¹Department of Pediatrics, School of Medicine, University of Texas Health Science Center, San Antonio, Texas, USA

²School of Nursing, University of Texas Health Science Center, San Antonio, Texas, USA

³Department of Computing Resources, University of Texas Health Science Center, San Antonio, Texas, USA

Correspondence to: M K Park, Department of Pediatrics, University of Texas Health Science Center, 7703 Floyd Curl Drive, San Antonio, TX 78229-7802, USA. E-mail: parkm@uthscsa.edu or parkm@uthscsa.dcci.com

OBJECTIVES: To investigate the prevalence of overweight among different ethnic and gender groups of children and adolescents in the San Antonio, Texas, area and to compare the prevalence with that of the US national figures.

DESIGN: Cross-sectional study

SUBJECTS: A total of 7208 schoolchildren in kindergarten through 12th grade. There were 4215 Mexican American (MA) (58.5%), 2040 non-Hispanic white (NHW) (28.3%) and 953 African American (AA) (13.2%) subjects.

MEASUREMENTS: Weight, height and skinfold thicknesses.

RESULTS: The body mass index (BMI, kg/m²) values of MA boys were almost consistently and significantly (P<0.05) larger than NHW boys and showed a tendency to be larger than AA boys, beginning as early as age 6 and continuing through age 17. Although rarely significant, a similar trend in ethnic difference was also noted for girls, with the smallest BMI seen in NHW girls. The subscapular skinfold thickness (SST) for MA boys and girls was significantly (P<0.05) larger than that for NHW counterparts and showed a tendency to be larger than AA counterparts. No significant ethnic differences were present in the triceps skinfold thickness (TST) for girls, but MA boys' TST were occasionally larger (P<0.05) than other ethnic-gender groups. Girls' TST were frequently larger (P<0.05) than boys for each ethnic groups. Using the population data from the National Health and Nutrition Examination Survey (NHANES) I as reference, the prevalence of overweight (BMI95th percentile) was greater in MA (15-28%) and AA (11-29%) boys and girls than in NHW (7-17%) counterparts. The combined prevalence of overweight and 'at risk of overweight' (BMI>85th percentile) was much larger in MA boys (40-50%), MA girls (34-52%), and AA girls (33-51%) than other subgroups. The onset of overweight is quite early, starting at 5-6 y of age, especially in girls. Compared to the data from national surveys, the prevalence of overweight found in this study is higher than reported nationally. We found a marked increase in the skinfold thickness, especially SST for boys, but the increase is less for girls.

CONCLUSIONS: The prevalence of overweight is higher in MA boys and girls and AA girls than other ethnic-gender groups in the San Antonio, Texas, area. The prevalence of childhood overweight in the San Antonio area is higher than national figures. The findings of increasing prevalence and early onset of childhood overweight are concerning, because these are known risk factors for diabetes and diseases of many other organ systems. Measures to prevent, reduce or treat childhood obesity are urgently needed.

International Journal of Obesity (2001) 25, 409-416

body mass index; ethnicity/race; Mexican Americans; African Americans; obesity/overweight; skinfold thickness

Introduction

The increasing prevalence of obesity in the last two decades in adults^1,2 as well as children and adolescents^3,4,5,6 is a national concern. Although ethnic differences exist, the increase in the prevalence of overweight is seen across ethnic and gender groups.^2,4,5,6,7 Obesity is more prevalent in Mexican American (MA) and African American (AA) adults^1,8,9,10 and children^5,8,11,12 than non-Hispanic white (NHW) counterpart. Ethnic differences are also found with fat patterning. Central (truncal) fat distribution is strongly associated with hyperinsulinemia and type 2 diabetes as well as other cardiovascular (CV) risk factors in adults^2,13,14 and children.^15,16,17 This pattern of fat distribution is more prevalent in AA and MA adults and children relative to NHW.^18,19

Because blood pressure levels are known to be correlated with body size,^20,21 anthropometric measurements were part of the San Antonio Children's Blood Pressure Study, which was conducted in children kindergarten through 12th grade from the San Antonio, Texas, area, during the periods of 1991-1995 and 1996-1998. Although the national trends of ethnic differences in the prevalence of obesity are known, the same information is lacking for this region in which Mexican Americans are the majority. The purposes of this portion of the report were (1) to examine the prevalence of overweight for each ethnic and gender group in a triethnic pediatric population, (2) to examine whether ethnic differences seen in adults are present in children, and (3) to compare the prevalence of obesity in this region with that of national surveys.

