Obesity and relative subcutaneous fat distribution among Canadians of First Nation and European ancestry

Transcription

1 Short Communication International Journal of Obesity (1998) 22, 1127±1131 ß 1998 Stockton Press All rights reserved 0307±0565/98 $ Obesity and relative subcutaneous fat distribution among Canadians of First Nation and European ancestry PT Katzmarzyk *1 and RM Malina 2 1 Department of Kinesiology and Health Science, York University, North York, Ontario, Canada; and 2 Institute for the Study of Sports, 213 IM Sports Circle, Michigan State University, East Lansing, MI, USA OBJECTIVE: To compare Canadians of Aboriginal (First Nation, ) and European ancestry (), with respect to obesity, subcutaneous fatness and relative subcutaneous fat distribution. DESIGN: Cross-sectional comparison. SUBJECTS: 118 First Nation and 472 European ancestry Canadians from Northern Ontario, youth (5 ±19 y) and adults (20±75 y). MSUREMENTS: Stature, mass and skinfold thicknesses at the triceps, biceps, medial calf, subscapular, suprailiac and abdominal sites. ANALYSIS: Prevalence of obesity was determined using the 85 th percentile of body mass index (BMI) from NHANES II as the cut-off. Principal components (PC) analyses were performed on the six skinfolds. The rst component (PC1) represented a trunk-extremity skinfold contrast. Differences in mean component scores between and were assessed using independent samples t-tests. RESULTS: Prevalence of obesity among was high, ranging from 29% in youth to 60% in adult females. demonstrated greater subcutaneous adiposity and greater values for PC1, with the exception of adult males, where the difference is in the expected direction, but is not signi cant. The direction of the differences indicates that the have a greater centralization of subcutaneous fat. CONCLUSIONS: Canadians generally have a greater prevalence of obesity, greater subcutaneous fatness and a more centralized distribution of body fat than those of European ancestry. Keywords: obesity; central fat distribution; adaptation; regional adiposity; Aboriginal; Native American Introduction Obesity and a central distribution of body fat are important risk factors for disease. 1±4 However, relative fat distribution among Native North Americans, a group with a high prevalence of metabolic diseases, has been under studied. The health of Native North Americans is generally poor, relative to the general population. A syndrome of diseases which are prevalent among native groups has been identi ed and has been called the `New World Syndrome'. This syndrome, which includes metabolic diseases such as obesity, diabetes and gall bladder disease, is presumed to have a signi cant genetic component. 5±7 The cause of high prevalence rates for metabolic diseases in North American Aboriginals is probably multifactorial and a better understanding of risk factors such as obesity, relative fat distribution and * Correspondence: Peter T. Katzmarzyk, Department of Kinesiology and Health Science, 352 Bethune College, York University, 4700 Keele Street, North York, Ontario M3J 1P3, Canada. Received 23 December 1997; revised 9 April 1998; accepted 18 June 1998 physique, may be important in understanding the etiology of the New World Syndrome. Relatively little data are available on subcutaneous fat distribution and physique in First Nation Canadians; this limits discussions of disease and risk factor prevention in these groups. 8 The purpose of this study is to compare Canadians of First Nation and European ancestry in terms of subcutaneous fatness, prevalence of obesity, and relative subcutaneous fat distribution. Methods Sample Data were collected during the spring and summer of 1996 (May±August) in the Northern Ontario town of Temagami and the First Nation community of Bear Island. All healthy individuals aged 5±75 y were eligible to participate. Permission to undertake the study was obtained from the Township Council of Temagami, the Temagami First Nation, the Timiskaming Board of Education and the University Committee for Research Involving Human Subjects (UCRIHS) at Michigan State University. Participants were recruited via advertising, using signs placed around the town, by telephone, door

