Objectives. We estimated the prevalence of overweight and diabetes among US immigrants by region of birth. Methods. We analyzed data on 34456 US immigrant adults from the National Health Interview Survey, pooling years 1997 to 2005. We estimated age- and gender-adjusted and multivariable-adjusted overweight and diabetes prevalence by region of birth using logistic regression. Results. Both men (odds ratio [OR] = 3.3; 95% confidence interval [Cl] = 1.9, 5.8) and women (OR = 4.2; 95% Cl=2.3, 7.7) from the Indian subcontinent were more likely than were European migrants to have diabetes without corresponding increased risk of being overweight. Men and women from Mexico, Central America, or the Caribbean were more likely to be overweight (men: OR = 1.5; 95% Cl = 1.3, 1.7; women: OR = 2.0; 95% Cl = 1.7, 2.2) and to have diabetes (men: OR = 2.0; 95% Cl = 1.4, 2.9; women: OR = 2.0; 95% Cl = 1.4, 2.8) than were European migrants. Conclusions. Considerable heterogeneity in both prevalence of overweight and diabetes by region of birth highlights the importance of making this distinction among US immigrants to better identify subgroups with higher risks of these conditions.

[No Keywords available]

Europe risk human sex difference middle aged Odds Ratio statistics diabetes mellitus Confidence Intervals obesity Aged Logistic Models Health Surveys Central America ethnology Mexico Hispanic Americans Confidence interval Young Adult United States Humans Hispanic Adolescent male Emigrants and Immigrants female risk factor Risk Factors Caribbean Region prevalence Article adult migration age Sex Factors Age Factors statistical model body mass Body Mass Index health survey