Methods

(A) Subject population and recruitment

This study was carried out in school children enrolled in kindergarten through 12th grade, in three selected school districts in the San Antonio, Texas, area. The initial phase of the San Antonio Children's Blood Pressure Study began in the fall of 1991 as a 4-y study (until 1995) to collect data from a biethnic population of MA and NHW children and adolescents. This phase of the study involved the South San Antonio Independent School District (ISD) and the Schertz-Cibolo-Universal City ISD. For a triethnic comparison, the study was extended in 1996 (to 1998) to include the San Antonio ISD to collect data primarily from African American (AA) children. The Schertz-Cibolo-Universal City ISD is a suburb of San Antonio with the majority of its students being non-Hispanic white (66.0%). The student body of the South San Antonio ISD is mostly Mexican American (94.0%). All schools in these two ISD were studied. There were four elementary schools, two middle schools, and one high school in the Schertz-Cibolo-Universal City ISD and 10 elementary schools, three middle schools, and two high schools in the South San Antonio ISD. Only three elementary schools, two middle schools and one high school were selected from the San Antonio ISD to recruit primarily African American children. Although the socioeconomic status was not clearly evaluated, participation in the school lunch program is an indicator of the socioeconomic status. Free school lunch program participation in each school district was as follows: (1) South San Antonio ISD¾95%; (2) Schertz-Cibolo-Universal City ISD¾25.9%; and (3) San Antonio ISD¾100%.

The School Boards of the three ISDs agreed to allow data collection in their schools. Parental consent was obtained and the child was also asked for his/her assent before data were collected. The letter and the consent forms were in English and Spanish. The Institutional Review Board of the University of Texas Health Science Center at San Antonio approved the research protocol.

Ethnic classification for African American (AA) children was made by the staff observer. The classification of Mexican American (MA), non-Hispanic white (NHW), non-Mexican Hispanics, and others was based on a previously published algorithm because MAs have different genetic makeup from other Hispanics.²² Scoring of the algorithm was done by computer so that the actual classification of children as MA, NHW or other was independent of observers' judgment.

Weight and height were measured to calculate BMI. Weight was measured with a scale calibrated daily with standard weights. Weight without shoes or heavy garments was measured to the nearest 0.25 pound and converted to kg later. Height without shoes was measured with the child standing straight against a 3 m rule taped to the wall and squared at the base with the floor. With the child looking straight ahead at an indicated point on the opposite wall, and heels, buttocks and occiput touching the wall, the edge of a plastic right triangle was positioned perpendicular to the meter rule to be touching the top of the head, with hair pushed down. The height was measured to the nearest 0.1 cm.

Triceps skinfold thickness was measured over the back of the right upper arm at the midpoint with the arm pendant. A full thickness pinch of skin was grasped horizontally 1 cm above the midpoint, and the skinfold thickness was read immediately after applying Lange calipers (square tipped). The calipers were released and the measurement recorded to the nearest 1 mm. Subscapular skinfold thickness was measured by pinching a full thickness of skin below the caudal end of the right scapula in the back, using the Lange calipers in the same manner as above. The Lange calipers were checked monthly for tension and measurement accuracy.

