the number of girls than boys. Evidence thereby substantiates a high prevalence of obesity among affluent, white, female children. 7 Although the evidence is neither comprehensive nor conclusive, the available studies strongly suggest that obesity tends to persist from childhood into adolescence and young adult life (Abraham & Norsieck, 1960; Abraham et al., 1971; Charney et al., 1976; Miller, Billewicz &Thomson, 1972; Mullins, 1958; & U.S. Public Health Service, 1966). The most authoritative estimate denotes 80 to 86 percent of obese children follow this course (Abraham &Norsieck, 1960; Knittle, 1972; Lloyd et al., 1961). Additionally, Abraham, et al. (1971) noted that four out of five of the obese female adu lts they stud i ed had been fobese as children~Furthermore, the odds against an obese child be- coming a normal weight adu1t--which are more than four to one at age l2--rise to 28 to one if weight has not been reduced by the end of adolescence (Stunkard &Burt, 1967). Mullins (1958) reported that 101 of 373 consecutive patients seen in an outpatient department of a London hospital were more than 20 percent above standard weight; of these, approximately one-third had been overweight since childhood. Thus, childhood obesity must be seriously considered a precursor of adu It obes i ty . \ C) According to Wolff (1955) significant correlations between weight in childhood and subsequent weight in adulthood do not occur until approximately age five. Knittle (1972) supports Wolff (1955) in suggesting age five to be an appropriate age to detect weight problems which could lead to difficulties in adulthood. Stuart and Sobel (1946) noted that between 5 and 11 years of age the amount of adipose tissue remained unchanged in females. Therefore, if obesity is highly prevalent among affluent, white, female children and tends 8 to progress into adulthood, and if age five can be considered an excellent time to identify obesity, then five-year-old, white, females from affluent families are appropriate subjects to involve in the investigation and detection of childhood obesity. It is imperative that obesity be detected and curtailed early in life in order to prevent problems frequently experienced by obese children and the compounded difficulties encountered when obesity continues into adulthood. In addition to important physiological ef- fects of obesity on the child and adult, the psychological aspects must also be considered significant. For example, Stunkard and Burt (1967) observed that three of a group of 10 obese children who reduced to normal weight during adolescence and maintained their weight loss reported the persistence of a body image disturbance 20 years later. Although the exact extent of the physiological hazards of adult obesity remains in question, the association of obesity with impaired health exists (Charney et a1., 1976; Kannel, Brand & Skinner, ~ 1967; Woodhouse, 1976). Hashim (1977) summarized the findings of \ several investigators in the declaration that when an individual is already obese, certain metabolic disorders may appear such as hyperlipidemia, diabetus mellitus, hyperinsulinism, and hypertension. In addition,Woodhouse (1976) concluded that obesity increases mortality rates associated with such disease as hepatic disease, respiratory disease, arteriosclerotic heart disease, and albuminuria. However, these disorders have been shown to decrease in severity or disappear after a weight loss has occurred (Hashim, 1977; Woodhouse, 1976). summary, if obesity is to be decreased in the American population, In 9 it appears that the time to intervene is childhood. Therefore, fac- tors related to the development of obesity in childhood need to be explored. Little attention has been given to distinguishing factors which generate obesity from those which perpetuate it. The specific etiology of obesity remains an uncertain and controversial issue. A great number of contributory mechanisms have been demonstrated in studies regarding the origin of childhood obesity: Endocrine dis- orders, hypothalamic lesions (Jones, 1972; Mobbs, 1970); decreased physical activity, physiological processes, genetic defects (Jones, 1972; Mobbs, Wei1, 1977); glucose-fat metabolism (Knittle, 1972; Jones, 1972); and many others. It seems probable that all these factors play a significant role and that obesity is truly a multifactorial syndrome (Gates et al., 1975; Penick &Stunkard, 1970; Wei1, 1977). It is well known that obesity can only occur when there is a~( greater calorie consumption than calorie expenditure (Gates, et a1. 1975; Owen et al., 1974). Although most authors in the pre- sent literature have cited excessive calorie intake as one of the principal causes of obesity at every age (Weil, 1977), there exists a discrepancy in findings regarding the food intake and calorie consumption of obese and nonobese school-age children (Jones, 1972; Mayer, 1975 &Wei1, 1977). Few studies concerning food intake and calorie consumption of preschool children have been documented in the literature. Of these studies, no investigation specifically examined the number of foods chosen by preschool children and the total calorie content of foods chosen as related to body fat and body weight. , 10 The variables, the number of food items chosen and the total calorie content of food chosen, need to be explored. More information regarding these variables as they relate to obesity in children is needed to assist the family nurse clinician in his/her educational and counseling efforts in order to prevent the occurrence of childhood obesity. In addition, further exploration of these variables could aid the family nurse clinician in weighing contributions of each etiologic factor in order to be more effective in the health care of the obese child and his/her family. Nursing involvement and research in areas concerning the prevention and detection of childhood obesity as well as counseling and educating the obese child and his/her family could have important effects upon decreasing the prevalence of obesity in the young. Minimal sources exists in the available literature related to the contributing factors and pathogenesis of obesity in man. Further- more, no studies have been reported of the related factors or etiology of obesity in the nursing literature. The literature review emphasis for this study will be placed on investigations involving the effect of two contributing factors of obesity, visual sight of food and food intake. The following topics regarding these two factors contributing to obesity (visual sight of food and food intake) will be addressed in this review measurement of obesity, external controls of food intake, calorie consumption and food choices of preschool children. Measurement of Obesity Confusion has developed widely through the years regarding the definition of obesity. It is well recognized today that the terms 11 "obesityll and "overweighttl were not clearly differentiated or defined by many investigators reporting in previous literature (Cannings & Mayer, 1966; Garn & Clark, 1976; Johnson et al., 1956; Lloyd et al., 1961; Neumann, 1977). Thus, the value and application of the findings recorded in these sources deserve close evaluation. Since body weight has been the most commonly used measure to reflect or define obesity, it appears appropriate to indicate the limitations and problems inherent in this approach. The obesity studies in the past have been primarily concerned with the over\veight rather than the obese in that they defined obesity in terms of excess weight for a given height rather than in terms of body fat. A child who is above average weight for his/her height may be a fat child, or he/she may be a muscular child. Conversely, a child who is underweight for his/her height may be simply lean or may be lacking in muscle or in bone mineralization. In 1977, Weil exempli- fied agreement with Garn (1976) that utilizing weight only as a determination of obesity below six or seven years of age is often misleading since the results most frequently offer an underestimation of fatness. According to Weil (1977) and Woolsey (1973), evaluation by weight can be useful to the clinician when observing for major deviations from a standard pattern, but the inclusion of independent techniques for measuring skeletal and body fat components is necessary. Knittle (1972) recognized the necessity for a definition~ which would clearly distinguish the terms "obese" from "overweight. IIi, He defined obesity as, "An excessive deposition and storage of body fat which clearly separates it from overweight, having no direct 12 implication of fatness" (r. 1050). In specifically defining pediatric obesity, Neumann (1977) and Wolff and Lloyd (1973) recently summarized the majority of pediatric obesity studies of school-age children "in terms of weight/height/age/sex' exceeding the "standard ll by 20 percent. Neumann critically noted deficits of several studies which defined obesity utilizing only height/weight/age/sex percentile charts for defining the standard normals. He concluded that utilizing only the percentile charts for defining obesity variables does not differentiate the childts weight problem as being one due to increased fat deposition from one due to both increased lean body mass and increased fat deposition. It, therefore, appears essential that studies of the obese state must include weight, height, and some estimate of total body fat in order to appropriately define obesity values (Committee on Nutrition Anthropology, 1956; Smith tL Brown, 1970). Although total body fat can be determined directly by the use of inert gases or indirectly by the measurement of body density, total body water, or total body potassium, these techniques are not readily available to the family nurse clinician (Brozek, Grande, Anderson & Keys, 1963; Forbes, 1964; Lesser & Zak, 1963; Owen, Jensen & Fomon. 1962). must~ The family nurse clinician and other health professionals, therefore, rely on less exact methods such as skinfold thickness and percentile charts. These methods are considered by many to be the best readily available tools to be used by the clinician in de-' tecting obesity (Newens & Goldstein, 1972; Weil, 1977) albeit they are not the most exact. 13 Several percentile charts have been designed to correlate height, weight, age and sex for children, but few of the charts actually include guidelines for defining obesity or overweight. Although the Children1s Bureau (1967) has recommended the use of sex specific weight for age and height for age percentiles, there are no recommendations for combining the weight for age and height for age classifications to arrive at estimations or definitions of overweight or obesity. Furthermore, these charts do not account for skinfold thickness measurements. Fomon (1977) does not specifically define obesity or overweight using the height for age and weight for age percentile charts pub 1i shed by the U. S. Department of Health, Educa ti on, and \-Je 1fare Public Health Service, but interprets those children ranking above the 95th percentile as candidates for further evaluation and treatment. A newer method, the skinfold thickness technique, is being utilized by clinicians to supplement the height and weight percentile charts in identifying obesity. Skinfo1d thickness has been unscienti- fically tested for centuries as lIa pinch of the flesh posity. ll to assess adi- The scientific method for measuring skinfo1d thickness en- tered into clinical practice upon the investigation and perfection of the calipers. The caliper skinfold method ;s well recognized as being the most accurate and reliable clinical technique of choice for assessing fatness in children at all ages (Mayer, 1972; Stuart &Sobel, 1946; Wei1, 1977). However, this technique alone should not consti- tute the entire obesity evaluation since other factors such as height and weight also deserve consideration (vleil, 1977; ~/oo1sey, 1973). 14 Pate (1978) and Tanner (1978) simply suggest that because skinfold thickness and percent of body fat are considered directly related, then the sum of appropriate skinfolds can be utilized as a measure of body composition and thus, obesity. The left triceps and left sub- scapular areas are most commonly used in evaluating the skinfold thickness of preschool children (Committee on Nutrition Anthropology, 1956; Garn &Clark, 1976; Hegstead. Darby, Filer &Shank, 1974; IIExcessive weight gain in infancy," 1970; Parizkova, 1961; Raugh & Schumsky, 1970; Seltzer & Wei1, 1977). r~ayer, 1965; Tanner & ~4hitehouse, 1962, 1975; In addition, the Harpenden skinfo1d caliper gauge is highly recommended for use with children over four years of age and adults (Jones, 1972; Edwards, 1955). Although the standard percentile rankings for tricep and subscapular skinfolds of four- and five-year-old children have been interpolated (Fomon, 1977), unfortunately, no American studies are available to substantiate these interpolations. Tanner and Whitehouse (1962, 1975) have illustrated correlations of subscapular and tricep skinfo1d thicknesses of British children ages 1 to 19 years of age and defined obesity as measurements ranking above the 97th percentile. The difference between British and American children's growth variables is unknown (Tanner, 1978). External Controls of Food Intake Schaeter and his co11egues (1968) are acknowledged as the originators of the most pertinent empirical \'iork on the eating behavior of humans, particularly obese individuals (Nisbett, 1972). 