Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Metabolism. Author manuscript; available in PMC 2010 November 1.
Published in final edited form as:
PMCID: PMC2763950

Height, Ethnicity, and the Incidence of Diabetes: the San Antonio Heart Study


Mexican Americans are more obese and have more diabetes than non-Hispanic whites, but are also shorter. Height is used in some diabetes prediction models. Therefore we examined the effect of height on the relationship between ethnicity and incident diabetes. Incident diabetes was ascertained in 1,730 participants in the San Antonio Heart Study (age range, 25 to 64 years) after 7.4 years of follow-up. Height predicted diabetes in neither men (odds ratio [OR] × 1 standard deviation [SD], 1.14 [0.85 – 1.51]) nor women (OR × 1 SD, 0.88 [0.70 – 1.11]) after adjusting for age and ethnicity. The area under the receiver operating characteristic curve (AUC) for predicting diabetes of a model that included waist circumference (in men, 0.775; in women, 0.781) was similar to that of models that included waist circumference + height (in men, 0.775, p = 0.702; in women, 0.783, p = 0.680) or waist-to-height ratio (in men, 0.764, p = 0.161; in women, 0.783, p = 0.619). The OR of incident diabetes according to ethnicity was lower in the model that was adjusted for the waist-to-height ratio than in the model that accounted only for waist circumference (in women, 1.45 (0.86 – 2.46) vs. 1.84 (1.10 – 3.08), p <0.001; in men, 2.00 (1.11 – 3.58) vs. 2.74 (1.52 – 4.95), p <0.001). In conclusion, the addition of height to adjust waist circumference does not increase the ability of waist circumference to predict diabetes, but may be useful in exploring differences in diabetic risk between populations of different race/ethnicity.

Keywords: Height, obesity, body mass index, type 2 diabetes, ethnicity, Epidemiology

1. Introduction

According to estimates from the National Health and Nutrition Examination Survey (NHANES) 2005-2006, 12.9% of the US population aged ≥20 years have diabetes (1). The number is likely to increase in the future because of changes in size and demographic characteristics of the US population as well as increasing prevalence (2). Diabetes rates are increasing because of the obesity epidemic (3) in all race/ethnic groups (4). Nevertheless, differences in diabetic risk between race/ethnic groups may be difficult to differentiate from the contribution of obesity to that risk for two reasons: first, race/ethnicity is frequently associated with distinct anthropometric measurements; second, standardized indices of obesity have been derived largely from studies in non-Hispanic white populations (5-7).

In San Antonio, Mexican Americans are more obese and have more type 2 diabetes than non-Hispanic whites (8), but are also shorter. Height is included in some clinical prediction models of type 2 diabetes (9,10). Therefore, height may explain to some extend the difference in diabetes risk between Mexican Americans and non-Hispanic whites. Therefore we compared the discriminatory capacity of height-adjusted waist circumference (11-14) with that of body mass index (BMI), waist circumference, and waist-to-hip circumference ratio in the San Antonio Heart Study (SAHS).

2. Materials and methods

2.1. Subjects

The SAHS was designed as a population-based study on type 2 diabetes and cardiovascular risk factors among Mexican Americans and non-Hispanic whites of San Antonio. SAHS had approved protocols by the Institutional Review Board of the University of Texas Health Science Center at San Antonio. All subjects gave written informed consent. Detailed descriptions have been already published (8,15,16). Briefly, all Mexican American and non-Hispanic white, men and nonpregnant women aged 25 to 64 years that resided in randomly selected households from low-, middle-, and high-income census tracts were invited to participate (response rate 65.3%). Mexican Americans and non-Hispanic whites were defined by previously published algorithm (17). Enrollment was carried out in two phases: from January 1979 to December 1982 and from January 1984 to December 1988. We used data only from phase 2 participants (n = 2,941), because waist circumference was not measured in phase 1 participants at baseline. Phase 2 participants were re-examined between October 1991 and October 1996. The median for the follow-up period was 7.4 years (range, 6.3 – 10.3 years). Acquisition of data and definition of variables and outcomes