Data so collected were entered into a database. Statistical analyses were performed using the SAS Software (SAS Institute Inc., Cary, NC) by the Computing Resources Department of the University of Texas Health Science Center at San Antonio, Texas. Children whose ethnic group was other then AA, MA and NHW were excluded from the statistical analyses, although they were included in data collection as required by the school districts. The remaining children were divided into six subgroups by ethnicity and gender. The mean and standard deviation (s.d.) of BMI and skinfold thicknesses were computed for each ethnic-gender group. Statistical tests for significant ethnic and gender differences in these variables were performed by Analysis of Variance followed by the Bonferroni test for each age group (age at last birthday in years). Statistical significance was accepted at the 5% level in this study. The prevalence of overweight in this population was computed using the reference data of BMI from the NHANES I.^23,24 BMI and skinfold thickness data from this study were compared with the results of national surveys.

Results

Data from 1006 subjects were excluded from analyses because of ethnicity or race other than AA, MA and NHW (n=612), ages older than 17 y (n=299), and incomplete data (n=95). Data from a total of 7208 children ranging in age from 5 to 17 y are included in the statistical analysis (Table 1). The number of MA children was much larger (58.5%) than that of NHW (28.3%) and AA (13.2%) children. The over-sampling of MA children could not be avoided because the school district policy required that every child who wanted to participate be screened. Despite our best efforts, a disproportionately small number of AA children participated in the study. The participation rate was 74% for the Schertz-Cibolo-Universal City ISD, 66% for the South San Antonio ISD, and 49% for the selected schools in the San Antonio ISD.

Body mass index

The BMI values of MA girls were almost always significantly (P<0.05) larger than those of NHW girls (Table 2). The BMI values of MA and AA girls were not significantly different. MA boys had consistently and significantly (P<0.05) larger BMI values than did NHW boys throughout the ages studied (Table 2). MA boys showed a strong tendency to have larger BMI values than AA boys, with statistical significance attained frequently. No significant gender difference was found for BMI (Table 2). Overall, the BMI values of NHW girls were the smallest and those of MA boys the largest.

Skinfold thicknesses

The subscapular skinfold thickness (SST) values of MA girls and boys were almost consistently and significantly (P<0.05) larger than those of NHW counterparts and showed a tendency to be larger than those of AA boys and girls (Table 3). One exception to this was the SST value of AA girls ages 15-17 y which was significantly larger (P<0.05) than that of MA and NHW girls. With regards to the triceps skinfold thickness (TST), no significant ethnic differences were present for girls, although MA girls tended to have the larger values than other ethnic groups (Table 4). MA boys had the largest TST, with occasional statistical significance attained. The TST of girls were often significantly (P<0.05) larger than those of boys in all three ethnic groups (Table 4).

Prevalence of overweight by BMI

Figures 1 and 2 show the prevalence of overweight in the triethnic pediatric population of the San Antonio, Texas, area compared with the age- and gender-specific 85th and 95th percentile cut-off points of the NHANES I population data.^23,24 Overweight in children and adolescents is defined as those with BMI values at or greater than the 95th percentile of the reference value.²⁵ Those children whose BMI is between the 85th percentile and the 94th percentile are defined as 'at risk of overweight'.²⁵ By this definition, the expected prevalence of overweight in a reference population should be 5% and that of 'at risk' should be 10%.

For girls, the combined prevalence of overweight and 'at risk' is much higher in AA and MA girls than NHW girls (Figure 1). The combined prevalence ranges between 33% and 51% for AA girls, between 34% and 52% for MA girls, and between 22% and 38% for NHW girls. The prevalence of overweight (BMI>95th percentile) is much lower in NHW girls (8-17%) than the other ethnic groups (15-29% for AA girls and 15-27% for MA girls). The prevalence of 'at risk of overweight' in NHW adolescent girls (13-22%) tends to be lower than that in the other ethnic groups (Figure 1). An early onset of marked overweight occurring in AA and MA girls at ages 5 and 6 is noteworthy.