15 Schaeter's (1968) theory supported by Nisbett (1972) and Bender (1976) states that obese individuals may differ from those of normal weight in that their eating is controlled largely by external or environmenta 1 factors. External factors are those which, while affecting food intake, are related to nutrition. These external factors include perceptual and sensory aspects such as social stimuli, taste properties of food, visual sight and availability of food. Because few studies have been conducted involving preschool children, and no studies have correlated the effect of environment upon food consumption with body weight and body fat of obese and nonobese preschool children, these variables will be reviewed for both children and adults. Social Stimuli Social stimuli which influence food intake can be illustrated during childhood. During this time, the child learns social rules affecting food etiquette and eating schedu1es, meal eating and snacking, helping and serving sizes, and a range of rules and roles associated with food deemed important by the family (Breckenridge, 1959; Bryan & Lowenberg, 1956; Jerome & Frese, 1978; t10bbs, 1970; tkCarthy, 1935). The concept of the family's role in influencing children's food habits, specifically, food intake, was demonstrated in a study conducted by Emmons and Hayes (1973). The authors reported that diet- ary customs and nutritional knowledge of mothers and their children more often pertained to habit, customs, likes and dislikes than to "nu tritiona1 reasons.1I Litman, Cooney, and Stief (1964) also 16 substantiated the role of the family in affecting children's food behavior. A child in a family atmosphere where overeating is the rule will tend to form a fixed eating pattern which will predispose him or her to obesity. In short, the family portrays an integral part of the social stimuli received by the child which in turn influences his food intake. Nisbett and Storms (1972) found the food intake of both normal and obese adults to be affected by the observati on of amounts of food eaten by others. These results confi rm the im- portance of social stimuli which is thought to influence cognitive contra 1 of food "j ntake. Taste Properties The taste properties of food appear to be relevant sensory factors involved in supporting the Nisbett's and Schaeter's external control of food intake hypothesis. Vance (1932), and Vance and Temple (1975) reported the palatability and the availability of food to have significant effects on the food consumed by preschool children. In addition, texture and temperature along with taste have been demonstrated to be important in the child's acceptance of food (Lamb & Ling, 1946; Lowenberg, 1948; Vance, 1932). Deck (1971) manipulated the taste properties of food for an adult population and relayed the observation that obese individuals ingested more pleasant tasting milk shakes than the nonobese subjects and considerably less was ingested by obese individuals when the milk shakes were adulterated with quinine. In addition, the obese adults demonstrated a higher sensitivity to taste and an increased 17 consumpti on of food cons i dered pl easant tas ti ng than nonobese adul ts. These observations provide evidence that the taste properties of food are significant factors involved in an "individual's food consumption. Nisbett's study in 1968 using adulterated and nonadulterated ice cream also examined responsiveness to taste in obese and nonobese adults. The difference between the amount of both liked and disliked ice cream eaten by obese subjects proved greater in the obese than for the normal and underweight subjects. In reference to the total intake found in the obese adult subjects, Price and Grinker (1973) and Cabanac and Duclaux (1970) supported Nisbett (1968) when they observed that obese adults ate more of everything presented to them than normal weight individuals. Therefore, evidence suggests that the taste properties of food be considered an important factor in the food consumption of preschool children and that these properties are highly influenced by social stimuli. Furthermore, research has suggested that obese adults are highly influenced by social st"imuli, are highly taste responsive, simply eat more food, and have a greater preference for good tasting food than normal \'/eight individuals. Sight and Availability of Food Sight, another external factor thought to affect food intake, appears to have great influence on the dietary practices of children. Commercial television affects the child's developing sense of what the culture deems appropriate to eat and also influences the child's choice of food and beverage (Jerome, 1975; Gussow, 1972, 1977). t1ayer and Neumann (1977), Metheny, Hunt. Patton. and Heyes (1962), and I'leil 18 (1977) noted a direct connection between the types of food presented on the television screen and those foods young children perceive as being desirable and appropriate to consume. A study by Lewis and Lewis (1974) revealed that 60 to 80 percent of fifth and sixth graders believed the television advertising in regard to drugs and snack foods. The stimulation of food intake by sight alone has thereby been demonstrated to significantly influence the food choices of children. The effects of sight on food intake have also been reported in several adult studies. Nisbett's (1968) classic study showed that in response to the sight of varying quantities of food following a deprivation period, obese subjects ate significantly more than normals when presented with an abundance of food. Nisbett's results sup- ported those of Mayer, Monello and Seltzer (1965) which suggested that the obese adult habitually will eat everything presented to them in a typical meal. Among these same subjects. Nisbett (1968) noted-- by utilizing a questionniare--that the greater the weight of the subject, the more often the plate was completely cleared of food, indicating the influence of sight and availability of food consumed by overweight individuals. Assuming the traditional roles in which females prepare more meals than males, females would be exposed to the sight of food more frequently than males. Hewitt's (1974) and Litman's et al., (1976) snacking studies which indicated females snack significantly more than males, would support a correlation drawn between the frequency, availability, and amount of food consumed with the visual exposure to food. Therefore, the visual sight of food can be 19 regarded as an important element externally affecting food intake of the obese individual, specifically, the obese, female adult. Nisbett (1972) and Hashim (1977) quoted an unpublished report by Ross (1970) who discovered that when saliency was manipulated obese adults ate significantly less when the food was dimly in sight. Johnson (1970) also quoted by Nisbett (1972) in his unpublished research has shown "overweight" subjects were more stimulated to obtain food when they could see than when they could not see as well, while this was not the case for the normal weight subjects. Penick and Stunkard (1970) summarized these studies when they concluded that the intake of the obese adult is considerably influenced by external environmental factors such as visual cues and effect of light. These studies lend credence to the external control of food intake theory which includes the concept that children's food consumption and choices are highly influenced by social stimuli, taste properties of food, and visual stimuli. The foregoing studies also provide evidence that obese adults are highly susceptible to external environmental stimuli, have increased sensitivity to taste properties, and respond positively when seeing readily available food. The high susceptibility to external cues presumed to characterize the obese adult may well lead the child--in a society with an abundance of food and cues which encourage eating--to chronic over indulgence and subsequent overweight or obesity. In understanding the contributing factors related to obesity and food intake, the terms "overweight ll and "obese" have been clearly differentiated. The variables involved in the external influences or 20 controls of food consumption were reviewed and discussed. It is necessary to next discuss the relationship between calorie intake and food choices of children as related to obesity and food consumption. Calorie Intake and Food Choices It is generally well accepted that weight gain and/or obesity occurs when there exists a positive calorie state greater than the energy expenditure of an individual over a period of time (American Academy of Pediatrics, 1967; "r~anagement Hashim, 1977; Stunkart, 1975). of the fat child," 1966; Several causes have been postulated to contribute to a positive calorie balance state in children, one being the simple excess of calories consumed (American Academy of Pediatrics, 1967; Asher, 1966; Eid, 1970; Hinton et a1., 1963; Mobbs, 1970; "Excessive weight gain in infancy," 1970; Mullins, 1958; Committee on Nutirition, 1968). Although results from four studies indicate obese adolescents consume the same or fewer calories than their nonobese peers (Hampton, Huenemann, Shapiro & rHtchell, 1967; Johnson et al., 1956; Mayer, 1975; Stefanick, Heald & ~4ayer, 1959). This pattern has not been verified in obese and nonobese preschool children. The average calorie intake by a five-year-old child of average or above average socioeconomic class has been calculated to be between 1,630 to 1,660 kilocalories per day (Abraham &Carrol, 1977; Owen et al., 1974). Even though no documentation is available, the majority of sources have assumed that obese children consume more calories than nonobese children (Huenemann, 1974; Weil, 1977). Some have speculated obese 21 children may consume fewer calories than nonobese children (Mayer, 1975; Weil, 1977). Surprisingly, no studies have been conducted to specifically investigate the relationship of the percentage of body fat and body weight to calories contained in foods and the numbers of food items chosen by preschool children. One related study by Gates et al., (1975) attempted to correlate these same factors in an adult population. Focusing their re- search on food choices of the obese and nonobese, Gates and his associates (1975) observed that obese adults not only chose larger quantities of food but also tended to choose more servings classified as high in calories. These investigators suggested their findings may be related to the externa1" cues theorized by Schaeter in 1968. l Al- though the Gates study had obvious limitations, a major obstacle regarding the observational techniques of food intake was overcome. The difficulty in obtaining accurate records of food intake from the obese individual had been recognized. Noting this difficulty, Gates and his associates carefully observed the food choices--for assumed consumption--of their population without knowledge of purpose or activity. Because this technique of unsuspecting observation was employed, anthropometric measures--skinfold thickness and height versus weight-were not made. Therefore, the individual was subjectively classified according to his/her appearance as obese, stocky, sturdy, or slender. The classification of obesity in this manner seriously limits the study's validity. Another obvious limitation of the Gates' study--as recognized by the investigators themselves--was that only types of food chosen 22 and the number of servings were evaluated, not actual food consumed or food wasted. Conclusions were drawn based on the supposition that all the food which was chosen by the adults in the cafeteria would be consumed by them. Even though the Gates' study exhibited definite limitations, it appears to reflect findings supportive of other investigators who have noted the attractiveness of the food (Schaeter, 1968; Nisbett, 1968), the availability of the food (Schaeter, 1968; Nisbett, 1968; Price & Grinker, 1973) and the pleasant taste of the food (Deck, 1971; Cabanac & Duclaux, 1970) to have a significant effect on the food intake by tile obese adul t. No study thus far has been conducted which controlled and considered all the variables thought to be significant in the evaluation of food intake of the obese and nonobese individual. These variables include the determination of exact food intake, the amount served and the amount wasted, antropometric measurements to accurately determine obesity, unknowing observations, and repeated observations to assure representation of usual eating habits or patterns {Gates et al., 1972). Further research considering these variables appears to be indicated in order to verify food and calorie consumption of the obese and nonobese adult and child. In addition to the environmental stimuli affecting food choices. the use of food and the foods preferred by the family i nfl uence the child's food preference. For some families. food often functions as a reward or punishment, thus affecting the chi1d's attitude towards eating it. A study by Litman and associates (1964) confirmed this 23 phenomenon by interviewing 1,039 Minnesota children as to what foods they would be "praised" or "scolded ll for eating or not eating. Fur- thermore, the manner in which the family emphasized food and eating has also been verified to influence the child's attitudes towards food (Mendleson, 1964). The food preferences of both parents and siblings share an important role in affecting the food choice patterns of family members (Bryan & Lowenburg, 1956; Litman et al., 1964; Sims & ~10rris, 1974). Frequently, a child will not be exposed to certain foods because they are disliked by another family member. According to Leverton and Coggs (1951), girls are more \'Ii11ing to vary their food choices than boys. Also, in relation to willingness to vary food choices, 60 per- cent of children over four years of age reportedly would readily try a new food (Owen et al., 1974). Evidence, therefore, suggests pre- school girls are more flexible in their food choice patterns than boys. Because of the problems attributed to obesity in adolescence and adulthood, emphasis for future research should be directed toward the study of calorie intake, food consumption, and environmental stimuli in relation to food intake of the young. In these areas of con- cern, no studies have been conducted or reported involving preschool children. Specifically, no literature is available to indicate whether or not preschool children with higher skinfold thickness values and body weights--when given a choice--will select more of those foods considered to be high in calories than those with lower skinfold thickness values and body weights. 