The Israeli population is characterized by its marked ethnic diversity. These ethnic groups (originating mainly from Yemen/Aden, the Middle East, North Africa and Europe/America) have kept traditional distinct lifestyle habits and exhibit different morbidity and mortality trends. The aim of the present study was to evaluate the associations among ethnic background, lifestyle patterns and 18-y all-cause mortality. A subgroup of 632 individuals aged 41-70 y, drawn from a larger stratified cohort from the Israel Glucose Intolerance, Obesity and Hypertension study, were personally interviewed, using a quantified food-frequency questionnaire, including most food items consumed by the different subpopulations in Israel. Physical activity was also evaluated, as well as smoking status. Weight, height and blood pressure (BP) measurements were taken. Predictors of mortality were assessed using Cox proportional hazards models. Over the 18-y follow-up period, 151 deaths occurred (24%). In comparison with Yemenites, the adjusted hazard ratios (HR) for all cause mortality were HR = 1.77 [95% confidence interval (CI): 1.01-3.09] for Europeans/Americans; HR = 1.63 (95% CI: 0.89-2.99) for those from a Middle Eastern background; and HR = 1.56 (95% CI: 0.82-2.97) for North Africans. Mortality risk was 43% lower among those consuming > or =25 g of dietary fiber daily [HR = 0.57 (95% CI: 0.41-0.72)], and 42% lower for those consuming <300 mg/d of cholesterol [HR = 0.58 (95% CI: 0.34-0.96)]. Accumulating an average of 0.5 h/d of moderate physical activity reduced mortality by 47% [HR = 0.53 (95% CI: 0.29-0.97)]. Smoking, increased systolic BP, older age and male sex increased mortality risk. We conclude that in our study, although ethnic origin and lifestyle habits are interrelated, each affects mortality independently.