For boys, the combined prevalence of overweight and 'at risk' is highest in MA boys (40-50%), with almost half of adolescents having BMI values greater than 85th percentile of the reference value.^23,24 The combined prevalence is lower for AA boys (23-36%) and NHW boys (22-37%) (Figure 2). Note that the onset of overweight is quite early in boys as well as in girls. MA boys have the highest prevalence of overweight (BMI>95th percentile) with the ranges between 21 and 28%. Among boys, the lowest prevalence of overweight is seen in NHW subjects (range of 7-16%). For AA boys, the prevalence of overweight ranges between 11 and 27%. The overall prevalence of 'at risk' ranges between 7% and 26% (Figure 2). AA boys had a lower prevalence of 'at risk' than the other ethnic groups.

Comparisons with national figures

In Table 5, the prevalence reported by Troiano et al⁵ and that found in this study are compared. For both studies, the reference values used (National Health Examination Survey (HES) cycles II and III) and the age breakdown are the same. The prevalence of overweight and 'at risk of overweight' in this study is much higher than that reported by Troiano et al⁵ from NHANES III. In all, 35-47% of MA boys and girls and AA girls have BMI greater than the 85th percentile in this study, in comparison to the figures reported by Troiano et al from the NHANES III (between 20 and 33%).⁵ The difference in the prevalence between these two studies is striking. The percentage difference in the prevalence of BMI>85th percentile (the combined prevalence of overweight and 'at risk') is on average 134%, with a high of 180%. The percentage difference for the prevalence of overweight (BMI>95th percentile) is on average 148%, with a high of more than 200% (Table 5).

Changes in skinfold thicknesses

The mean values of the SST in this study were compared with the race-specific 50th percentile values from NHANES I data for AA and NHW boys²⁶ and AA and NHW girls.²⁷ The reference values for MA children were from the Hispanic Health and Nutrition Examination Survey (HHANES),²⁸ since the ethnicity-specific MA data are not available from NHANES I. The increase in the SST was modest in girls of 6-11 y and less in older girls (Table 6). There was almost no increase in the SST in NHW girls aged 12-17 y. AA and MA girls showed much larger percentage increases than NHW girls. The SST was markedly increased in boys of all ages, with a larger increase seen in AA and MA boys than in NHW boys (Table 6). The largest increase (70-128%) is seen in AA boys. In comparison with 50th percentile values for TST from NHANES I (1971-1974) for AA and NHW children²⁶ and from HHANES (1982-1984),²⁸ for MA children, there was a small percentage increase in the TST in girls 6-11 y for all ethnic groups. However, for MA and NHW girls aged 12-17, there was an actual percentage decrease (13-20%) in this measurement. In AA adolescent girls the value increased (not shown). For boys, there was a modest increase (9-48%) in the percentage change in all ages studied. The increase was most marked in AA boys and least so in NHW boys.

Discussion

Our data show (a) that an increase in the prevalence of overweight is seen in all ethnic and gender groups, as has been reported nationally,^6,12 (b) that there are several important ethnic differences in BMI and skinfold thickness measurements in children and adolescents, and (c) that the magnitude of the increase in the prevalence is greater in this study than has been reported nationally.

MA boys and girls often have larger BMI values than NHW counterparts while AA children, especially girls, have BMI values equally as large as MA girls. Data from NHANES II,²⁹ NHANES III,⁶ and HHANES²⁸ show the same ethnic differences in BMI as noted above among the three ethnic groups of children. These three national surveys^28,29,30 also show the same pattern of ethnic differences in BMI among the three ethnic groups of adults. With regard to the SST, MA children tended to have larger values than other ethnic groups and NHW girls had the smallest SST measurements. NHANES II²⁹ and HHANES²⁸ surveys also showed the same pattern of ethnic differences in the SST in adults as found in this study, with white women having the lowest SST. Thus, the results of our study show that the ethnic differences in BMI and the fat distribution pattern reported in the adult are already established as early as 5-6 y of age.