2 1128 to door visits and letters sent home from the principal of the public school (for children aged 5±15 y). A total of 590 subjects, 118 First Nation () and 472 European ancestry () participated in the study. The sample represented approximately 50% of the total population living in the area. The response rate for the study could not be determined due to the variety of recruitment strategies employed. Subjects were assigned to either the or group based on self-ascribed ethnic status. Measures Several anthropometric dimensions were taken with standard techniques. 9 Stature was measured using a eld anthropometer (GPM, Seritex Inc., Carlstadt, NJ) to the nearest mm with the subject bare footed, standing on a at surface, with mass evenly distributed on both feet, whereas body mass was assessed to the nearest 0.2 kg using a spring scale (Medixact Proshape, Sunbeam-Oster, Schaumburg, IL) resting on a hard at surface. Skinfolds at the biceps, triceps, medial calf, subscapular, suprailiac and abdominal sites were measured with Holtain (Holtain LTD, Crymych, UK) calipers to the nearest 0.2 mm. Derived indices included the body mass index (BMI, mass (kg)=stature (m) 2 ), the sum of skinfolds (SUM, subscapular abdominal supra-iliac biceps triceps medial calf), and the trunk-to-extremity skinfold ratio (TER, subscapular ab-dominal suprailiac= biceps triceps medial calf). All measurements were taken by a single observer (PTK). Measurement variability and reliability were assessed using the technical error of measurement (TEM) for replicate measurements taken on a sample of 64 subjects. Replicate measurements were made at least one day apart and no more than two months apart (mean 10 d). The TEMs for stature (0.54 cm), body mass (0.72 kg) and the skinfolds (range 0.94 for triceps to 1.64 for abdominal), are comparable to the major national surveys in the US. 10±13 Statistical analysis The sample was divided into youth (5±19 y) and adults (20±75 y) for the analysis. Prevalence rates for obesity were determined using the BMI as the criterion. For youth, the cut-off was the age- and gender-speci c 85 th percentiles of National Health and Examination Survey (NHANES) II data. 14 For adults, the cut-off was the 85 th percentile NHANES II data for 20±29 year olds (BMI 27.8 in males and BMI 27.3 in females). Principal components analysis (PCA) was applied to the skinfolds following the procedures described by Deutsch et al 15 and Baumgartner et al. 16 Skinfolds were transformed to natural logarithms and adjusted for the effects of age and the mean skinfold of the individual by applying the following regression: Y ˆ age age 2 age 3 meansk, within each gender and age group. The standardized residuals were retained for the PCA, to represent the skinfold adjusted for the effects of age and subcutaneous fatness. PCA was applied to the skinfold residuals within each gender and age group separately. The resulting principal component scores represent indices of subcutaneous fat distribution which are independent of overall fatness. Generally, the rst principal component (PC1) indicates a trunk-extremity contrast, whereas the second component contrasts upper and lower body skinfolds, although there may be differences in the interpretation of the later component among samples. 16,17 Independent samples t-tests were used to determine differences in PC1 between and. All analyses were performed using SPSS procedures. 18 Results The physical characteristics of the sample are described in Table 1. There were few gender differences; however, adult males had greater TERs than Table 1 Physical characteristics of the sample Age (y) Stature (cm) Mass (kg) BMI (kg=m 2 ) SUM (mm) TER (mm=mm) n Mean s.d. Mean s.d. Mean s.d. Mean s.d. Mean s.d. Mean s.d. Males Females * Males ** ** 0.36 Females ** 0.26 Males Females * 0.36 Males ** ** Females ** * *, ** *, ** *, ** 0.35 ˆ European ancestry; ˆ First Nation Canadians. *P < 0.05 between males and females, ANCOVA with age as the covariate. **P < 0.05 between and, ANCOVA with age as the covariate.