Anthropometric measurements were gathered by trained personnel after participants had removed their shoes and upper garments and put on an examination gown (15). Waist and hip circumferences were measured with a non-stretchable standard tape at the level of the umbilicus and greater trochanter, respectively. The average of two readings was taken to reduce the measurement error. Blood specimens were obtained after a 12- to 14-hour fast. A 75-g oral glucose load (Orangedex; Custom Laboratories, Baltimore, MD) was administered to ascertain diabetes status at the baseline and follow-up examinations. Fasting and 2-hour glucose concentrations were measured with an Abbott Bichromatic Analyzer (South Pasadena, CA) in the laboratory of the Division of Clinical Epidemiology in San Antonio (16).

BMI was used as a measure of overall obesity and waist circumference, waist-to-height ratio, and waist-to-hip ratio as measures of central obesity. Diabetes was defined according to the plasma glucose cut-points of the 2003 American Diabetes Association (fasting glucose ≥7.0 mmol/l, 2-hour glucose ≥11.1 mmol/l) (18). Regardless of glucose values, subjects who reported current therapy with antidiabetic medications were considered to have diabetes. The outcome of interest was incident diabetes. In nondiabetic individuals at the baseline examination, the presence of diabetes at the follow-up examination was synonymous of incident diabetes.

2.2. Statistical analyses

Statistical analyses were performed with the SAS statistical software (SAS Institute Inc. Cary, NC). To account for the effect of age, anthropometric measurements and markers of obesity were compared by one-way analysis of covariance and incident diabetes by logistic regression analysis. The relation of height to age and waist circumference was assessed by generalized linear models. In separate models, significant associations were tested for linearity (adding squared terms for height) as well as for interactions of ethnicity (adding an interaction term for ethnicity × height). Diabetic risk associated with height and obesity was determined by logistic regression analysis. Interaction terms for ethnicity × height and ethnicity × obesity were analyzed. The area under the receiver operating characteristic curve (AUC) was chosen as the parameter to compare the predictive discrimination between models (19). The effect of each obesity index on the difference in incident diabetes according to ethnicity was ascertained by logistic regression analysis in sex-specific models that included age as a covariate. A bootstrap method with 1000 replications was applied to compare the OR of incident diabetes according to ethnicity for the model that was adjusted for age + waist circumference with the OR for each one of the other non-nested models. We considered significant a p value <0.050.

3. Results

Diabetes status was ascertained in 1,734 of 2,511 (69.1%) nondiabetic participants who were alive at follow-up. Ascertainment rates of incident diabetes were similar in Mexican Americans and non-Hispanic whites (68.8% vs. 69.5%, p = 0.714), but younger individuals were less likely to return to the follow-up examination than older individuals (p <0.001). A total of 1,730 participants were eligible for analysis, because relevant information was missing in 4 individuals. After adjusting for age and ethnicity, participants who were eligible for analysis (730 men and 1,000 women) did not differ from those who were excluded (n = 356 men and 425 women) in terms of any of the baseline anthropometric measurements (p >0.1 for all comparisons).

Table 1 shows age-adjusted baseline characteristics of the eligible participants by sex and ethnicity. Mexican Americans were shorter and more obese than non-Hispanic whites. However, waist circumference was comparable in men of both ethnic groups.

Table 1
Age-adjusted baseline characteristics by sex and ethnicity

In linear regression analysis, there was an indirect relationship between height and age (in men, β estimate ± standard error -0.09 ± 0.02, p <0.001; in women, β -0.09 ± 0.02, p <0.001). There was not significant interaction of ethnicity on the association of height and age (in men, p = 0.818; in women, p = 0.346). In addition, there was a direct association between waist circumference and height after accounting for the effect of age (in men, β 2.36 ± 0.56, p <0.001; in women, β 1.74 ± 0.74, p = 0.019). This relationship was linear in both men (squared terms of height, p = 0.986) and women (height2, p = 0.498) and was subject to no significant interaction of ethnicity in men (p = 0.891). In women, however, interaction terms of ethnicity × height were close to significance (p = 0.054). Waist circumference was related to height in non-Hispanic white women (β 3.80 ± 1.04, p <0.001), but it wasn't in Mexican American women (β 0.62 ± 0.98, p = 0.524).

