To determine if the suicides in a Finnish psychiatric hospital with a history of a suicide epidemic were clustered and if the sex distribution of the suicides followed a random pattern, the authors conducted statistical analyses of the temporal distribution and sex distribution of 59 consecutive inpatient suicides over the years 1967-1992 in the hospital. They found no statistically significant temporal clustering and that the sex distribution of the suicides was random. They conclude that inpatient suicide epidemics seem to be rare and separate events, related more to temporary micro-social factors and to the psychopathology of individual patients than to permanent characteristics of a particular hospital.