3 Table 2 Prevalence of obesity a Prevalence n n % Males Females Males Females Males Females Males Females* ˆ European ancestry; ˆ First Nation Canadians. * and groups signi cantly different in estimated prevalence at P 0.05 (chi square). a Obese de ned as 85th percentile NHANES II age- and genderspeci c reference data in the groups aged 5± 19 y and 85th percentile NHANES II for people aged 20± 29 y: BMI 27.8 in males and BMI 27.3 in females in the groups aged 20 ±75 y. adult females. Similarly, in all groups, except youth, females had greater subcutaneous fatness than males (SUM). Among adults, males were taller and heavier, and males had higher BMIs than females. Differences between and were also apparent in some parameters. adult females had greater mean body mass, BMI, TER and SUM, than. Among adult males, had greater BMI and SUM than. Among youths, males had greater TER and SUM than ; however, the only difference in female youth was the TER, which was greater in. The prevalence rates for obesity are presented in Table 2. Among youth, prevalences ranged from 12.3% in females to 29.4% in females. Among adults, prevalences ranged from 33.3% in females to 60.0% in females. Prevalence rates were greater for in each age=gender group; however, the difference was only statistically signi cant in adult females (chi square ˆ 10.6, P ˆ 0.002). The results of the principal components analyses are presented in Figure 1. The rst principal component (PC1) explains between 33±40% of the variance. The second component explains between 19±24% of the variance; however, loadings follow no apparent pattern and are not shown. In each gender and age group, with the exception of adult males, PC1 represents a trunk-extremity contrast; thus, it is an indicator of relative subcutaneous fat distribution. The direction of the factors indicates that higher mean component scores are associated with more fat on the trunk relative to the extremities. In adult males, PC1 appears to contrast abdominal vs extremity fatness. This perhaps re ects the propensity to store fat preferentially in the abdominal region with age in adult males. Mean scores for PC1 differ among and (Table 3). PC1 is signi cantly greater in in every case except for adult males, where the difference is in the same direction, but not signi cant. The direction 1129 Figure 1 Results of principal components analyses of skinfolds in males aged (a) 5±19 y and (b) 20±75 y, and in females aged (c) 5± 19 y and (d) 20±75 y. Skinfolds: TRI ˆ triceps; BIC ˆ biceps; CALF ˆ Medial calf; SUB ˆ subscapular; SUP ˆ Suprailiac; ABD ˆ abdominal; EV ˆ eigen value.

4 1130 Table 3 Mean rst principal component (PC1) scores and results of t-tests for differences between First Nation () and European ancestry () Canadians of differences indicates that have a greater central or truncal distribution of subcutaneous fatness than. Discussion The prevalence of obesity among American Indian children and youth varies by geographic location and tribal af liation. Malina 19 examined prevalences of obesity in North American children and youth (de ned as 85th percentile NHANES II BMI) from 11% in Chippewa females to 78.3% in Southwestern Arizona Native females. There was also a trend over time, with older samples having lower estimated prevalences of obesity than more recent samples. The prevalences of obesity were generally greater than those reported by Broussard et al 20 among Native American adolescents: 24.5% in males and 25.0% in females ( 85th percentile age-speci c NHANES II BMI 14 ). Based on comparisons to 95th percentiles of the BMI in the NCHS data set, 11.2% and 12.5% of Navajo girls and boys, respectively, exceeded the cut-off. 21 Among adults, Broussard et al 20 estimated that prevalences of overweight (BMI 27.8 in males and BMI 27.3 in females) in Native Americans 18 y at 33.7% in males and 40.3% in females, which are lower than those estimated in the present study. The prevalence of overweight among the Navajo (BMI 27.8 in males and BMI 27.3 in females) was estimated at 30.3% in males and 50.0% in females, 22 which is similar for females, but lower than that for males in the present study. Age-speci c prevalence rates for overweight (BMI 27.8 in males and BMI 27.3 in females) among the Pima ranged from 31±78% for males 20 y and from 48±87% for females 20 y. 23 Studies of the prevalence of overweight among Native groups generally indicate higher prevalences in females than males. 