More Mexican Americans developed incident diabetes than non-Hispanic whites (in men, 12.7% vs. 6.4%, p <0.001; in women, 14.4% vs. 6.5%, p <0.001). In men, height did not predict future diabetes (odds ratio [OR] × 1 standard deviation [SD], 0.85 [95% confidence intervals {CI} 0.87 – 1.08]). Neither was height a predictor after adjusting for age and ethnicity (OR × 1 SD, 1.14 [0.85 – 1.51]). In women, however, height was associated with lower diabetic risk (OR × 1SD, 0.72 [0.59 – 0.88]), but this association was no longer statistically significant after adjusting for age and ethnicity (OR × 1 SD, 0.88 [0.70 – 1.11]).

All indices of obesity predicted incident diabetes (Table 2). The odds of predicting diabetes associated with waist circumference did not change by introducing height into the model. In this model, height did not predict diabetes (in men, OR × 1 SD, 1.01 [0.75 – 1.36]; in women, OR × 1 SD, 0.82 [0.65 – 1.05]). In a separate waist circumference + height model, an interaction term for ethnicity × height was not statistically significant (in men, p = 0.215; in women, p = 0.123). Neither was significant the effect of ethnicity on the relationship between waist-to-height ratio and incident diabetes (in men, p = 0.562; in women, p = 0.113).

Table 2
Obesity indices as predictors of incident diabetes (7.4 year follow-up period)

The AUC for predicting incident diabetes of sex-specific models that included age, ethnicity, and waist circumference as independent variables did not change by adding height into the model or by using a height-adjusted measure of waist circumference (Table 3). In men, all models that included any index of obesity had similar AUCs for predicting incident diabetes. In women, however, the AUC of the model that included waist circumference was greater than that of models that included BMI or waist-to-hip ratio.

Table 3
Comparisons between AUC of obesity indices for predicting incident diabetes (7.4 year follow-up period)

The OR of incident diabetes according to ethnicity (Mexican Americans vs. non-Hispanic whites) was lower in the model that was adjusted for the waist-to-height ratio than in the model that did not account for height, reflecting in a more appropriate fashion that inclusion of height in the model attenuated the impact of ethnicity on incident diabetes even after adjustment for waist girth (Table 4). The OR of incident diabetes according to ethnicity for the model that was adjusted for the waist-to-height ratio was similar to that for the model that was adjusted for BMI (in men, p = 0.740, in women, p = 0.312).

Table 4
Effect of each obesity index on the relation of ethnicity to incident diabetes (7.4 years of follow-up) by multivariable logistic regression analysis

4. Discussion

Waist circumference predicts diabetes better than BMI or waist-to-hip ratio in women, but all indices are similar in men. Correcting waist circumference for height does not help to explain a larger proportion of the risk of diabetes in either men or women, but attenuates more the impact of ethnicity on incident diabetes.

Height is considered a marker of superior intrauterine and childhood nutrition and growth (20,21). Height has been inversely associated with cardiovascular disease risk (20-22) although this association is not always observed (23,24). Height has been also associated with lower diabetic risk and is included in some clinical prediction models for type 2 diabetes (9,10). However, the relation of height to insulin resistance and type 2 diabetes may not retain statistical significance after accounting for the effect of other risk factors (25). In our study, height is not associated with diabetes incidence in men. It is associated in women, but statistical significance is not observed after taking into account the effect of age and ethnicity. Recent reports have also indicated that leg length, a marker of prepubertal growth (26), is the component of stature that is related to insulin resistance, type 2 diabetes, and cardiovascular disease (25,27,28). The ability of leg length to predict type 2 diabetes requires further research, but this anthropometric measure is not available in the SAHS.

