Citation

Allebeck P, Allgulander C. Acta Psychiatr. Scand. 1990; 81(6): 565-570.

Affiliation

Department of Medicine, Karolinska Institute, Huddinge University Hospital, Stockholm, Sweden.

Copyright

(Copyright © 1990, John Wiley and Sons)

DOI

unavailable

PMID

2378251

Abstract

The role of psychiatric illness versus social and behavioural risk factors for suicide in young men was analysed in a longitudinal study of 50,465 conscripts. Data collected in 1969-1970 on social background, personality characteristics, use of alcohol and drugs, psychological assessment and psychiatric diagnosis were linked to records from the national psychiatric case register and the national cause-of-death register through 1983. A total of 247 deaths from suicide occurred in the cohort during the follow-up. By means of multivariate analysis, the role of different social and behavioural characteristics was assessed in relation to that of psychiatric diagnoses, with suicide as dependent variable. A psychiatric diagnosis in inpatient care (n = 2247) was the strongest predictor of suicide, with an odds ratio (OR) of 11.3 (8.3-15.4), controlling for social and behavioural risk factors. Schizophrenia was the diagnosis with the highest suicide risk: OR = 13.3 (8.2-21.6). A psychiatric diagnosis at conscription (n = 5877) was not associated with a significantly increased risk of suicide. Several indicators of poor social background, deviant behaviour, substance abuse and disrupted interpersonal relations were associated with a significantly increased suicide risk, also after controlling for psychiatric illness. Although mental illness requiring inpatient treatment was the most powerful predictor of suicide, less than half the cohort had received such treatment. Social and behavioural risk factors are thus important for prevention on the population level.

Language: en