20±25 Results of the present study indicate that adult males have a similar rate of obesity as females, 51.4% and 60.0%, respectively. This trend is also evident in adults. Mean s.d. Mean s.d. t P Males Females Males Females < Males have a rate of 38.3% and the females have a rate of 33.3%. Young and Sevenhuysen 25 indicated that obesity among four Northern Canadian First Nation groups was primarily of the central type, as gauged by the subscapular=triceps skinfold ratio and the waist to hip ratio (WHR). Although raw data were not presented, the authors indicated that 36% of men and 11% of women had a WHR > 0.99, which corresponds to approximately the 95th and 99th percentiles of the French reference data for men and women, respectively. 25 Also, 29.8% of men and 22.4% of women had a subscapular=triceps ratio > Adult males had a somewhat different rst principal component, and the difference between and was not signi cant in this group. Similarly, the TER was not signi cantly different among and adult males. The results suggest that among adult males, differences in relative subcutaneous fat distribution are not as apparent as in other groups. Among Canadian Eskimos, a truncal fat distribution (ratio of triceps=subscapular suprailiac skinfolds) was greater in women than men. 26 The results are also somewhat consistent with the data presented by Johnston et al 27 for urban Native American school children, which indicated that Native American females, but not males, tend to carry proportionally more subcutaneous fat on the trunk, relative to reference data. It has been hypothesized that a central distribution of subcutaneous fat is an adaptation to the cold, as similar observations in women have been made in other cold adapted populations, such as the Evenki reindeer herders of the central Siberian taiga 28 and Mongolian pastoralists. 29 Caution must be used when interpreting these results, however, as relative subcutaneous fat distribution is fat dependent. In other words, as overall fatness increases, the ratio of trunk=extremity fatness also increases. 30,31 There is an apparent relationship between truncal fat distribution and adult-onset diabetes among Native groups. Hall et al 22 reported that among female Navajo, there was an increased risk of diabetes as the WHR increased. A similar trend was evident in the males, but it was not signi cant. Mean WHR was 0.90 for females 20 y and 0.96 for males 20 y. SzathmaÂry and Holt 32 used principal components analyses and hierarchical analysis of variance to examine the association between fat distribution and blood glucose levels among the Dogrib in the Northwest Territories. There was a signi cant association between truncal fat distribution (as assessed by principal components of skinfolds) and elevated blood glucose levels. 32 Given that Canadians have a greater incidence of obesity, diabetes and related metabolic disorders than the general population, some of the differences may be explained by differences in risk factors for disease, such as relative subcutaneous fat distribution. There is a need, however, to better document the genesis of the central

5 pattern of subcutaneous fat distribution among Native Americans during childhood and adolescence. Sample sizes in the present study, as well as in other studies of youth are generally small when grouped by age and gender. Acknowledgements The data collection upon which this research is based was conducted while P.T. Katzmarzyk was a doctoral candidate in the Department of Kinesiology at Michigan State University. References 1 NIH. National institutes of health consensus development conference statement. Ann Intern Med 1985; 103: 1073 ± Ducimetiere P, Richard JL, Cambiae F. The pattern of subcutaneous fat distribution in middle-aged men and the risk of coronary heart disease: The Paris Prospective Study. Int J Obes 1986; 10: 229± Ducimetiere P, Richard JL. The relationship between subsets of anthropometric upper versus lower body measurements and coronary heart disease risk in middle-aged men: The Paris Prospective Study I. Int J Obes 1989; 13: 111± DespreÂs JP, Moorjani S, Lupien PJ, Tremblay A, Nadeau A, Bouchard C. Regional distribution of body fat, plasma lipoproteins, and cardiovascular disease. Arteriosclerosis 1990; 10: 497 ± Gardner LI, Stern MP, Haffner SM, Gaskill SP, Hazuda HP, Relethford JH, Ei er CW. Prevalence of diabetes in Mexican Americans: Relationship