Does genetic ancestry explain racial/ethnic disparities in diabetes? results from a longitudinal study

Type 2 diabetes (T2D) is more prevalent in US minority populations, particularly among black and Hispanics Americans compared to whites. Recent epidemiological studies suggest that genetic factors related to biogeographic ancestry (BGA) may explain some of this disparity. Therefore, the objective of this research was to evaluate the contribution of BGA to racial/ethnic disparities in incident T2D. The Boston Area Community Health (BACH) Survey is a racially/ethnically diverse, (black, Hispanic, white) longitudinal cohort of men and women. Participants aged 30-79 at baseline (2002-05) were followed approximately 5 and 7 years later for T2D incidence (n=2732). The BACH Survey collected information relating to: sociodemographics, behaviors, health status, and physical and biochemical parameters, including 63 Ancestry Informative Markers to quantify BGA. We analyzed the relationship between % African and % Native American BGA on incident T2D (OR and [95% CI]). In age- and gender-adjusted models, the odds of developing T2D were 14% higher (1.14 [1.07-1.22]) for every 10% increase in African ancestry, relative to European ancestry. Results were similar for Native American ancestry although not statistically significant (1.14 [0.97-1.33]). These results were eliminated with adjustment for socioeconomic factors (African ancestry: 1.07 [0.99-1.15]; Native American ancestry: 0.99 [0.83-1.18]). BGA only accounted for 0.3% of the variation in T2D incidence (by comparison, income, education, and occupation combined accounted for 1.4%; body mass index 1.6%). We conclude that while African and Native American ancestry are associated with incidence of T2D in minimally adjusted models, the effect is likely explained by differences in socioeconomic factors.