Abdominal visceral fat is a major determinant of clustering of disorders associated with the metabolic syndrome (29). However, evidence on the discriminatory capacity of markers of obesity for predicting future diabetes is conflicting. Some studies favor the use of waist circumference (15) or waist-to-hip ratio over BMI (30), but others favor BMI (31,32). Moreover, some studies do not favor any index over the rest (33), and others back the importance of both types of indices (central and overall obesity) (34). Clearly, diabetic risk is driven by obesity in all ethnic groups, but the preeminence of one type of obesity over the rest remains to be elucidated by more precise markers of fat distribution (such as those derived from imaging techniques).

Waist circumference may be the best anthropometric marker of abdominal visceral fat (12), but it is not clear whether height affects the relationship between waist circumference and abdominal visceral fat. Height has no effect in some studies (11,12), but has significant effect in others (13,14) —linear in women and quadratic in men (14). We cannot examine the influence of height on the relationship between waist circumference and abdominal visceral fat, because precise measures of fat distribution are not available in the SAHS. Nevertheless, height does not affect the discriminatory capacity of waist circumference for predicting future diabetes.

If the model that includes waist-to-height ratio does not explain a larger proportion of the risk of diabetes than the model that includes waist circumference, how is that waist-to-height ratio attenuates more the impact of ethnicity on incident diabetes? Mexican Americans, both men and women, have lower stature than non-Hispanic whites, although only Mexican American women have a higher waist circumference. Therefore differences in a height adjusted of waist circumference between Mexican Americans and non-Hispanic whites are higher than differences in waist circumference in both men and women. Therefore adjusting waist circumference for height lessens the relationship between ethnicity and incident diabetes. Additionally, the significance anthropometric measurements may not be applicable uniformly to all populations (5,6), and neither may be fat distribution ((35-38). Non-Hispanic whites have more abdominal visceral fat than African Americans and less than Asian Americans (35,37). In the Insulin Resistance Atherosclerosis Family Study, Mexican Americans have more abdominal visceral fat than African Americans, even though their BMI was similar (36). Thus, it is plausible that the greater diabetic risk in Mexican Americans may be related not only to more obesity but also to a relatively greater amount of abdominal visceral fat.

In summary, diabetic risk associated with waist circumference may be independent of height within a given population. However, height may be an anthropometric measure to consider in studies that examine differences in diabetic risk between populations of different race/ethnicity.


This work was supported by grants from the National Heart, Lung and Blood Institute (RO1-HL24799 and RO1-HL36820).


Conflicts of Interest/Disclosures: The authors declare that there is no duality of interest associated with this manuscript.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


