The present study focuses on the associations between self-rated long-standing psychiatric illness, ethnicity, all-cause mortality and violent death (accidents and suicide), in a sample of 39,155 Swedish-born and foreign-born individuals. The study was designed as a longitudinal follow-up study, covering the period between 1 January 1979 and 31 December 1996. The data were analysed by a proportional hazard model and the results are given as hazard ratios (HR) with 95% confidence intervals (CI). Self-reported long-standing psychiatric illness was a strong risk factor for total mortality: women had an HR of 2.13 (CI = 1.78-2.54) and men an HR of 1.84 (CI = 1.53-2.21), when adjusted for background factors such as country of birth, civil status and socio-economic factors. Finnish men had an increased risk of all-cause mortality compared to Swedes in the final model, when adjusted for socio-economic factors. Long-standing psychiatric illness was also a strong risk factor for violent death, with an HR of 3.51 (CI = 2.32-5.32). The risk of violent death was 2.4 times higher for men than for women. The conclusions of the present study are that self-reported long-standing psychiatric illness is a strong predictor of an increased all-cause mortality and increased mortality from violent death. The increased age-adjusted mortality risk for foreign-born men could be explained by disadvantaged social and economic conditions. Only Finnish men demonstrated an independent increased all-cause mortality risk.