24 It appears that if obesity and its related disorders are to be effectively alleviated in our population, efforts should be directed primarily toward the young preschool child (Knittle &GinsbergFellner, 1975). The family nurse clinician needs to be actively in- volved in the prevention and detection of obese children and take action to counsel and educate the obese child and his/her family. Be- cause many families today tend to pass off the appearance of obesity in their child as being fat which he or she will eventually outgrow, early identification of the obese child involving careful measurement of weight, height, and skinfolds seems significant. If identi- fied early in childhood, the clinician is able to impart nutritional and environmental education at an appropriate state in the development of obesity. More research should be conducted to clarify factors thought to contribute to the development of obesity in order to aid the family nurse clinician and other health professionals in weighing the relative contribution of each factor for subsequent treatment, and curtailment. ~revention, CHAPTER III CONCEPTUAL FRAMEWORK Because many researchers have experienced difficulties in defining obesity and its related contributing factors, many questions remain unanswered. Several percentile charts have been designed to correlate height, weight, age, and sex for the assessment of obesity in children, but few of the charts actually include guidelines for differentiating obesity from overweight. Neumann (1977) reported that the utilization of only percentile charts for defining childhood obesity variables does not differentiate the child's weight problem as being one due to increased fat deposition from one due to both increased lean body mass and increased fat deposition. No specific methods or standards have been established for four- and five-year-old children (Owen, 1978). In addition to the problems of defining preschool obesity, confusion concerning the etiology remain. Many have speculated con- cerning the cause of obesity, but no one theory has yet been generally accepted. Although the specific etiology is unknown, several factors have been demonstrated to be contributors to obesity in adults. Two of these factors include the visual stimulation of food and the calorie content of foods consummed (Mayer et al., 1965; Nisbett, 1968, 1972; Penick &Stunkard, 1970; Schaeter, 1968). It is desirable to investigate the effect these two factors (visual stimulation of food and calories contained in consumed foods) 26 have as contributors to obesity in preschool children. Furthermore, it is necessary that studies regarding obesity in the preschool child include not only weight and height, but an estimate of total body fat in order to appropriately assess adipose deposition. The estimation of body fat in preschool children is best obtained clinically when skinfold thickness is measured through the use of skinfo1d calipers (Mayer, 1972; Stuart & Sobel, 1946; Weil, 1977). This study proposed the investigation of the relationship between the visual stimulation of food, calories contained in food choices and the anthropometric measurements--height, weight, and mean tricep and mean subscapular skinfo1d measurements--of five-year-01d, white, female children. Hypotheses Hypothesis 1 Five-year-old, white, females of greater body weight and body fat will choose a significantly greater number of food items than those five-year-old, white, females of lesser body weight and body fat. Hypothesis 2 Five-year-01d, white, females of greater body weight and body fat will obtain a significantly higher total caloric content in their food choices than those five-year-old, white, females of lesser body weight and body fat. 27 Definitions of Terms Body Weight Body weight was defined as that measurement which is obtained by weighing the child on a calibrated standing scale to the nearest half of a pound. Body Fat Body fat was defined as the mean left tricep and mean left subscapular skinfold measurement obtained by utilizing the Harpenden skinfold caliper gauge. Food Items Food items was defined as those artificial foods selected from the general assortment of foods displayed which are representative of real foods. Total Calories Total calories was defined as the calculated summation of calories designated by Nasco West, Inc. to be contained in the food portions represented by the food replicas. CHAPTER IV METHODOLOGY AND RESEARCH DESIGN Aim The purpose of this study was to determine the similarities or differences between (1) Measurements of body weight and skinfold thickness of preschool females and the number of food items they will choose when given the opportunity, and (2) measurements of body weight and skinfold thickness of preschool females and the total calories contained in their food choices. Population The population consisted of five-year-old, white, females attending preschools in the Salt lake School District in Salt lake City, Utah. The children attending these schools were from families whose gross annual income--as reported by data collected by the Salt Lake City School District--was between $9,500 and $13,000. These income levels were consistent with those defined as upper-middle and upper socioeconomic classes by Owen et al., (1974) in the Ten State Nutrition Survey. Sample Selection and Exclusions The technique for selection of schools--to be utilized according to income range--was similar to that used by Ginsberg-Fellner and Carmel (1974) in their study of preschool children. At the time the 29 children were selected from the schools it was discovered that the actual gross annual income ranged above and below that reported by the Salt Lake City School District. At that time~ the investigator made the decision to include those children of higher and lower income brackets. One hundred and two, five-year-old, white, females who met the criteria for the sample collection were selected as the sample population. $38~000 The actual gross annual income ranged from while the mean gross annual income was $6~200 to $14~500. Criteria for Sample Selection Children were excluded according to a modified medical history questionnaire completed by the child's parent or guardian at the time the consent form was signed (see Appendices A and B). The medical history questionnaires were numerically coded to the consent forms and the master code list to assure the anonymity of the child (see Appendix H). The children who exhibited or had a past history of the following conditions or characteristics were exluded from participating in this investigation. 1. Head injury or encephalitis--suggesting possible hypo- thalamic involvement--{Jones, 1972; Mobbs, 1970). 2. Cushing syndrome (Jones~ 1972; Puzynski & Biehusen, 1962) . 3. sone~ 4. Currently taking the following medications: Corti- Amphetamines, or appetite suppressants. Adrenal cortical disorders (liThe overweight child,H 1970; Puzynski & Biehusen~ 1962). 30 5. Blind or have difficulty seeing--suggesting inabili- ties in distinguishing food types. 6. Deaf--suggesting an inability to hear instructions by the investigator. 7. Thyroid disorder--suggesting possible effects of metabolism--(Knitt1e, 1972}. 8. Diabetus mellitus--suggesting possible effects of glucose on weight gain and control of food intake--(IIThe overweight chi 1d, 9. II 1970; Mayer, 1975}. Children who are not considered to be white or Caucasian (Garn &Clark, 1976). These exclusions were necessary to eliminate the possibility that the children's food choices would be mediated by -internal or external factors other than visual cues and established preferences. Des ign The design of this study was descriptive. The study did not involve a control group as the procedure was the same for the entire sample population. Instruments Scale The preschool children were weighed on a beam balance scale, Model 1522, manufactured by Continental. The scale had nondetachable weights as recommended by Hegstead et a1., (1974) and Owen et al., (1974) and measured weight to the nearest one-fourth pound (Sims & 31 Morris, 1974; Smith & Brown, 1970). The scale was calibrated prior to each procedure by weighing a standardized, one-fourth pound weight (Fomon, 1977). Tape Measure A plastic-coated, flexible, cloth tape measure was used in this study to obtain height. The tape measure was calibrated in cen- timeters. Right Angle Block A right angle block was u'sed in conjunction with a tape measure to obtain height. The homemade, l2-inch right angle block was con- structed from three-fourth inch lightweight plywood. The handle was designed from a wooden block (3" x 3") and was firmly attached to the horizontal angle of the right angle block. Skinfold Calipers The skinfold calipers were used to measure the skinfold thickness of each child because it had been demonstrated to be ali accurate and reliable method for the assessment of fatness in children of all ages (Owen et al., 1974; Weiner & Lourie, 1969). The Harpenden skin- fold caliper gauge was chosen for use over others such as the Lange and the Best calipers because many anthropologists support its utilization for children after the age of three years (Fomon, 1977; Keys, 1956; Owen et al., 1974; Tanner, 1959; Tanner & Whitehouse, 1962, 1975). According to Edwards (1955) and Fomon (1977), the Harpenden skinfold caliper gauge qualified as a highly reliable tool for measuring skinfold thickness. The Harpenden skinfold caliper gauge, manufactured 32 by British Indicators Limited, was constructed to exert a constant pressure of 10 grams per square millimeter at all openings and was calibrated as such by the manufacturer. The instrument was received new--direct from the manufacturer--and was not recalibrated during the procedure as advised by the manufacturer. Artificial Food Items The artificial food items were made of solid, durable latex and were colored to resemble real foods. The artificial food items were constructed in child-size portions and their proposed calorie content was calculated by Nasco West, Inc., the manufacturer (see Appendix I for artificial food listing and calories). The artificial food assortment from which the specific food selections were made was based on studies indicating the types of foods preferred by five-year-old, upper-middle and upper-class females (Owen et al., 1974), and studies demonstrating foods preferred by preschool children (Breckenridge, 1959; Bryan & Lowenburg, 1956; Jerome &Frese, 1978; Lamb & Ling, 1946; McCarthy, 1935; Vance &Tample, 1975). Twenty artificial foods were originally to be utilized in this study. In order to designate these specific food replicas from which selections would be made by the children, percentage calculations were necessary. Data from studies by Owen et al., (1974) illus- trated the average percent contribution of nine food groups to the daily intake by five-year-old female children. These percentages pro- vided the means to calculate the appropriate number of foods from 33 each of the nine food groups in order to determine the total artificial food items for selection. Because the percentages used to cal- culate the number of artificial food items from each of nine food groups were rounded to the nearest whole number, 23 items were designated as being the total number from which the selection would be made. A random selection of 23 food items was obtained from each of the nine food groupings containing the specific food preferences of preschool children. ing items: The artificial food selection included the follow- pork, pizza, cornbread, raisins, pie, jello, hamburger, milk, chicken, peaches, ice cream, soda crackers, cheese, popcorn, tuna fish, graham crackers, bologna and bread, baked potato, french fries, banana, pickle, tomato, and corn. The artificial foods were classified into two groups (Group A and B) similar to the classifications of foods used by Gates et al., (1975). Group A foods--generally considered to be higher in calories than Group B--including the follwoing kinds of foods: