Abstract

OBJECTIVE To determine the extent of, and reasons for, ethnic differences in type 2 diabetes incidence in the U.K. RESEARCH DESIGN AND METHODS Population-based triethnic cohort. Participants were without diabetes, aged 40–69 at baseline (1989–1991), and followed-up for 20 years. Baseline measurements included fasting and postglucose bloods, anthropometry, and lifestyle questionnaire. Incident diabetes was identified from medical records and participant recall. Ethnic differences in diabetes incidence were examined using competing risks regression. RESULTS Incident diabetes was identified in 196 of 1,354 (14%) Europeans, 282 of 839 (34%) Indian Asians, and 100 of 335 (30%) African Caribbeans. All Indian Asians and African Caribbeans were first-generation migrants. Compared with Europeans, age-adjusted subhazard ratios (SHRs [95% CI]) for men and women, respectively, were 2.88 (95%, 2.36–3.53; P < 0.001) and 1.91 (1.18–3.10; P = 0.008) in Indian Asians, and 2.23 (1.64–3.03; P < 0.001) and 2.51 (1.63–3.87; P < 0.001) in African Caribbeans. Differences in baseline insulin resistance and truncal obesity largely attenuated the ethnic minority excess in women (adjusted SHRs: Indian Asians 0.77 [0.49–1.42]; P = 0.3; African Caribbeans 1.48 [0.89–2.45]; P = 0.13), but not in men (adjusted SHRs: Indian Asians 1.98 [1.52–2.58]; P < 0.001 and African Caribbeans, 2.05 [1.46–2.89; P < 0.001]). CONCLUSIONS Insulin resistance and truncal obesity account for the twofold excess incidence of diabetes in Indian Asian and African Caribbean women, but not men. Explanations for the excess diabetes risk in ethnic minority men remains unclear. Further study requires more precise measures of conventional risk factors and identification of novel risk factors. The global prevalence of type 2 diabetes continues to increase, with the Indian subcontinent predicted to contribute the greatest increase in the number of people with diabetes by 2030 (1). Indian Asian migrant populations also experience greater prevalence of diabetes than host white populations (2,3). Although prevalence of diabetes in sub-Saharan Africa remains low, the prevalence in African-origin populations in other areas of the world is elevated compared with that of white populations (4,5). Few studies have explored explanations for ethnic differences in diabetes incidence. The Atherosclerosis Risk in Communities (ARIC) study found that although adiposity, lifestyle, and socioeconomic factors accounted for nearly 50% of the excess diabetes risk in African American women, none of the excess could be explained in men (6), echoing findings from the National Health and Nutrition Examination Survey (NHANES) (7,8). However, previous studies did not explore the role of insulin resistance and ectopic fat distribution. Further, there are no longitudinal studies to explain the excess diabetes risk in Indian Asians compared with Europeans. We have reported a threefold prevalence of diabetes in men and women aged 40–70 years of Indian Asian and African Caribbean origin compared with Europeans in the SABRE (Southall And Brent REvisited) cohort. We now report incidence of diabetes and potential explanations for ethnic differences in incidence in this unique cohort with a 20-year follow-up to ages 60–89 years.