There has been a dramatic (8%) increase in the prevalence of overweight in the last two decades, more precisely between the period of 1976-1980 (NHANES II) and 1988-1991 (NHANES III, phase I).¹ Flegal et al³⁰ reported the prevalence of obesity (defined as body mass index (BMI, kg/m²)30) to be 22% of men and 24.9% of women in the NHANES III (1988-1994). The prevalence was higher in Mexican American (MA) men (25.4%) than white (20%) and African American (AA) (21.3%) men and it was higher in MA women (34.2%) and AA women (30%) than white women (22%). Parallel with the increase in prevalence of obesity in adults,¹ an alarming increase in the prevalence of childhood obesity has occurred during the same period in the USA.^3,4,6 Troiano et al,⁵ reported the prevalence of obesity (called overweight in pediatrics, BMI95th percentile) from the NHANES III study (1988-1994) to be 10-12% for NHW and AA boys and 15-17% for MA boys. It was 14-16% for AA and MA girls and 9% for white girls,⁶ showing the same ethnic trend as reported in the adults.³⁰ Thus, the trends in ethnic difference are the same for adults and children in the NHANES III^5,30 and in our triethnic pediatric population in the San Antonio area.

This rapid increase in the prevalence of obesity is concerning because many obese children become obese adults^31,32 and, once established, obesity is difficult to treat. It becomes a chronic condition with complications of many organ systems,² including cardiovascular diseases, type 2 diabetes mellitus, gallbladder disease, osteoarthritis, sleep apnea, and certain types of cancer.^2,33,34 None of the treatment modalities available will cure the condition.³³ At best, the condition is palliated on continued treatment at an enormous cost. In 1995, the total cost attributable to obesity amounted to $99.2 billion.³⁵ Approximately $51.6 billion of these dollars (5.7% of the national health expenditure) were direct medical costs.³⁵

Unlike adults, obesity-related diseases are rarely manifested in children and adolescents. However, some of these children have already shown signs of chronic complications of obesity, such as type 2 diabetes, usually not seen until middle age.³⁶ Many of these children may actually have high blood pressure, high cholesterol, and other diseases in their early stages,^17,37,38,39 and are at risk for psychosocial dysfunction.⁴⁰ Many diseases, including coronary artery disease, are silent during childhood and adolescence, but there is strong evidence that coronary artery 'clogging' is going on during childhood.⁴¹ The early onset of overweight seen in this study is also concerning. The onset of overweight may be even earlier in the preschool-age group, which we did not study. Reports from the US⁴² and the UK⁴³ indicate that some preschool children are already obese. Hernandez et al⁴² reported 32% of children in a Head Start program to be overweight. Reilly et al⁴³ reported that 20% of 4-y-old children in the UK were overweight. Therefore, the prevention and/or remediation of obesity is an important measure that could prevent cardiovascular diseases, hypertension³² diabetes³⁶ and other diseases in youth and thus should be started in childhood.^41,44 Interventions during childhood are more likely to have long-term success.⁴⁴

The reasons for the higher prevalence of overweight in this area are not entirely clear. It may be a local phenomenon, as San Antonio is dubiously known as the third 'fattest' city in the country.⁴⁵ It may also reflect a continued increase in the prevalence of overweight since the completion of the NHANES III (1991-1994), but this is unlikely at least for MA and NHW children because our data collection was only 1 y later than the national survey. Most of the data on AA children were, however, obtained 4-5 y later (in 1996 to 1998) than the NHANES III and may reflect the continued increase in obesity rather than the local phenomenon. Thus, the reasons for the higher prevalence of obesity in this region may be a combination of the two possibilities.

The strength of this study is that it was carried out in a triethnic population by a small number of investigators, so that a direct comparison of anthropometric data is possible among the three ethnic groups. The weaknesses of the study includes (a) a disproportionately small number of AA children participated and an over-sampling of MA children (59%) was unavoidable because of the required policy of School Districts to collect data from all volunteering students, (b) the participation rate was less than optimal, (c) data collection from predominantly AA students lagged 2 y, (d) detailed information on the socioeconomic status or the level of acculturation are lacking, and (e) no information is available on the difference between participants and non-participants. Small numbers of AA subjects in the study may have resulted in a less than ideal situation for statistical analyses for this particular ethnic group. The socioeconomic and acculturation status would undoubtedly have contributed to the ethnic difference shown in this study.^46,47,48 The prevalence of obesity and its ethnic differences obtained in this study may not be appropriate for other regions.