pasta, rice cereals, starchy vegetables, bread, desserts, snack chips, crackers, soft drinks, lemonade, jams, jelly, table fats, gravy, sauces (Gates et al., (1975). Group B foods--generally considered to be lower in calories than Group A--included the following types of foods: milk, meat, eggs, cheese, legumes, non- starchy vegetables, fruit, juices (Gates et al., (1975). Measurements Weight The five-year-old female children were weighed on a calibrated beam scale. The children were weighed following the emptying of their 34 bladders, and wearing only lightweight clothing similar to those weighed in Smith and Brown's (1970) study of Hawaiian preschool children. Not requiring all the clothing to be taken off decreased pos- sible chilling and prevented undue distress. No corrections were made for weighing the children in lightweight clothing. The weights were measured to the nearest quarter pound (Sims &Morris, 1974; Smith &Brown, 1970), converted to kilograms, and recorded on the data collection form (see Appendix D for detailed weight measurement procedure). Height The height of the female preschool child was measured by using a plastic-coated, flexible cloth measuring tape calibrated in centimeters which was fixed to a vertical flat \'/all. The child stood erect and barefoot with her buttocks, heels, shoulders, and the back of the child's head touching the wall. A lightweight block squared at right angles against the wall was placed on the top of the child's head and the measurement was recorded to the nearest centimeter (Fomon, 1977; Hegstead et al., 1974) on the data collection form (see Appendix E for detailed height measurement procedure). Skinfold Measurement The Harpenden skinfold caliper gauge was utilized in obtaining the tr;cep and subscapular skinfold thickness measurement of fiveyear-old preschool females (American Academy of Pediatrics Committee of Nutrition, 1967; IIExcessive weight gain in infancy," 1970; Garn, Rowen &McCann, 1971; Malena, 1972; Shepard, Jones, Ishii, Kaneko & 35 Olbrecht, 1969; Sims &Morris, 1974; Sloan & Shapiro, 1972). The left side was utilized in each case as recommended by the majority of anthropologists for the purpose of technique standardization (Edwards, 1955; Tanner, 1959; Tanner &Whitehouse, 1962, 1975). The left subscapular and left tricep skinfold thickness of 50 preschool children were measured prior to the data collection. This measurement exercise aided the investigator by increasing her ability in handling the Harpenden skinfold caliper gauge, in accurately locating the measurement site, and in correctly reading the caliper gauge. The Harpenden skinfold caliper gauge was not recalibrated following original calibration as recommended by British Indicators Limited, the manufacturer. The left tricep skinfold measurement was obtained with the child's bare, left arm relaxed to her side. The area to be measured by using the Harpenden skinfold caliper gauge was located and marked half-way between the tip of the left acromion process and the tip of the olecranon process of the ulna. The investigator picked up the skin between the left thumb and forefinger, one centimeter above the half-way point marked on the upper arm. The calipers were then ap- plied at the point marked which was one centimeter below the fingers grasping the skinfold. The right hand fingers of the investigator released the handle of the calipers to permit full force of the spring to be exerted on the skinfold while continuing to pinch the skinfold between the left thumb and forefinger. Immediately following the re- lease of the caliper handle, the resulting measurement on the caliper gauge was recorded to the nearest 0.1 millimeter. The left tricep 36 skinfold measurement procedure was repeated three times, and the average of the three measurements was calculated and recorded on the data collection form (see Appendix F for detailed tricep measurement procedure). The left subscapular skinfold measurement was obtained by the investigator with the child's bare, left shoulder and left arm relaxed. The skin thickness was picked up by the investigator between the left thumb and forefinger and measured using the Harpenden skinfold caliper gauge or calipers. The calipers were applied one centimeter above a mark which was made just inferior and lateral to the inferior angle of the left scapula, in the line of the natural skin cleavage. The skinfold continued to be pinched by the investigator while the handle of the calipers was released from the investigator's right hand to permit full force of the spring to be exerted on the skinfold. Three seconds following the release of the caliper handle, the resulting measurement on the caliper gauge was recorded to the nearest 0.1 milThe left subscapular skinfold measurement procedure was limeter. repeated on the data collection form (see Appendix G for detailed subscapular skinfold measurement procedure). Procedure Sequence 1. The procedure was consistently conducted prior to lunch or snack time (Deck, 1971; Hashim, 1977; Nisbett & Storms, 1972; Pliner, 1973). 2. Consent was personally obtained through the child's parent or guardian and completed the day prior to the procedure (see Appendix A for consent form). The child's parent or guardian received a 37 telephone call from the investigator to answer questions related to the study. The short medical questionnaire was completed at that time by the child's parent or guardian (see Appendix B for medical questionnaire form). The medical questionnaire, the data collection form, and the consent form were numerically coded to the master code list to assure anonymity of the children. 3. Prior to the procedure, the scales were calibrated and the measuring tape was securely positioned on the wall. The Harpenden skinfold caliper gauge and scales were positioned out of the childls sight until needed in the procedure sequence. 4. The food replicas were placed on a table (3 x 3 1 in adequate light. 1 X 9') The food replicas were positioned on the table according to their randomly selected positions (see Appendix C for food positioning chart). 5. The investigator introduced herself to the child and asked the child--if she needed to or was able to-_lIgo to the bathroom" at this time, and instructed her to return to the specified room as soon as possible. 6. The investigator then asked the child upon returning from the bathroom to, "Pl ease step over to the table where all the food ; s placed. II 7. The investigator suggested to the child, "Ild like for you, (name of child) to playa game with me. II 8. The investigator then asked the child U(name of child) let's pretend that these foods are real and that we are in a cafeteria or restaurant where you can have all the food you want to eat from these foods on this table." 38 9. The investigator then gave these instructions to the child: lI(name of child), I have to leave the room for a couple of minutes, while I'm gone you'll have time to choose what you1re going to have for dinner. I want you to put all the food that you want to eat on this tray and I'll be right back." 10. The investigator left the room with the child confronting the food selection. 11. The investigator returned in three minutes and asked the child, "Have you chosen a 11 that you want to eat?" 12. The investigator then gave the child 30 additional seconds to choose any more items desired. This technique helped to assure the child that she could have chosen anything she wanted. 13. The investigator then completed the game by choosing three foods which she supposedly wanted to eat. 14. The investigator thanked the subject for playing this game with her. 15. The investigator asked the child to step on the scales to be weighed and the measurement of the resulting weight was recorded to the nearest quarter pound on the data collection form (see Appendix D for weight measurement procedure). 16. The investigator asked the child to step up to the measur- ing tape on the wall to measure how tall she was. The measurement of the resulting height was recorded on the data collection form (see Appendix E for height measurment procedure). 17. The skinfo1d thickness measurement was taken (see Appendi- ces.F and G for tricep and subscapular measurement procedure). 39 18. The child was again thanked for her participation in the activity. 19. The food choices made by the child were recorded by the investigator on the data collection form (see Appendix H for data collection form). 20. The foods were then replaced in their respective position on the table to be ready for the next child's selection. Statistical Procedures In the analysis of the collected data two statistical calculations--the means and standard deviations--were obtained to describe the variables. The calculated means of the study's variables--total number of foods chosen, total calories of foods chosen, number of foods chosen from Group A, number of foods chosen from Group B, height, weight, and mean left tricep and left subscapular skinfold measurements--described features of the average, five-year-old, white, female of the 102 participants. Therefore, particular characteristics of the typical five-year-old, white, female--in terms of these variables-were demonstrated and thus examined. dard deviations were calculated. Related to the mean, the stan- Computation of the standard devia- tions allowed examination of the amount the sources of each variable actually varied about its mean or average. Standard deviations there- by indicate the amount of variability and range demonstrated by the resulting data in relation to the mean (Bartz, 1974). In addition to describing characteristics of the scores within the variables themselves, simple correlations and significance probabilities were employed to examine possible relationships between the 40 outcome variables (the total number of foods chosen, total calories of foods chosen, number of foods chosen from Group A, number of foods chosen from Group B) and the measurement variables (height, weight, mean measurements of left tricep and left subscapular skinfolds). Two functions considered important in determining correlations between variables included: to aide in developing a single value which will-- at a glance--indicate strength or amount of the relationship, and to predict results of one variable from the knowledge of another variable (Bartz, 1974). Two variables were considered to be positively correlated when the calculated correlation coefficient was significantly large. The closer the correlation coefficient approached 1.0, the stronger the correlation or relationship between the variables. The signifi- cance of the relationship between variables was determined by computing the significance levels or probabilities of the correlation coefficients. The significance levels were determining using the mean and standard deviations of the variables. ficance accepted was set at £~.Ol. The level of statistical signiIf the probability that the results happened by chance was .01 or less then the results were considered significant at the .01 level. When a relationship between two variables was thought to be caused by a third variable, partial correlations were considered necessary to be computed. Furthermore, when specific measures were taken to control the third variable--possibly affecting a relationship between two variables--then a clearer significance of relationships and correlations was obtained. Utilizing partial correlations in the sta- 41 statistical analysis of special variables, control of the variables-height and weight was possible. The significance of the relationship demonstrated between variables was determined by computing the significance levels or probabilities of the partial correlations. cance accepted was set at Q~.Ol. The level of statistical signifiIf the probability that the results happened by chance was .01 or less, then the results were considered significant at the .01 level. ~ \. :.., .' CHAPTER V ANALYSIS OF DATA Hypotheses The original hypotheses of this study were: (l) five-year-old, white, females of greater body weight and body fat will choose a significantly greater number of food items than those five-year-old, white females of lesser body weight and body fat, and (2) five-year-old, white, females of greater weight and body fat will obtain a significantly higher total calorie content in their food choices than those five-year-old females of lesser body weight and body fat. It has been demonstrated that preschool five-year-old, white females of greater body weight and particularly greater body fat--as signified by mean skinfold thickness measurements--indicated food choices of greater numbers and total calorie content than those preschool females of lesser body weight and body fat. The statistical hypotheses of this study were as follows: Hypothesis 1 The outcome variables (which include the number of itenls chosen from Group .A and Group B and the total number of foods chosen) are correlated with the measurement variables of height, weight, and mean measurement of the left subscapular and left tricep skinfolds. 