OBJECTIVES: The aim was to study the effect of a 5-year diet intervention on 24-year mortality in middle aged men with combined hyperlipidaemia. SETTING: We studied 104 initially healthy men (in 1972) aged 40-49 years with baseline values of total serum cholesterol >6.45 mmol L-1 and fasting triglycerides >2.55 mmol L-1, within the randomized diet and smoking cessation trial of the Oslo study (n = 1232). METHODS: The participants were randomized to a 5-year diet intervention or a control group. The diet consisted of a traditional lipid-lowering diet with emphasis on reduction of saturated fat, total caloric intake and body weight. The groups were initially well balanced with regard to traditional risk factors for mortality. RESULTS: Thirty-three subjects died during the 24-year observation period [17 of cardiovascular disease (CVD) and 12 of cancer]. In the diet intervention group, mortality was 51% lower (RR = 0.49, 95% CI 0.22-0.91, P = 0.022) as compared with the control group. This difference remained significant in a Cox regression analysis after adjusting for age and smoking status (RR = 0.47, 95% CI 0.23-0.96, P = 0.038). CONCLUSION: This study indicates that the investigated 5-year diet intervention significantly reduces late mortality in healthy middle-aged men with combined hyperlipidaemia.