Introduction: Prior estimates of lifetime risk (LTR) for developing type 2 diabetes (diabetes) among US adults have been derived using Markov modeling, thus were not based on longitudinal data. We use pooled data from several US community-based cohorts to estimate the LTR of diabetes according to age, sex and body mass index (BMI).
Hypothesis: Higher BMI is associated with higher LTR of diabetes irrespective of the index age; and men have a higher LTR of diabetes compared to women.
Methods: We pooled individual-level data from 7 US-based cohorts: Framingham Heart Study; Framingham Offspring Heart Study; Coronary Artery Risk Development in Young Adults (CARDIA); Atherosclerosis Risk in Communities (ARIC); Cardiovascular Health Study (CHS); Multi-Ethnic Study of Atherosclerosis (MESA); and Jackson Heart Study (JHS). Participants were free of diabetes (fasting blood glucose ≥126 mg/dl or use of medications) at index age, and were categorized by BMI (normal weight [<25 kg/m2], overweight [25-29 kg/m2] and obese (BMI≥30 kg/m2)]. The LTR of diabetes, at index age 55, was derived modified Kaplan-Meier estimator adjusting for competing risk of death free from diabetes.
Results: Our sample consisted of 14139 Black and White participants (56% female). At index age 55 years, the 25-year adjusted risk for diabetes was 5.3% (95% confidence interval: 4.1-6.6%) in men and 2.8% (2.2-3.5%) in women with normal weight (Figure). The corresponding risks for diabetes among overweight participants were 11.0% (9.7-12.3%) in men and 8.5% (7.3-9.7%) in women. In obese individuals, the 25-year risk for diabetes was 24.3% (21.9- 26.6%) in men, and 20.4% (18.6-22.2%) in women.
Conclusions: In this sample of middle-aged US adults, long-term risk for diabetes was substantially higher in overweight and obese participants. These data strongly suggest that lowering the population burden of obesity is critical for lowering the population burden of diabetes.