Abstract
BACKGROUND: There is a paucity of studies on the role of early risk factors for the choice of methods for violent suicide attempts. Adolescent risk factors for the choice of violent or non-violent methods for suicide attempts and the risk of subsequent suicide were studied using a longitudinal design. METHODS: A national Swedish cohort of 48 834 18-20-year-old young men conscripted for military service from 1969 to 1970 was followed through official registers during a 37-year period. Two questionnaires concerning their psychosocial background were answered by each conscript. Cox proportional hazard regression analyses were used to estimate the risk for different methods of attempted suicide and later suicide. RESULTS: A total of 1195 (2.4 %) men had made a suicide attempt and of these, 133 (11.1 %) committed suicide later. The number of suicide victims among the non-attempters was 482 (1 %). Half of the suicides occurred during the same year as the attempt. Suicide victims had earlier onset of suicidal behaviour and had more often used hanging as a method of attempted suicide than those who did not later commit suicide. The early risk factors for both violent and non-violent methods of suicide attempt were quite similar. CONCLUSION: Violent suicide attempts, especially by hanging, are associated with a clearly elevated suicide risk in men and require special clinical and public health attention. The early risk factors related to the choice of either a violent or a non-violent suicide attempt method are interlinked and circumstantial factors temporally close to the suicide attempt, such as access to a specific method, may partly explain the choice of method.