43 Hypothesis 2 When height and weight variables are controlled, a positive partial correlation remains between the mean measurement of the left tricep and left subscapular skinfolds and the total number of foods chosen, the total calories of those foods chosen, and the number of foods chosen from Group A and Group B. Statistical Examination of Data Descriptive Statistics The mean and standard deviations were calculated on variables which included: the total number of foods chosen, total calorie con- tent of foods chosen, the number of foods chosen from Group A, the number of foods chosen from Group B, height, weight, mean left tricep, and mean left subscapular skinfold measurements (Table 1). Because there was a wide range in each of the variables, the standard deviations were large in relation to the means of the Table 1 Descriptive Statistics Variables Total number of foods chosen Total calories of foods chosen No. of foods chosen from Group A No. of foods chosen from Group B Height (cm) 'tJeight (kg) Mean left tricep skinfold measurement (mm) Mean left subscapular measurement (mm) j·1ean Standard Deviation 5.28 2.17 741 .97 344.44 3.17 1.74 2.54 1.45 4.32 cm 113.85 cm 2.57 kg 19.86 kg 10.61 mm 2.l8mm 6.72 mm 1 .84 mm 44 variables. An example of this high degree of variability was demon- strated by the numbers of food items chosen by the children. The range of this variable (number of food items chosen) varied from one to 10. In short, the standard deviations of several variables were almost one-half their mean, therefore, the data presented a wide range of variability. Simple Correlations The data were analyzed by correlation coefficients and significance levels or probabilities. Correlation coefficients were computed to demonstrate possible relationships between the outcome variables the total number of foods chosen, total calorie content of foods chosen, the number of foods chosen from Group A, the number of foods chosen from Group B}, and the measurement variables (height, weight, mean left tricep skinfold measurement, mean left subscapular skinfold measurement). The correlation coefficients of the outcome and measurement variables are displayed in Table 2. In determining the strength of the relationship between the variables by utilizing correlation coefficients--the results indicate a low but positive correlation between the outcome and measurement variables. Therefore, since the outcome variables were positively correlated to the measurement variables, then statistical hypotheses 1 which states this correlation can be accepted. The strongest correlation or relationship (.68) was noted between the mean left tricep skinfold measurement and the total calories of foods chosen. In addition, a relatively strong relationship was indicated between the Table 2 Correlation Coefficients Variables Total Number Total Calories of Foods of Foods Chosen Chosen Number of Foods Chosen From Group A Number of Foods Chosen From Group B Height 0.24 0.23 0.10 0.04 Weight 0.30 0.40 0.17 0.08 Mean left tricep skinfo1d measurement 0.47 0.68 0.43 0.27 Mean left subscapular skinfo1d measurement 0.47 0.57 0.36 0.20 +:> U1 46 mean left subscapular skinfold measurement and total calories of foods chosen (~= .57). As noted in Table 2, the measurement variables (mean left tricep and mean left subscapular skinfold measurements) correlated with all four outcome variables to a greater extent than did height and weight variables. Table 3 illustrates that for correlations between the measurement variables (mean left tricep measurement and mean left subscapular skinfold measurements), and outcome variables (the total number of foods chosen and the total calories of foods chosen) the probability was less than .01 which is highly significant. The mean tricep and subscapular skinfold measurements and the number of foods chosen from Group A were also found to be correlated with a significance probability of .01 or less as demonstrated in Table 3. Furthermore, the mean tricep skinfold measurement was found correlated to the number of foods chosen from Group B at a .01 or less significance level. While the mean tricep and subscapular measure- ments were significantly correlated to almost all the outcome variables, the significance level was greater than .01 for the correlations between measurements variables--weight and height--and outcome variables--number of foods chosen from Group A and Group B. There- fore, despite small positive correlation coefficients, a notable significant probability of less than .01 existed between many variables. These significant probabilities indicated a low probability that the resulting correlation coefficients occurred only by random chance. Table 3 Significance Levels of Correlation Coefficients Total number of Foods Chosen Total Calories of Foods Chosen Height 0.02 0.02 0.31 0.67 Weight <.01 <.01 0.09 0.41 Mean left tricep skinfold measurement <.01 <.01 <.01 0.01 Mean left subscapular skinfold measurement <.01 <.01 <.01 0.04 Variables Number of Foods Chosen From Grou~ A Number of Foods Chosen From Grou~ B ~ ......., 48 Partial Correlations Partial correlations were calculated between the measurement variables (mean left tricep and mean left subscapular measurements), and the outcome variables (total number of foods chosen, total calories of foods chosen, number of foods chosen from Group A, and number of foods chosen from hroup B). Table 4 illustrates partial correlations between measurement variables and outcome variables when statistical measures were undertaken to control the effect of height and weight variables upon other correlations. When measures were taken to control the effects of height and weight upon the relationship of other variables, correlations between the measurement and outcome variables remained significant. The re- sulting data demonstrated support of the statistical Hypothesis 2 which states that when height and weight variables are controlled, positive correlations remain between the mean left tricep, the mean left subscapular skinfold measurement,and the total number of foods chosen from Group A and Group B. The strongest of these correlations was found to exist between the total calories of the foods chosen and mean measurements of the left tricep skinfolds (.59) and the left subscapular skinfolds (.43). Because the total calories of foods chosen by the children produced discrete data rather than continuous data, the results were of particular significant value. Furthermore, the significance probability of .01 or less as shown in Table 5, indicated a significantly low probability that the resulting partial correlations occurred by random chance. The significance levels Table 4 Partial Correlations Vari abl es Total Number of Foods Chosen Total Calories of Foods Chosen Number of Foods Chosen From Group A Number of Foods Chosen From Group B Mean left tricep skinfold measurement 0.39 0.59 0.43 0.27 Mean left subscapular skinfold measurement 0.38 0.43 0.33 0.19 +:» ~ Table 5 Significance levels of Partial Correlations Variables Tota 1 Number of Foods Chosen Mean left tricep skinfo1d measurement <.01 <.01 <.01 <.01 Mean left subscapular skinfo1d measurement <.01 <.01 <.01 0.01 Total Calories of Foods Chosen Number of Foods Chosen From Grou~ A Number of Foods Chosen From Grou~ B 0'1 o 51 indicated that although the values of the partial correlations were low, the positive relationship which they represent were notably significant. Summarizing the results, five-year-old, white females of greater body weight and particularly greater mean tricep and subcapular skinfold measurements have been observed to choose significantly more food items and more food items with higher total calorie content than five-year-old, white females of lesser body weight and mean tricep and subscapular skinfold measurements. In addition, it has been demonstrated that as the measurements of weight and particularly mean tricep and subscapular skinfold measurements of five-year-old, white females increase, one could expect to observe an increase in the total number of foods chosen and predict an increase in the total calorie content of the foods chosen. From these results and conclusions, the original hypotheses can, therefore, be accepted. These concluding results indicate a significant role for nurses who are involved in the care of obese children and their families to become cognizant of the possible effects visual food cues and the calories contained in available foods may have upon the \'Jeight statis of the preschool child. Furthermore, the emphasis of utilizing the Harpenden skinfold caliper gauge in the identification and evaluation of obesity in preschool children by the family nurse clinician and other health professionals is paramount. CHAPTER VI DISCUSSION OF FINDINGS The original hypotheses of this study were (1) five-year- old, white females of greater body weight and body fat will choose a significantly greater number of food items than those five-year-01d, white females of lesser body weight and body fat, and (2) five-year- old, white females of greater weight and body fat will obtain a significantly higher total calorie content in their food choices than those five-year-01d females of lesser body weight and body fat. It was demonstrated that preschool five-year-old, white females of greater body weight and particularly greater body fat--as signified by mean skinfold thickness measurements--indicated food choices of greater numbers and total calorie content than those preschool females of lesser body weight and body fat. Because this investigation produced results which indicated this phenomenon occurred with a significance level of .01 or less, the original hypotheses were thereby accepted. The purpose of this study was to determine similarities or differences between (1) measurements of body weight and skinfold thickness of preschool females and the number of food items they would choose when given the opportunity, and (2) measurements of body weight and skinfold thickness of preschool females and the total calories contained in their food choices. The implementation of guide- lines for the identification of childhood obesity and the recognition 53 of contributing factors require investigation in order to possibly curtail life-long obesity patterns and their associated complications. Two factors considered to have significant influence in contributing to the state of childhood obesity are the visual stimulation of food cues and the consummation of a calorie surplus. No American studies have previously been conducted to develop guidelines for defining obese and nonobese four- and five-year-old females. Although the anthropometric similarities and differences between British and American preschool girls is currently unknown (Tanner, 1978), it is interesting to note the related findings. The mean left tricep measurement (10.6mm) and the mean left subscapular skinfold measurement (6.72mm)in this study were found to closely approximate the findings of a British study involving 1,000 five-year-old females (Tanner &Whitehouse, 1975). Tanner and Whitehouse (1975) found the mean left tricep of five-year-old females to be approximately 10.2mm and the mean left subscapular measurement to be approximately 6.2mm. Furthermore, when utilizing the only available defini- tion for obese five-year-old females--according to Tanner and \-/hitehouse's British study--13.2 percent of the 102 preschool children tested would be considered obese. It was noted by the investigator that many of the children who participated in the study did not appear lIoverwei ght" or "obese through observati on a lone. ll ~~hen measured, several children were found to reveal unexpected measurements which were in the upper range for weight and skinfold thickness relative to the total sample population. This impression indicated to the investi- gator that observation alone was a poor technique for evaluating body 54 weight and percent of body fat and could easily mislead the family or health professional to overlook weight problems or obesity in the child. In addition to the unexpected skinfold measurements obtained from several children, other ancillary findings were noted by the investigator. The income of the children's families--who, according to the school district, had gross annual incomes of $9,000 to $13,000-actually had mean incomes of $14,500 with a range of $6,200 to $38,000. These findings indicate that the data on which the school board based their evaluation of income range according to schools should be recalculated for accuracy. Other ancilllary findings observed in this investigation included the children's hunger response, and the number of children who claimed of having eaten breakfast. All 102 of the sample population stated they were hungry when asked. Furthermore, all but two of the 102 children claimed to have eaten breakfast. No attempts were made to examine what meaning breakfast or hunger had to each child. Height--frequently utilized in evaluating growth and development--was the first anthropometric measurement to be obtained from the children. In the statistical analysis, height did not correlate significantly to the outcome variables--number of food items chosen, total calories of foods chosen, number of foods chosen from Groups A and B (> .01 significance level). Therefore, along with weight, height was statistically controlled in computing the partial correlations. When controlling for the variation produced by these vari- ables----height and weight--the correlation between the mean tricep 55 and subscapular skinfo1d measurements and the outcome variables remained significant (R ~.01 significane level). These results, sub- stantiate