Population impact of strategies for identifying groups at high risk of type 2 diabetes

Background. To assess the incidence of diabetes among sub-groups of the population defined by the presence of one or more simple risk factors, and to investigate population stratification as a means of identifying groups at high risk of diabetes. Methods. Data from EPIC-Norfolk (1993-1998), a population-based cohort study of 24,714 men and women aged 40-78 years without selfreported diabetes at baseline, were analyzed. During 12 years of follow-up, 608 new cases of diabetes were recorded. Results. Age (RR 1.03; 95% CI 1.02, 1.04), parental history of diabetes (RR 2.15; 95% CI 1.80, 2.57), BMI (RR 1.76; 95% CI 1.53, 2.02) and physical activity (RR 0.72-0.77 (reference sedentary)) were independently related to risk of diabetes. Sedentary, obese individuals aged over 55 years, with a parental history of diabetes were 18 times more likely to develop diabetes than those in the lowest risk group. Conclusion. Sedentary, obese men and women over 55 years with a parental history of diabetes form a readily identifiable group, which could be targeted for screening and primary prevention. Groups such as that defined by physical inactivity alone would be more suitable for population level approaches. (c) 2006 Elsevier Inc. All rights reserved.