In conclusion, we found in this study an early onset of childhood obesity and an early establishment of the adult pattern of ethnic differences in BMI and central fat distribution. The reports of childhood onset of type 2 diabetes³⁶ and cardiovascular risk factors^17,49 are particular concerns. The prevention and/or remediation of overweight is an important measure that could prevent hypertension,^32,50 diabetes,³⁶ and other diseases in young adults and these efforts should be started in childhood.⁴¹

Acknowledgements

The authors wish to thank the school board members, superintendents, school teachers, and nursing staff of the South San Antonio Independent School District (ISD), the Schertz-Cibolo-Universal City ISD, and the San Antonio ISD for their enthusiastic support for the project. We wish to thank all staff members of the San Antonio Triethnic Children's Blood Pressure Study for their dedication to the project.

This study was supported by grant MCH-480612 and MCH-480747 from the Maternal and Child Health Program (Title V, Social Security Act), Health Resources and Services Administration, Department of Health and Human Services.

1 Kuczmarski RJ, Flegal KM, Campbell SM, Johnson CL. Increasing prevalence of overweight among US adults. The National Health and Nutrition Surveys, 1960-1991. JAMA 1994; 272: 205-211, MEDLINE

2 Pi-Sunyer FX. Medical Hazards of Obesity. JAMA 1993; 119: 655-660,

3 Freeman DS, Srinivasan SR, Valdez RA, Williamson DF, Berenson GS. Secular increase in relative weight and adiposity among children over two decades: the Bogalusa Heart Study. Pediatrics 1997; 99: 420-426, MEDLINE

4 Gortmaker SL, Dietz WH, Sobol AM, Wehler CA. Increasing pediatric obesity in the United States. Am J Dis Child 1987; 141: 535-540, MEDLINE

5 Troiano RP, Flegal KM, Kuczmarski RJ, Campbell SM, Johnson CL. Overweight prevalence and trends for children and adolescents: the National Health and Nutrition Surveys, 1963-1991. Arch Pediatr Adolesc Med 1995; 149: 1085-1091, MEDLINE

6 Troiano RP, Flegal KM. Overweight children and adolescents: description, epidemiology, and demographics. Pediatrics 1998; 101: 525-539,

7 Malina RM, Zavaleta AN, Little BB. Body size, fatness and leanness of Mexican American children in Brownsville, Texas: Changes between 1972 and 1983. Am J Public Health 1987; 77: 573-577, MEDLINE

8 Update: prevalence of overweight among children, adolescents, and adults¾United States, 1988-1994. Morb Mort Wkly Rep 1997; 46: 199-202,

9 Haffner SM, Stem MP, Hazuda H, Rosenthal M, Knapp A. The role of behavioral variables and fat patterning in explaining ethnic differences in serum lipids and lipoproteins. Am J Epidemiol 1986; 123: 830-839, MEDLINE

10 Stevens J, Keil JE, Rust PF, Tyroler HA, Davis CE, Gazes PC. Body mass index and body girths as predictors of mortality in black and white women. Arch Intern Med 1992; 152: 1257-1262, MEDLINE

11 Winkleby MA, Robinson TN, Sundquist J, Kraemer HC. Ethnic variation in cardiovascular disease risk factors among children and young adults. JAMA 1999; 281: 1006-1013, MEDLINE

12 Martorell R, Mendosa FS, Castillo RO. Genetic and environmental determinants of growth in Mexican-Americans. Pediatrics 1989; 84: 864-871, MEDLINE

13 Kissebah AH. Central obesity: measurement and metabolic effects. Diabetes Rev 1997; 5: 8-20,

14 Reaven GM. Role of insulin resistance in human disease. Diabetes 1988; 37: 1595-1609, MEDLINE

15 Brambilla P, Manzoni P, Sironi S et al. Peripheral and abdominal adiposity in childhood obesity. Int J Obes Relat Metab Disord 1994; 18: 795-800, MEDLINE