the importance of the clinical utilization of skinfo1d thickness measurements in addition to total body weight in order to illustrate a more complete and definite data base on which to structure methods for identification, prevention, and treatment of obesity in children. The preschool females who had greater body weights and particularly greater mean skinfold thickness measurements demonstrated food choices of greater numbers than those preschool females of lesser body weight and body fat (R ~ .01 significance level). The increase in the number of foods chosen by children of greater body fat--according to the mean measurements of tricep and subscapular skinfo1d--imp1ies consideration of Schaeter's external control theory (1968). Schaeter demonstrated the numbers of foods eaten by obese adults to be largely controlled by the visual stimulation of foods. As food replicas were utilized in the selection process and food consumption was not available for observation, appropriate testing of Schaeter's external control theory on the preschool females was not possible. Therefore, total support of this theory was limited; however, the results suggest at least partial support. Because the number of foods chosen by preschool girls was related to greater body weights and skinfold thickness measurements in this study (Tables 4 and 5), it seems reasonable to consider the environmental stimuli of foods for those children with higher \veights and skinfold measurements as being an -important factor in controlling 56 or preventing childhood obesity. This concept supports many v/ho are currently studying the effects of advertising and television on children s ea ti ng patterns (Lewi s I & Levli s, 1974; ~"ayer & Neumann, 1977; Hetheny et al., 1962; Hen, 1977). Attempting to control the types and amounts of foods available to the child would, in theory, effect the strength and amount of stimuli to presumably eat. This theory necessitates scientific investigation in order to substantiate the visual stimulation of food as it relates to obese children and the possible gains achieved when utilized in weight control or weight reduction programs for the children and their families. If this approach were to be evaluated, more emphasis would be placed on the consideration of the nutritional value and amount of food consumed by the child rather than the counting of calories contained in the food eaten da ily. In addition to the correlations regarding the total number of foods chosen and the measurement variables--weight and skinfold thickness--a significant correlation existed between mean skinfold measurements and the number of foods chosen from Group A (£~ .01 significance level.) These results indicated the fatter the child--in terms of skinfold thickness--the greater was the possibility of her choices being foods considered higher in calories. Although the number of foods chosen from Group A and Group B, and the total number of foods chosen from the selection provided important correlations of significance, the variables produced limited data. The variables, number of foods chosen from Group A, the number of foods chosen from Group B, and the total number of foods chosen 57 produced continuous rather than discrete data which decreased the value of the correlation coefficients and also significantly increased the standard deviations about the means. An example can be demon- strated whereby the data did not differentiate when one child chose three foods from Group A and another child chose three different foods from the same Group A. Furthermore, because only the total number of foods chosen and the number chosen from the respective food groupings (A or B), were considered in this investigation--rather than specific types of foods chosen--it ~/as highly possible that the number of foods chosen by one child could have been totally different--in relation to calorie content or kind of food--from those chosen by another child. The mean total calories of foods chosen with presumed consumption by the preschool girls was found to be 740 kilocalories. If the foods were chosen for actual consumption whereby these calories, along with two meals of equal calorie content. were consumed, an average of 2,200 kilocalories would be presumably eaten by the average five-yearold girls per day. It remains a controversial issue whether or not the fatter children actually eat food containing more calories than do their slimmer peers. In this study. a low but positive correlation existed between the measurement variables. particularly the mean skinfold thickness measurements and the total calorie content of the foods chosen. which may well be due to the effects of such things as the environmental stimulation of foods or family habits, and not due to the child's appetite or desire to actually eat all the foods chosen. Obvious reeducation of parents and nursery school directors about the importance of preventing obesity is paramount (Ginsberg- 58 Fellner &Carmel, 1974). The value of food, the purpose and impor- tance of monitoring calorie intake and the environmental stimulus of food, and the significance of early identification of obesity are appropriate elements to be included by the nurse in family teaching plans. The family nurse clinician could be instrumental in the insti- gation of such programs to decrease the prevalance of obesity in childhood and thus curtail life-long obesity patterns in the American population. CHAPTER VII LIMITATIONS The purpose of this study involving the two contributing factors of obesity--visual food cues and food calorie count--was to determine the similarities and differences in food choices or preferences of five-year-old. white females of varying weight and body fat. Specifically, the investigation attended to the number of artificial food items chosen from an assortment and their total calorie content. In the course of this investigation. several limitations were noted. Because artificial foods \'/ere utilized as representative foods for the child's selection. actual food consumption and waste could not be measured. Therefore. consumption of the "food" items chosen by the children could only be assumed. Furthermore. the artificial food replicas had no smell or taste. were cold in temperature. and had similar consistency thereby limiting the sensory effects of the food to sight alone. Had the children been able to sample real food items. different choices. and a greater or lesser number of food choices. there may have been a different selection of foods chosen. In addi- tion to these limitations regarding artificial foods, no previous studies have reported using food replicas in exploring the effects of food stimuli and calorie content of foods chosen by children or adults. It is not known if a child will choose the same artificial foods as natural foods. 60 The absence of measures to control foods eaten for breakfast provided one further limitation. Although steps were instigated to control the time at which the selection of artificial foods was made-in relation to snack time and lunch time--there were no controls to regulate the intake of breakfast foods. Since studies have demon- strated the effects preloads have on the amount of food consumed (Deck, 1971; Hashim, 1977; Nisbett & Storms, 1972; Pl"iner, 1973, considera- tion of this additional factor appears important. In addition to the possible effects of preloads, the food selection environment in which the artificial foods were chosen may not be similar to that which the child ;s accustomed to associating with meals or eating--and could thereby affect her food choices. No reported observations were conducted to validate that the selection made by the preschool child was representative of normal food choice patterns. The Harpenden skinfold caliper gauge used to measure the left tricep and left subscapular skinfold thickness in the assessment of body fat was not retested for calibration between children. Although British Indicators Limited originally calibrated the instrument and advised recalibration only periodically by the manufacturer--British Indicators Limited--no measures were instigated to recheck the caliper gauge. CHAPTER VIII SUMMARY AND RECOMMENDATIONS This study was designed to investigate the contribution of visual food cues and the calorie content food choices toward the development of childhood obesity. These factors were investigated in 102 five-year-old, white females from families of upper-middle and upper-socioeconomic classes. The girls were taken from their pre- school class, one at a time, and asked to visually inspect the selection of food replicas and then to choose those "foods to eat for dinner. ll they would like Height, weight, and tricep and subscapular skin- fold thicknesses were obtained from each child following the food choices. The results of correlating weight and mean skinfold thick- ness measurements to the variables; total number of foods chosen, the total calories contained in the foods chosen, and the number of foods chosen from the foods classified into high and low calorie food groupings--Group A and Group B--proved to be significantly correlated. When statistical measures were utilized to control the effect of body weight and height upon relationship of other variables, the resulting partial correlations between the number of foods chosen, the total calories contained in those choices, and the mean skinfold thickness measurements remained significant (Q.. level). < .01 significance The statistical analysis indicated large standard deviations about the means which signified high variability of the data but 62 despite the large standard deviations statistical evaluation of correlation coefficients were highly significant (£ ~ .01). On the basis of these findings, it is evident that a five-year-old, white female child with greater body weight and particularly greater body fat--as signified by mean skinfold thickness measurements--can be expected to choose more numbers of food items with higher total calorie content than those five-year-old, white female children of lesser body weight and body fat (£ ~ .01 significance level). Therefore, the original hypotheses were accepted. Those nurses who are actively involved in the health maintenance of individuals in ambulatory care settings are in excellent positions to have an effective impact in decreasing the prevalance of obesity. Since there is a greater success rate of treating obesity in childhood (Ginsberg-Fellner &Carmel, 1974), nurses should assume leadership roles in the planning, coordinating, and developing of programs to screen for and identify obesity in young children. It ap- pears important that as nurses utilize the Harpenden skinfold caliper gauge in the evaluation of childhood growth and development, the data must be made available to establish standards on which to structure baselines for clinical evaluations. It is also necessary that nurses be concerned with developing family teaching plans in which the purpose and importance of monitoring the calorie content of available foods, the effects of environmental stimulus of food on consumption, and the significance of early identification of obesity in childhood ;s emphasized. Parents made aware of the effects of television, social attitudes, and the 63 environmental stimulation of food cues on food consumption could have positive effects in decreasing the incidence of obesity in the young. In coordination with other health professions, nurses could be valuable in setting standards for the progressive development of school lunch programs, instigating classroom methods of teaching nutrition to children, and offering parents nutritional guidelines on which to base family food consumption. In the event that this investigation would be expanded, several reco~TIendat;ons are offered. First, by increasing the sample size, the variability might decrease and thus increase the degree of validity. Second, in examining the relationship of other factors to childhood obesity, many correlations may be developed. Examples of correlations which can be developed in the event this study is expanded include: (1) Parental education status as related to knowledge of food value and nutrition, (2) the results obtained when utilizing real food in the design and number of types of foods chosen by preschool children, (3) the relationship of parental anthropometric measure- ments and those of their children, and (4) the food preferences and attitudes of parents as related to those of their children. This investigation also implicated several areas related to childhood obesity requiring further study through nursing research. The effects of controlling the environmental stimulation of food on weight reduction programs for children is one aspect of childhood obesity to be investigated. Is it possible to control weight gain or obtain weight loss by manipulating the available food cues in the child's environment? Furthermore, can this manipulation of 64 environmental food cues affect the child1s food attitudes and preferences in a positive way? If manipulation of food cues can aid in con- trolling weight gain and effect the food attitudes of young children, then how can television be utilized to aid in the orevention of childhood obesity and weight gain? The current methods of obesity detection involving height, weight, and age in normative standards do not