to percent of gene pool derived from Native American sources. Diabetes 1984; 33: 86± Weiss KM, Ferrell RE, Hanis CL. A new world syndrome of metabolic diseases with a genetic and evolutionary basis. Yrbk Phys Anthropol 1984; 27: 153± SzathmaÂry EJE. Insulin dependent diabetes mellitus among Aboriginal North Americans. Ann Rev Anthropol 1994; 23: 457± Young TK. Diabetes mellitus among Native Americans in Canada and the United States: An epidemiological review. Am J Hum Biol 1993; 5: 399± Lohman TG, Roche AF, Martorell R (eds). Anthropometric Standardization Reference Manual. Human Kinetics: Champaign, Malina RM, Hamill PVV, Lemeshow S. Selected body measurements of children 6± 11 years, United States, Vital and Health Statistics, Series 11, No National Center For Health Statistics: Hyattsville, MD, Malina RM. Anthropometry. In: Maud PJ, Foster C (eds). Physiological Assessment of Human Fitness. Human Kinetics: Champaign, 1995, 205± Johnston FE, Hamill PVV, Lemeshow S. Skinfold thickness of children 6 ±11 years, United States, Vital and Health Statistics, Series 11, No National Center For Health Statistics: Hyattsville, MD, Chumlea WC, Guo S, Kuczmarski RJ, Johnson CL, Leahy CK. Reliability for anthropometric measurements in the Hispanic Health and Nutrition Examination Survey (HHANES 1982±1984). Am J Clin Nutr 1990; 51: 902s±907s. 14 Najjar MF, Rowland M. Anthropometric reference data and prevalence of overweight, United States, 1976 ± 80. Vital and Health Statistics, Series 11, No National Center For Health Statistics: Hyattsville, MD, Deutsch MI, Mueller WH, Malina RM. Androgyny in fat patterning is associated with obesity in adolescents and young adults. Ann Hum Biol 12: 275± Baumgartner RN, Roche AF, Guo S, Lohman T, Boileau RA, Slaughter MH. Adipose tissue distribution: The stability of principal components by sex, ethnicity and maturation stage. Hum Biol 1986; 58: 719± Malina RM, Huang Y-C, Brown KH. Subcutaneous adipose tissue distribution in adolescent girls of four ethnic groups. Int J Obes 1995; 19: 793± SPSS. SPSS Reference Guide. SPSS Inc: Chicago, Malina RM. Ethnic variation in the prevalence of obesity in North American children and youth. Crit Rev Food Sci Nutr 1993; 33: 389± Broussard BA, Johnson A, Himes JH, Story M, Fichtner R, Hauck F, Bachman-Carter K, Hayes J, Frohlich K, Gray N, Valway S, Gohdes D. Prevalence of obesity in American Indians and Alaska Natives. Am J Clin Nutr 1991; 53: 1535s ± 1542s. 21 Sugarman JR, White LL, Gibert TJ. Evidence for a secular change in obesity, height, and weight among Navajo Indian schoolchildren. Am J Clin Nutr 1990; 52: 960± Hall TR, Hickey ME, Young TB. The relationship of body fat distribution to non-insulin-dependent diabetes mellitus in a Navajo community. Am J Hum Biol 1991; 3: 119± Knowler WC, Pettitt DJ, Saad MF, Charles MA, Nelson RG, Howard BV, Bogardus C, Bennett PH. Obesity in the Pima Indians: Its magnitude and relationship with diabetes. Am J Clin Nutr 1991; 53: 1543s ±1551s. 24 McIntyre L, Shah CP. Prevalence of hypertension, obesity and smoking in three Indian communities in northwestern Ontario. Can Med Assoc J 1986; 134: 345± Young TK, Sevenhuysen G. Obesity in northern Canadian Indians: Patterns, determinants and consequences. Am J Clin Nutr 1989; 49: 786± Schaefer O. Are Eskimos more or less fat than other Canadians? A comparison of skinfold thickness and ponderal index in Canadian Eskimos. Am J Clin Nutr 1977; 30: 1623± Johnston FE, McKigney JI, Hopwood S, Smelker J. Physical growth and development of urban native Americans: A study of urbanization and its implications for nutritional status. Am J Clin Nutr 1978; 31: 1017 ± Leonard WR, Katzmarzyk PT, Comuzzie AG, Crawford MH, Sukernik RI. Growth and nutritional status of the Evenki reindeer herders of Siberia. Am J Hum Biol 1994; 6: 339± Beall CM, Goldstein MC. High prevalence of excess fat and central fat patterning among Mongolian pastoral nomads. Am J Hum Biol 1992; 4: 747± Garn SM, Ryan AS, Robson JKR. Fatness-dependence and utility of the subscapular=triceps ratio. Ecol Food Nutr 1982; 12: 173± Malina RM. Regional body composition: Age, sex and ethnic variation. In: Roche AF, Heyms eld SB, Lohman TG (eds). Human Body Composition. Human Kinetics: Champaign, 1996, 217 ± SzathmaÂry EJE, Holt N. Hyperglycemia in Dogrib Indians of the Northwest Territories, Canada: Association with age and a centripetal fat distribution of body fat. Hum Biol 1983; 55: 493±