16 Daniels SR, Morrison JA, Sprecher DL, Khoury P, Kimball TR. Association of body fat distribution and cardiovascular risk factors in children and adolescents. Circulation 1999; 99: 541-545, MEDLINE

17 Dietz WH. Health consequences of obesity in youth: childhood predictors of adult disease. Pediatrics 1998; 101: 518-525,

18 Mueller WH. Ethnic differences in fat distribution during growth. Liss: New York, 1988,

19 Greaves KA, Puhl J, Baranowski T, Gruben D, Seale D. Ethnic differences in anthropometric characteristics of young children and their parents. Hum Biol 1989; 61: 459-477, MEDLINE

20 Report of the Second Task Force on Blood Pressure Control in Children¾1987. Pediatrics 1987; 79: 1-25,

21 Lauer RM, Burns TL, Clarke WR. Assessing children's blood pressure¾consideration of age and body size: the Muscatine study. Pediatrics 1985; 75: 1081-1090, MEDLINE

22 Hazuda HP, Comeaux PJ, Stern MP, Haffner SM, Rosenthal M, Franco LJ. A comparison of three indicators for identifying Mexican-Americans in epidemiologic research. Am J Epidemiol 1986; 123: 96-112, MEDLINE

23 Must MF, Dallal GE, Dietz WH. Referencer data for obesity: 85th and 95th percentiles of body mass index (wt/ht²) and triceps skinfold thickness. Am J Clin Nutr 1991; 53: 839-846, MEDLINE

24 Must A, Dallal GE, Dietz WH. Reference data for obesity: 85th and 95th percentiles of body mass index (wt/ht²)¾a correction. Am J Clin Nutr 1991; 54: 773,

25 Barlow SE, Dietz WH. Obesity evaluation and treatment: expert Committee Recommendations. Pediatrics 1998; 102: e29, MEDLINE

26 Cronk CE, Roche AF. Race- and sex-specific reference data for triceps and subscapular skinfold and weight/height². Am J Clin Nutr 1982; 35: 347-354, MEDLINE

27 Abrahams S, Lawenstein FW, O'Connell DE. Preliminary findings to the first Health and Nutrition Survey, United States, 1971-1972: anthropometric and clinical findings. National Center for Health Statistics: Washington, DC, 1975,

28 National Center for Health Statistics. Anthropometric data and prevalence of overweight for Hispanics: 1982-84. DHHS Publication no. (PHS) 89-1689, 1989,

29 Najjar MF, Rowland M. Anthropometric reference data and prevalence of overweight, United States, 1976-80. DHHS Publication no. (PHS) 87-1688, 1987,

30 Flegal KM, Carroll MD, Kuczmarski RJ. Overweight and obesity in the United States: prevalence and trends. Int J Obes Relat Metab Disord 1998; 22: 39-47, MEDLINE

31 Serdula MK, Ivery D, Coates RJ, Freedman DS, Williamson DF, Byers T. Do obese children become obese adults? A review of the literature. Prev Med 1993; 22: 167-177, Article MEDLINE

32 Must A, Jacques PF, Dallal LM, Bajema CJ, Dietz WH. Long-term morbidity and mortality of overweight adolescents. A follow-up of the Harvard growth study of 1922 to 1935. New Engl J Med 1992; 327: 1350-1355, MEDLINE

33 Bray GA. Obesity. In: Harrison's Principle of Internal Medicine 14th edn. McGraw-Hill: New York, 1998,

34 Kaplan NM. The deadly quartet. Upper-body obesity, glucose intolerance, hypertriglyceridemia, and hypertension. Arch Intern Med 1989; 149: 1514-1520, MEDLINE

35 Wolf AM, Colditz GA. Current estimates of the economic costs of obesity in the United States. Obes Res 1998; 6: 97-106, MEDLINE

36 Pinhas-Hamiel O, Dolan LM, Daniels SR, Standiford D, Khoury PR, Zeitler P. Increased incidence of non-insulin-dependent diabetes mellitus among adolescents. J Pediatr 1996; 128: 608-615, MEDLINE