provide reliable and accurate estimations of body fat in relation to size. The contribu- tion calipers could provide in the evaluation of growth and the identification of obesity in preschool children is another area of childhood obesity which would appropriately concern researchers in nursing. What variance would the use of calipers produce in the substantiation of guidelines defining obesity in four- and five-year-old children? In conclusion, there are many notable aspects of childhood obesity which should be examined through nursing research. Those in- volved in nursing research could be instrumental in furthering the knowledge of contributing factors and techniques for early identification and prevention of obesity in the young. Were these factors and techniques to be explored, more information would become available to formulate guidelines on which to base clinical practice and thus, assist in decreasing the prevalence of the obesity in the preschool years. APPENDIX A INFORr'1ED CONSENT FOR HUMAN RESEARCH PROJECT THE UNIVERSITY OF UTAH COLLEGE OF NURS I NG GRADUATE PROGRAt-1S 66 I, (first name) (middle name) (last name) , herewith agree to involve my child, , as a subject (first name) (middle name) (last name) in the investigation of Food Choices of Preschool Children under the supervision of The investigation aims to determine the food preferences or choices of preschool female children age five. The procedures to which my child will be subjected are food selection from an assortment of artificial replicas of food and measurements of height, weight, and skinfold thickness. As part of the procedure, I understand I will be request!d to complete a short medical questionnaire concerning my child. are no expected risks. There I may benefit from these procedures by in- creasing my knowledge of foods preferred and chosen by preschool children as well as contributing to new knowledge that may benefit others concerned with the nutrition of the growing child. Because the medical questionnaire is only utilized as a screening tool and will be destroyed immediately following acceptance of my child into the investigation, I understand that confidentiality will be protected. I understand that I am free to withdraw my child from participation in the investigation at any time, and that she will obtain the best care otherwise available. I have read and fully understand the foregoing information. Date Parental or Guardian Signature Code # Address Monthly take-home salary of family: APPENDI X B MEDICAL QUESTIONNAIRE 68 Instructions: Concerning your child's present and past medical history, please mark "yes" or "no" in the spaces provided to the following questions: Yes Has your child ever experienced a severe head injury? Does your child have or has she ever had encephalitis (inflammation of the brain)? Does your child have or has she ever had cushings syndrome (a cortisone related disease)? Is your child currently taking the medication cortisone? Is your child currently taking amphetamines? Is your child currently taking any medication to suppress her appetite? Does your child have or has she ever had any adrenal cortical disorders? Is your child blind? Does your chi 1d have or has she ever had thyro; d problems? Does your child have or has she ever had diabetes mell itus? Is your child considered a white/Caucasian? No APPENDIX C RANDOMLY SELECTED POSITIONS OF FOOD REPLICAS 70 French Fries Mil k Pizza Raisins Cornbread Jello Baked Potato Tuna Popcorn Soda Crackers Banana Peaches Chicken Pork Chop Bologna Sandwich Ice Cream Pie Sheet Cake Tomato Corn Pi ck 1e Graham Crackers Cheese Hamburger APPENDIX D WEIGHT MEASUREMENT PROCEDURE 72 1. The investigator will calibrate the scales by weighing a standardized one-fourth pound weight (Fomon, 1977). 2. The investigator will ask the child to go to the bathroom to try to urinate all he/she is able to before he/she is weighted (Smith &Brown, 1970). 3. The investigator will instruct the child to remove her shoes and any heavy clothing she may be wearing (i.e., sweater, coat, hat) . 4. The investigator will instruct the child to stand evenly on the scales with both feet and with hands by his/her sides. 5. The investigator will report the reading to the nearest one-fourth pound. No corrections will be made for weight of clothing (Smith &Brown, 1970). 6. The weight will then be converted into kilograms and re- corded by the investigator on the data collection form. APPENDIX E HEIGHT MEASURB1ENT PROCEDURE 74 1. A plastic-coated. flexible cloth measuring tape calibrated in centimeters will be firmly fixed to the vertical flat surface of a wall. 2. The child will be asked by the investigator to stand with his/her buttocks. heels~ shoulders. and head touching the tape and with her heels together, looking straight ahead. 3. A lightweight block squared at right angles against the wall will be placed on the crown of the child's head and the measurement is noted by the investigator to the nearest centimeter. 4. The height will be recorded by the investigator on the data collection form. APPENDIX F TRICEP SKINFOLD MEASUREMENT PROCEDURE 76 1. The Harpenden skinfold calioer gauge or calipers will be calibrated by British Indicators Limited, the manufacturers. 2. The investigator will instruct the child to remove his/ her shirt and underclothing covering the upper trunk and arms. 3. Using a felt-tipped pen, the investigator will mark the tip of the left acromion process and the tip of the left olecronon process. 4. Using a calibrated tape measure, the investigator will measure the distance halfway between the two marks, marking this halfway point with the pen. S. The investigator will instruct the child to relax his/her left arm to the side of his/her body. 6. The investigator will pick up the skin one centimeter above the half-way point marked on the upper arm between the left thumb and forefinger. Only subcutaneous fat tissue should be he1d be- tween the fingers for measurement, not the muscle tissue. 7. The investigator will apply the calipers at the point marked which is one centimeter below the fingers which are grasping the skinfold. 8. The investigator will then continue pinching the skinfold while the right-hand fingers of the investigator will release the handle of the calipers to permit full force of the spring to be exerted on the fold. 9. The gauge on the calipers will be noted by the investiga- tor immediately following the release of the caliper's handle. results will be measured to the nearest 0.1 millimeter. The 77 10. The average of the three measurements will be calculated and recorded to the nearest 0.1 millimeter on the data collection form by the investigator (Tanner &Whitehouse, 1962, 1975). APPENDIX G SUBSCAPULAR SKINFOLD MEASURENENT PROCEDURE 79 1. The Harpenden skinfold caliper gauge or calipers will be calibrated by British Indicators Limited, the manufacturers. 2. The investigator will instruct the child to remove his/her shirt and underclothing covering the upper trunk and arms. 3. The investigator will instruct the child to relax his/her left shoulder and left arm. 4. The investigator will locate the measurement site just below the angle of the left scapula, lateral to the inferior angle and will mark this point with a felt-tipped pen. 5. The investigator will pick up the skin one centimeter above this marked pOint between the left thumb and forefinger. Only subcutaneous fat tissue should be held between the fingers for measurement, not the muscle tissue. 6. The investigator will apply the calipers at the point marked which is one centimeter below the finger's grasp of the skinfo 1d. 7. The investigator will then continue pinching the skinfold while the right-hand fingers of the investigator will release the handle to the calipers to permit full force of the spring to be exerted on the fold. 8. The gauge on the calipers will be noted by the investiga- tor three seconds following the release of the caliper's handle. The re$ults will be measured to the nearest 0.1 millimeter. 9. 10. The procedure will be repeated three times. The average of the three measurements will be calculated and recorded to the nearest 0.1 millimeter on the data collection form by the investigator (Tanner & Whitehouse. 1962.1975). APPENDIX H DATA COLLECTION FORM Code No. : Body Measurements: Ski d Thickness: Facility: Height: Tricep: Time of Procedure: Height: Scapula: Date: AVAILABLE FOOD CHOICES CALORIES PER SERVING FOODS SELECTED Total number of food items selected: Total calorie content of selected foods: Number of items chosen from Group A foods (considered high in calories--pasta, rice cereals, starchy vegetables, bread, desserts, snack chips, crackers, soft drinks, drinks, lemonade, jams, jelly, table fats, salad dressings, gravy, sauces): Number of items chosen from Group B foods (considered low in calories--milk, meat, eggs, cheese, legumes, nonstarchy vegetables, fruit, juices): APPENDIX I ARTIFICIAL FOOD LISTING AND CALORIES 83 The types of food represented by the food replicas were based on the percent of food types chosen by middle- and upper-middle-class preschool females, and studies demonstrating preferred by preschool children (Breckenridge, 1959; Bryan & Lowenburg, 1956; Jerome, 1978; Lamb & Ling, 1946; McCarthy, 1935; Vance & Temple, 1975). Description Portion Graham crackers Raisins Beef patty Pickle Sheet cake Pizza Jell 0 mol d Popcorn Light bologna Pork chop, fried Banana Chicken thigh (fried) Pumpkin pie Bread (white) Hamburger bun Vanilla ice cream French fried potatoes Salted soda crackers Whol e mil k American cheese Tuna Sweet corn on cob Peaches (canned) Tomato slice Baked potato Two Two tablespoons One Dill (medium) Yellow (without icing) One sector (5 1/4") One-half cup One cup Large (round slice) Small size Medium Medium size 1/6th pie One slice ~1ed i urn size One small scoop Medium-sized portion Three Eight oz. One sl ice (1 oz.) 1/4th cup One Two halves One One medium Calories 50 55 220 15 100 225 50 50 65 325 100 100 275 150 100 150 100 150 165 100 125 100 75 5 125 REFERENCES 85 Abraham, S. &Carrol, M. D. Dietar~ intake findings vital and health statistics, data from the natlon health survey. Hyattsville, Maryland: U.S. Department of Health, Education, and Welfare, 1977, (Series 11, No. 202). Abraham, S., Collins, G., & Norsieck, M. Relationship of childhood weight status to morbidity in adults. NDMHA Health Reports, 1971, 86, 273. Abraham, S. &Nordsieck, M. Relationship of excess weight in children and adults. Public Health Reports, 1960, ~(3), 262-273. American Academy of Pediatrics Committee of Nutrition. childhood. Pediatrics, 1967, 40, 455-467. Obesity in American Academy of Pediatrics, Committee Statement. The ten-state nutrition survey: A pediatric perspective. Pediatrics, 1973, ~(6), 1095-1099. Asher, P. Fat babies and fat children: The prognosis of obesity in the very young. Archives of Disease in childhood, 1966, iL, 674-680. Bartz, A. E. Basic statistical concepts in education and the behavioral Science. Minneapolis: Burgess Publishing Company, 1974. Bender, A. E. Food preferences of males and females. Proceedings of the Nutrition Society, 1976, ~, 181. Breckenridge, M. Food attitudes of five to twelve year old children. The Journal of the American Dietetic Association, 1959, ~, 704-709. British Indicators Limited, Sutton Road, St. Abans, Harts, England. Brozek, J., Grande, F., Anderson, J. T., & Keys, A. Densitometric analysis of body composition. Annals of the New York Academy of Science, 1963, llQ, 113-120. Bryan, M. S. & Lowenberg, M. E. The father's influence on young children's food preferences. The Journal of the American Dietetic Association, 1956, 34, 30-35. Brusch, H. Emotional aspects of obesity in children. 1975, .1., 91-101. Pediatric Annals, Cabanac, M. & Duclaux, R. Obesity of satiety, aversion to sucrose. Science, 1970, 168. 496-497. 86 Cannings, H. & Mayer, J. Obesity, its possible effect on college acceptance. The New England Journal of Medicine, 1966, 275, 11721180. Charney, E., Goodman, H. C., McBride, M., Lyon, B., & Pra~t, R. Childhood antecedents of adult obesity: Do chubby 1nfants become obese adults? The New England Journal of Medicine, 1976, 295(1), 6-9. Children's Bureau, U.S. Department of Health, Education, and Welfare. Evaluation of the nutritional status of preschool children. Washington, D.C.: U.S. Government Printing Office, 1967. Committee on Nutrition, American Academy of Pediatrics. childhood. Pediatrics, 1967, 40,455-459. Obesity in Committee on Nutrition Anthropology. Recommendations concerning body measurements for the characterization of nutritional status. Human Biology, 1956, 28(2), 111-123. Committee on Nutrition. Measurement of skinfold thickness in childhood. Pediatrics, 1968, 42, 538-554. Deck, E. Emotion, obesity, and crime. 1971 . New York: Academic Press, Edwards, D. A. W. Estimation of the proportion of fat in the body by measurement of skinfo1d thickness. The American Journal of Clinical Nutrition, 1955, ~(1), 35-36. Eid, E. E. Follow up study of physical growth of children who had excessive weight gain in first six months of life. British Medical Journal, 1970, £,74-65. Emmons, L. &Hayes, M. Nutrition knowledge of mothers and children. Journal of Nutritional Education, 1973, ~(2), 134-139. Excessive weight gain in infancy. 