Prospective association of a Lifestyle Risk Factor Index with type 2 diabetes in the Multiethnic Cohort

PURPOSE: This study examined behaviors captured in a composite Lifestyle Risk Factor Index (LSRI) in relation to type 2 diabetes (T2D) incidence across five ethnic groups in the Multiethnic Cohort, considering the cumulative and interactive effects of lifestyle factors.

METHODS: The study population included 165,383 European American (EA), African American (AA), Native Hawaiian (NH), Japanese American (JA), and Latino (L) participants. The LSRI score, assessed by baseline questionnaire, assigns 1 point each for no current smoking, physical activity (>/= 150 min/week), consuming < 1 (women) or < 2 (men) alcoholic drinks/day and adhering to >/= 3 of 7 dietary recommendations. Hazard ratios with 95% confidence intervals were estimated by Cox regression.

RESULTS: During a mean follow-up of 17 years, 44,518 (27%) incident T2D cases were identified. Adherence was highest for moderate alcohol (86%) and no current smoking (84%), followed by physical activity (81%) and diet (22%). A 1-point increase in LSRI was associated with a 6% lower incidence of T2D (HR = 0.94; 95%CI 0.93-0.95) in the BMI-adjusted model. No current smoking, physical activity, and healthy diet (without BMI adjustment only) were inversely and moderate alcohol consumption positively associated with T2D incidence. The LSRI was associated with lower T2D risk in BMI-adjusted models for participants with AA, L, and EA ancestry and among JA before BMI adjustment.

CONCLUSIONS: These results confirm that a combination of lifestyle behaviors is critical in T2D prevention. However, not all LSRI components impact T2D risk equally, and both, associations and the impact of BMI adjustment, vary by ethnic group.