1. Cowie CC, Rust KF, Ford ES, et al. Full accounting of diabetes and pre-diabetes in the U.S. population in 1988-1994 and 2005-2006. Diabetes Care. 2009;32:287–94. [PMC free article] [PubMed]
2. Mainous AG, Baker R, Koopman RJ, et al. Impact of the population at risk of diabetes on projections of diabetes burden in the United States: an epidemic on the way. Diabetologia. 2007;50:934–40. [PMC free article] [PubMed]
3. Gregg EW, Cheng YJ, Cadwell BL, et al. Secular trends in cardiovascular disease risk factors according to body mass index in US adults. JAMA. 2005;293:1868–74. [PubMed]
4. McBean AM, Li S, Gilbertson DT, Collins AJ. Differences in diabetes prevalence, incidence, and mortality among the elderly of four racial/ethnic groups: whites, blacks, Hispanics, and Asians. Diabetes Care. 2004;27:2317–24. [PubMed]
5. Alberti KG, Zimmet P, Shaw J. The metabolic syndrome--a new worldwide definition. Lancet. 2005;366:1059–62. [PubMed]
6. Grundy SM, Cleeman JI, Daniels SR, et al. Diagnosis and management of the metabolic syndrome: an American Heart Association/National Heart, Lung, and Blood Institute Scientific Statement. Circulation. 2005;112:2735–52. [PubMed]
7. World Health Organization. Appropriate body-mass index for Asian populations and its implication for policy and intervention strategies. Lancet. 2004;363:157–63. [PubMed]
8. Haffner SM, Hazuda HP, Mitchell BD, Patterson JK, Stern MP. Increased incidence of type II diabetes mellitus in Mexican Americans. Diabetes Care. 1991;14:102–8. [PubMed]
9. Schulze MB, Hoffmann K, Boeing H, et al. An accurate risk score based on anthropometric, dietary, and lifestyle factors to predict the development of type 2 diabetes. Diabetes Care. 2007;30:510–5. [PubMed]
10. Schmidt MI, Duncan BB, Bang H, et al. Identifying individuals at high risk for diabetes: The Atherosclerosis Risk in Communities study. Diabetes Care. 2005;28:2013–8. [PubMed]
11. Han TS, McNeill G, Seidell JC, Lean ME. Predicting intra-abdominal fatness from anthropometric measures: the influence of stature. Int J Obes Relat Metab Disord. 1997;21:587–93. [PubMed]
12. Pouliot MC, Despres JP, Lemieux S, et al. Waist circumference and abdominal sagittal diameter: best simple anthropometric indexes of abdominal visceral adipose tissue accumulation and related cardiovascular risk in men and women. Am J Cardiol. 1994;73:460–8. [PubMed]
13. Ashwell M, Cole TJ, Dixon AK. Ratio of waist circumference to height is strong predictor of intra-abdominal fat. BMJ. 1996;313:559–60. [PMC free article] [PubMed]
14. Schreiner PJ, Terry JG, Evans GW, Hinson WH, Crouse JR, III, Heiss G. Sex-specific associations of magnetic resonance imaging-derived intra-abdominal and subcutaneous fat areas with conventional anthropometric indices. Am J Epidemiol. 1996;144:335–45. [PubMed]
15. Wei M, Gaskill SP, Haffner SM, Stern MP. Waist circumference as the best predictor of noninsulin dependent diabetes mellitus (NIDDM) compared to body mass index, waist/hip ratio and other anthropometric measurements in Mexican Americans--a 7-year prospective study. Obes Res. 1997;5:16–23. [PubMed]
16. Burke JP, Williams K, Gaskill SP, Hazuda HP, Haffner SM, Stern MP. Rapid rise in the incidence of type 2 diabetes from 1987 to 1996: results from the San Antonio Heart Study. Arch Intern Med. 1999;159:1450–6. [PubMed]
17. Hazuda HP, Comeaux PJ, Stern MP, Haffner SM, Eifler CW, Rosenthal M. A comparison of three indicators for identifying Mexican Americans in epidemiologic research: methodological findings from the San Antonio Heart Study. Am J Epidemiol. 1986;123:96–112. [PubMed]
18. Genuth S, Alberti KG, Bennett P, et al. Follow-up report on the diagnosis of diabetes mellitus. Diabetes Care. 2003;26:3160–7. [PubMed]
19. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988;44:837–45. [PubMed]