184-185. Nutrition Reviews, 1970, 28(7), Fomon, S. J. Nutritional disorders of children, prevention, screening, and follow-up. Rockville, Maryland: Health Services Administration, 1977, (Department of Health, Education, and Welfare Publication No. HSA 77-5104). Forbes, G. B. Lean body mass and fat in obese children. 1964, 34, 308-315. Garn, S. M. The origins of obesity. Children, 1976, 130, 465-467. Pediatrics, American Journal of Diseases in Garn~ 87 S. M. &Clark, D. C. American Academy of Pediatrics, Ad Hoc Committee to review the ten state nutrition survey, trends in fatness and origins of obesity. Pediatrics, 1976, 57(4), 443356. -- Garn, S. M., Rowen, N. N. & McCann, M. B. Relative values of different fat folds in a nutritional survey. The American Journal of Clinical Nutrition, 1971, 24, 1380-1381. Gates, J. C., Huenemann, R. L., & Brand, R. J. Food choices of obese and nonobese persons. Research, 1975, 67, 339-343. Ginsberg-Fellner, F. &Carme'l, H. A. The prevalence of obesity in nursery school children, a preliminary study. Unpublished manuscript, 1974. Goldb1oom, R. B. Obesity in childhood. Journal, 1975, Ill, 139. Canadian Medical Association GUssow, J. Counternutritional messages of television ads aimed at children. Journal of Nutritional Education, 1972, i, 48-61. Gussow, J. Using and abusing the mass media. tic Association, 1977, 38, 2-8. Journal Canadian Diete- Hampton, M. C. Nutritional intake of teenages. The Journal of the American Dietetic Association, 1967~ 50~ 385-396. Hampton, M. C., Huenemann~ R. L., Shapiro, L. R. & Mitchell, B. W. Caloric and nutrient intakes of teenagers. The Journal of the American Dietetic Association, 1967, 50~ 385-396. Hashim, S. A. Hunger and satiety in man. Nutrition, 1977, ~, 107-118. Current Concepts of Hegstead, D. M., Darby, W. J., Filer, L. J., &Shank, R. E. Comparison of body weights and lengths or heights of groups of children. Nutrition Reviews, 1974, 32(9), 302-326. Hewitt, M. I. Satiety sensations and control of body weight. Journal of the Indiana State Medical Association, 1974, 67(10), 914-918. Hinton, M. A., Eppright, E. S., Chadderdon, H., & Wo1lins, L. Eating behavior and the dietary intake of girls twelve to fourteen years old. The Journal of the American Dietetic Association, 1963, 43, 233-242. Huenemann, R. L. Environmental factors associated with preschool children. The Journal of the American Dietetic Association, 1974, 64, 480-485. Jacobson, E. The self and the obese world. of Children, 1954, ~, 75-127. 88 Psychoanalytical Study Jerome, N. W. Chi1dren ' s television advertising: A vehicle for prosocia1 nutritional messages in 1975. In E. M. Mazze (Ed.), Combined proceedings of marketing in turbulent times and marketing: The chal1enqes and the opportunities. American Marketing Association, 1975. Jerome, N. l~. & Frese, D. J. The child family preference profiles for fresh fruits and ready to eat cereals. University of Kansas, 1978. Unpublished manuscript. Johnson, M. L., Burke, B. S., &Mayer, J. The prevalence and incidence of obesity in a cross-section of elementary and secondary school children. The American Journal of Clinical Nutrition, 1956, 4, 231237. Johnson, W. G. The effect of prior taste and food visabi1ity in food directed instrument performance of the normal and the obese. Doctoral dissertation, Columbia University, 1970. Jones, H. E. The fat child. The Practitioner, 1972, 208, 212-219. Kannel, W. B., Brand, N., & Skinner, J. J. The relation of adiposity to blood pressure and development of hypertension, the Framingham study. Annals of Internal Medlcine, 1967, §I, 48-59. Keys, A. Recommendations concerning body measurements for the characterization of nutritional status. Human Biology, 1956, 28, 111. Knittle, J. L. Obesity in childhood. §l(6), 1048-1052. Journal of Pediatrics, 1972, Knittle, J. L. & Ginsberg-Fellner, F. Can obesity be prevented? Pediatric Annals, 1975, 1, 27-38. Lamb, M. W. & Ling, B. C. An analysis of food consumption and preferences of nursery school children. Children Development, 1946, l?J4), 187-217. Lerner, R. M. & Gelert, E. Body build identification, preference, and aversion in children. Developmental Psychiatry Bulletin of New York Academic Medicine, 1960, 36, 296-302. Lesser, G. T. & Zak, G. Measurement of total body fat in man by simultaneous absorption of two inert gases. Annals of the New York Academy of Science, 1963, 40, 110-113. Leverton, R. M. & Coggs, M. C. Food choices of Nebraska children. The Journal of Home Economics, 1951, l, 176-178. 89 Lewis, C. E. & Lewis, M. A. The impact of television commercials on health related beliefs and behavior of children. Pediatrics, 1974, 53(3),431-435. Litman, T. J., Cooney, J. P., & Stief, R. The views of r1innesota school children. Journal of Nutritional Education, 1964, ~, 134-140. Lloyd, J. K., Wolff, O. H., & \~helen, W. S. long term study of height and weight. 1961, £, 145-148. Childhood obesity, a British t4edical Journal, Lowenberg, M. Food preferences of young chil dren. The Journa 1 of the American Dietetic Association, 1948, 24, 430-434. Malena, R. M. Skinfold--bodyweight correlations. The American Journal of Clinical Nutrition, 1972, ~, 861-863. Management of the fat child. 961-962. British Medical Journal, 1966, (5520), Mayer, J. Obesity. Annual Review of Medicine, 1963, }!, 126-131. Mayer, J. Obesity. Postgraduate Medicine, 1972, 66-69. Mayer, J. Obesity during childhood. 1975, l, 73-80. Current Concepts of Nutrition, Mayer, J., Monello, L., & Seltzer, C. Hunger and satiety sensations in man. Postgraduate Medicine, 1965, E, 97-102. Mayer, J. & Neumann, C. G. Obesity in preschool and school-age children. Pediatric Clinics of North America, 1977, 24, 117-122. McCarthy, D. Children's feeding problems in relation to the food aversions in the family. Child Development, 1935, ~(l), 277284. Mendleson, M. Psychological aspects of obesity. North America, 1964, 48{5), 1373-1385. Medical Clinics of Metheny, N. Y., Hunt, F. E., Patton, M. B., & Heyes, H. Nutritional sufficiency findings and family marketing practices. Journal of Home Economics, 1962, 54{4), 297-303. Miller, F. J. W., Billewicz, W. Z., &Thomson, A. M. Growth from birth to adult life of 442 Newcastle upon Tyne children. Britain Journal of Preventive and Social Medicine, 1972, 26, 224228. Mobbs, J. Childhood obesity. International Journal of Nursing Studies, 1970, I, 13-18. 90 Mullins, A. G. The prognosis in juvenile obesity. sease in Chi1dhood~ 1958, 33, 307-714. Archives of Di- Nathan, S. Body image in chronically obese children as reflected in fi gure drawi ngs. \]ourna 1 of Personal i ty Assessment, 1973, R, 457-563. Nasco West, Inc. 1524 Princeton Avenue, Modesto, California, 95352. National Center fQr Health Statistics. Weight by height and age of adults, 1966, (United States 1960-1962, Series 11, No. 14). Neumann, C. G. Obesity in pediatric practice, obesity in the preschool and school-age child. Pediatric Clinics of North America, 1977, 24(l},117-119. Newens, M. E. & Goldstein, H. Height, weight and assessment of obesity in childhood. Britain Journal of Preventive and Social Medicine, 1972, 26, 33-50. Nisbett, R. E. Eating behavior and obesity in men and animal. vances in Psychosomatic Medicine, 1972, L, 173-193. (a) Ad- Nisbett, R. E. Hunger, obesity, and the ventromedial hypothalamus. Psychological Review, 1972, 79(6), 433-453. (b) Nisbett. R. E. Taste, deprivation and weight determinants of eating behavior. Journal of Personality and Social Psychology, 1968, lQ., 107-116. Nisbett, R. E. &Kanouse, D. E. Obesity, food deprivation and supermarket shopping behavior. Journal of Personality and Social Psychology, 1970,.11., 289-294. Nisbett, R. E. &Storms, M. D. Cognitive and social determinants of food intake. Chicato: Aldine, 1972. Owen, G. M. Personal communication. ember, 1978. University of Michigan, Sept- Owen, G. M.• Kram, K. M., Garry, P. J .• Lowe, J. E., & Lubin, A. H. A study of nutritional status of preschool children in the United States. 1968-1970. Pediatrics, 1974, ~(4), 597-646. Owen, G. M., Jensen, R. L., & Fomon, S. J. Sex-related difference in total body water and exchangeable chloride during infancy. Journal of Pediatrics, 1962, 60(6),858-868. Parizkova. J. Total body fat and skinfold thickness in children. Metabolism, 1961, lQ., 794-807. 91 Pate, R. R. Personal cOlTlmunication. Univeristy of Virgina. Dec .• 1978. Penick, S. B. & Stunkard, A. J. Newer concepts of obesity. Clinics of North America, 1970, 54(3), 745-754. Medical Pliner, P. Effect of liquid and solid preloads on eating behavior. Physiology and Behavior, 1973, 11. 285-290. Price, J. M. & Grinker, J. Effects of degree of obesity, food deprivation, and palatability on eating behavior of humans. Journal of Comparative Physiological Psychology, 1973, 85, 265-271. Puzynski, D. L. & Biehusen, F. C. Adrenocortical adenoma with Cushings Syndrome and virilism in a five-year-01d child. Journal of Pediatrics, 1962, 60(6), 836-840. Raugh, J. L. &Schumshy, D. A. An evaluation of triceps skinfold measures from urban school children. Human Biology, 1970, 40, 679-688. Raugh, J. L., Schumsky, D. A., &Witt, T. Height, weight, and obesity in urban school children. Child Development, 1967. 38, 515530. Ross, L. D. Cue and cognition controlled eating among obese and normal subjects. Doctoral dissertation, Columbia University, 1970. Schaeter, S. Some extraordinary facts about obese humans and rats. American Psychologist, 1971, ~, 129-144. Schaeter, S. Obesity and eating. Science, 1968, 161, 751-756. Seltzer, C. C. &Mayer, J. Simple criterion of obesity. ~4edicine, 1965, 38,101-107. Postgraduate Sims, L. S. & Morris, P. M. Nutritional status of preschoolers. The Journal of the American Dietetic Association, 1974, §i, 492-499. Shepard, R. J., Jones, G., Ishii, K., Kaneko, M. & Olbrecht. A. J. Factors affecting body density and thickness of subcutaneous fat. The American Journal of Clinical Nutrition, 1969, 22, 1175. Sloan, A. W. & Shapiro, M. Comparison of skinfold measurements. Human Biology, 1972, 44, 29-36. Smith, D. ~: & Brown, M. L. Anthropometry in preschool children in Hawall. The American Journal of Clinical Nutrition, 1970, 23(7), 932-939. 92 Stefanick, P. A., Heald, F. P., & Mayer, J. Caloric intake in relation to energy output of obese and nonobese adolescent boys. The American Journal of Clinical Nutrition, 1959, L, 55-62. Stuart, H. C. &Sobel, E. H. The thickness of the skin and subcutaneous tissue by age and sex in childhood. Journal of Pediatrics, 1946, 28(6), 631-637. Stunkard, A. Satiety is a conditioned response. cine, 1975, 37(5), 383-387. Psychosomatic Medi- Stunkard, A. & Burt, V. Obesity and body image: II age at onset of disturbance in the body image. American Journal of Psychiatry, 1967, 123, 1443-1468. Stunkard, A., d'Aquili, E., Fox, S., & Filion, R. D. L. Influence of social class on obesity and thinness in children. The Journal of the American Medical Association, 1972, 221(6), 579-584. Stunkard, A. & Mendleson, M. Obesity and the body image: I characteristics of disturbances in body image of some obese persons. American Journal of Psychiatry, 1967, 123, 1296-1336. Stunkard, A. & McLaren-Hume, M. The results of treatment for obesity. American Medical Association Archives of Internal Medicine, 1959, 103, 79-85. Tanner, J. M. The measurement of body fat in man. the Nutrition Society, 1959, ~, 148. Tanner, J. M. Personal communication. Proceedings of December 12,1978. Tanner, J. M. &Whitehouse, R. H. Standards for subcutaneous fat in British children. British Medical Journal, 1962, 1, 446-450. Tanner, J. M. &Whitehouse, R. H. Revised standards for triceps and subscapular skinfolds in British children. Archives of Disease in Childhood, 1975, 50, 142-143. Ten state nutrition survey, 1968-1970. United States Department of Health, Education, and Welfare, 1973, (No. HSH 72-8130, I-V). The overweight child. British Medical Journal, 1970, ~, 64-65. U.S. Public Health Service. Obesity and health. Washington, D.C.: U.S. Department.of Health, Education and Welfare, U.S. Government Printing Office, 1966. Vance, T. F. Food selections of nursery school children. Development, 1932,1, 169-177. Child 93 Vance, T. F. &Temple V. M. The food preferences of preschool children: A comparison of rural children with children of the Iowa State College Nursery School. Child Development, 1975, §Z.' 222-228. Walker, A. R. P. & Richardson, B. D. Lancet, 1971,.,£,1146. Overnutrition in children. Weil, W. B. Current controversies in childhood obesity. Pediatrics, 1977, iL(2) , 175-187. Weiner, J. S. & Lourie, J. A. Human biology: A guide to field methods. Internation biological program handbook no. 9. Oxford: Blackwell Scientific Publication, 1969. Wetzel, N. S. Physical fitness in terms of physique, development, and basal metabolism. Journal of American Medical Association, 1941, 116 (12), 1187-1195. Wolff, O. H. Obesity in childhood: A study of the birth weight, the height, and the onset of puberty. Quarterly Journal of Medicine, 1955, £i(94) , 109-122. Wolff, O. H. & Lloyd, J. K. 1973, 32, 195-198. Childhood obesity. Proc. Nutrition Soc., Woolsey, T. D. Skinfold thickness of children six to eleven years. Rockville, Maryland: U.S. Department of Health, Education and Welfare, 1973, (No. 73-1602). Woodhouse, S. P. Obesity as a risk factor. Medical Journal of Australia Special Supplement, 1976, 1, 11-12. Young, C. M., Moore, N. S., & Richard, S. The problem of the obese patient. The Journal of the American Dietetic Association, 1955, ll, 1115-1118.