37 Webber LS, Harsha DW, Phillips GT, Srinivasan SR, Simon JW, Berenson GS. Cardiovascular risk factors in Hispanic, white and black children: the Brooks County and Bogalusa Heart Studies. Am J Epidemiol 1991; 133: 704-714, MEDLINE

38 Steinberger J, Moorhead C, Katch V, Rocchini AP. Relationship between insulin resistance and abnormal lipid profile in obese adolescents. J Pediatr 1995; 126: 690-695, MEDLINE

39 Berenson GS, Radhakishnamujrthy B, Srinivasan RS et al. Plasma glucose and insulin levels in relation to cardiovascular risk factors in children from a biracial population¾the Bogalusa Heart Study. J Chronic Dis 1981; 34: 379-391, MEDLINE

40 Monello LF, Mayer J. Obese adolescenct girls: an interdisciplinary study of adolescent obesity. J Pediatr 1972; 13: 35-39,

41 Berenson GS, Wattigney MS, Bao W, Srinivasan SR, Radhakrishnamurthy B. Rationale to study the early natural history of heart disease: the Bogalusa Heart Study. Am J Med Sci 1995; 310: S22-S28, MEDLINE

42 Hernandez B, Tphold CR, Graham MV, Singer L. Prevalence and correlates of obesity in preschool children. J Pediatr Nurs 1998; 13: 68-76, MEDLINE

43 Reilly JJ, Dorosty AR, Emmett PM. Prevalence of overweight and obesity in British children: cohort study. Br Med J 1999; 319: 1039-1041,

44 Epstein LH, Volaski A, Wing RR, McCurley J. Ten-year follow-up of behavioral, family based treatment for obese children. JAMA 1990; 264: 2519-2523, MEDLINE

45 Mapping US obesity. Tufts Univ Health Nutr Lett 1997; 15: 1,

46 Mei Z, Scanlon KS, Grummer-Strawn LM, Freedman DS, Yip R, Trowdridge FL. Increasing prevalence of overweight among US low-income preschool children: the Center for Disease Control and Prevention pediatric nutrition surveillance, 1983 to 1995. Pediatrics 1998; 101: E12, MEDLINE

47 Brown PJ. Cultural perspectives on the etiology and treatment of obesity. 2nd edn. Raven Press: New York, 1993,

48 Rosner B, Prineas R, Loggie J, Daniels SR. Percentiles for body mass index in US children 5 to 17 years of age. J Pediatr 1998; 132: 211-222, MEDLINE

49 Shear CL, Freeman DS, Burke GL, Harsha DW, Berenson GS. Body fat patterning and blood pressure in children and young adults. The Bogalusa Heart Study. Hypertension 1987; 9: 236-244, MEDLINE

50 Diehl AK, Stern MP. Special health problems of Mexican-Americans: obesity, gallbladder disease, diabetes mellitus, and cardiovascular disease. Adv Intern Med 1989; 34: 73-96, MEDLINE

Figure 1 Prevalence of overweight and 'at risk of overweight' for girls. The current data were compared with the population reference data of the National Health and Nutrition Examination Survey I (NHANES I) (1971-1974).

Figure 2 Prevalence of overweight and 'at risk of overweight' for boys. The current data were compared with the population reference data of the NHANES I (1971-1974).

Table 1 Number of study subjects according to age groups, ethnicity and gender

Table 2 BMI (kg/m²) (mean±s.d.) according to ages, ethnicity and gender, with results of statistical tests

Table 3 Subscapular skinfold thickness (mm) (mean±s.d.) according to ages, ethnicity and gender, with results of statistical tests

Table 4 Triceps skinfold thickness (mm) (mean±s.d.) according to ages, ethnicity and gender, with results of statistical tests

Table 5 Comparison of prevalence of overweight (%) between NHANES III and the present study (with reference data from NHES cycles II and III)

Table 6 Changing trends in the mean values of the subscapular skinfold thickness (%)