The aim of the study was to estimate life expectancies in different diagnostic groups for individuals treated as inpatients at Swedish psychiatric clinics. All individuals, older than 18 y and alive on the first of January 1983, who had been registered in the National Hospital Discharge Registry by a psychiatric clinic in 1978-82, were monitored for mortality during 1983 by using the National Cause of Death Registry. The study group consisted of 91 385 men and 77 217 women. The patients were divided into nine diagnostic groups according to the principal diagnosis registered at the latest discharge. Actuarial mathematics was used to construct life expectancy tables, which present the number of years expected to live, by gender and diagnostic group. Expectancies of life were significantly shortened for both genders and in all nine diagnostic groups (with one exception). Mental disorders in general are life shortening. This fact should be recognised in community health when setting health priorities. It should also be addressed in curricula as well as in treatment and preventive programmes.