20. Hebert PR, Rich-Edwards JW, Manson JE, Ridker PM, Cook NR, O×Connor GT, Buring JE, Hennekens CH. Height and incidence of cardiovascular disease in male physicians. Circulation. 1993;88:1437–43. [PubMed]
21. Silventoinen K, Zdravkovic S, Skytthe A, et al. Association between height and coronary heart disease mortality: a prospective study of 35,000 twin pairs. Am J Epidemiol. 2006;163:615–21. [PubMed]
22. Song YM, Smith GD, Sung J. Adult height and cause-specific mortality: a large prospective study of South Korean men. Am J Epidemiol. 2003;158:479–85. [PubMed]
23. Kannam JP, Levy D, Larson M, Wilson PW. Short stature and risk for mortality and cardiovascular disease events. The Framingham Heart Study. Circulation. 1994;90:2241–7. [PubMed]
24. Samaras TT, Elrick H, Storms LH. Is short height really a risk factor for coronary heart disease and stroke mortality? Med Sci Monit. 2004;10:RA63–76. [PubMed]
25. Asao K, Kao WH, Baptiste-Roberts K, Bandeen-Roche K, Erlinger TP, Brancati FL. Short stature and the risk of adiposity, insulin resistance, and type 2 diabetes in middle age: the Third National Health and Nutrition Examination Survey (NHANES III), 1988-1994. Diabetes Care. 2006;29:1632–7. [PubMed]
26. Dangour AD, Schilg S, Hulse JA, Cole TJ. Sitting height and subischial leg length centile curves for boys and girls from Southeast England. Ann Hum Biol. 2002;29:290–305. [PubMed]
27. Gunnell D, May M, Ben-Shlomo Y, Yarnell J, Smith GD. Leg length, insulin resistance, and coronary heart disease risk: the Caerphilly Study. J Epidemiol Community Health. 2001;55:867–72. [PMC free article] [PubMed]
28. Langenberg C, Hardy R, Breeze E, Kuh D, Wadsworth ME. Influence of short stature on the change in pulse pressure, systolic and diastolic blood pressure from age 36 to 53 years: an analysis using multilevel models. Int J Epidemiol. 2005;34:905–13. [PubMed]
29. Carr DB, Utzschneider KM, Hull RL, et al. Intra-abdominal fat is a major determinant of the National Cholesterol Education Program Adult Treatment Panel III criteria for the metabolic syndrome. Diabetes. 2004;53:2087–94. [PubMed]
30. Rosenthal AD, Jin F, Shu XO, et al. Body fat distribution and risk of diabetes among Chinese women. Int J Obes Relat Metab Disord. 2004;28:594–9. [PubMed]
31. Chan JM, Rimm EB, Colditz GA, Stampfer MJ, Willett WC. Obesity, fat distribution, and weight gain as risk factors for clinical diabetes in men. Diabetes Care. 1994;17:961–9. [PubMed]
32. Tulloch-Reid MK, Williams DE, Looker HC, Hanson RL, Knowler WC. Do measures of body fat distribution provide information on the risk of type 2 diabetes in addition to measures of general obesity? Comparison of anthropometric predictors of type 2 diabetes in Pima Indians. Diabetes Care. 2003;26:2556–61. [PubMed]
33. Sargeant LA, Bennett FI, Forrester TE, Cooper RS, Wilks RJ. Predicting incident diabetes in Jamaica: the role of anthropometry. Obes Res. 2002;10:792–8. [PubMed]
34. Wang Y, Rimm EB, Stampfer MJ, Willett WC, Hu FB. Comparison of abdominal adiposity and overall obesity in predicting risk of type 2 diabetes among men. Am J Clin Nutr. 2005;81:555–63. [PubMed]
35. Araneta MR, Barrett-Connor E. Ethnic differences in visceral adipose tissue and type 2 diabetes: Filipino, African-American, and white women. Obes Res. 2005;13:1458–65. [PubMed]
36. Mitchell BD, Zaccaro D, Wagenknecht LE, et al. Insulin sensitivity, body fat distribution, and family diabetes history: the IRAS Family Study. Obes Res. 2004;12:831–9. [PubMed]
37. Hoffman DJ, Wang Z, Gallagher D, Heymsfield SB. Comparison of visceral adipose tissue mass in adult African Americans and whites. Obes Res. 2005;13:66–74. [PubMed]
38. Park YW, Allison DB, Heymsfield SB, Gallagher D. Larger amounts of visceral adipose tissue in Asian Americans. Obes Res. 2001;9:381–7. [PubMed]