The 46 study participants were all offenders who had been referred for forensic psychiatric treatment because of a "serious mental disorder;" all had been legally discharged from compulsory treatment between January 1, 1992, and December 31, 1999. Demographic data and information on diagnosis, duration of hospital stay, and the most serious offense were obtained from medical files. The cases were then linked to the National Cause-of-Death Register. Standardized mortality rates were calculated for the 46 subjects by dividing the observed mortality by the expected mortality as computed from population data for the same time period. Median follow-up time was 53 months. The analysis found a significantly elevated SMR for the sample that was 13.4 times higher than that of the general population, mostly due to suicide. Although the cohort size was small, it was representative, and it provides data from another country for the expanding international pool that confirms the high risk of premature, generally self-inflicted death among mentally disordered offenders. This indicates that resettlement and rehabilitation services after release from secure treatment should focus as much on mortality risk as reoffending. 14 references