Risk factors for diagnosed intentional self-injury: A total population-based study.

BACKGROUND: Few studies investigate predictors of intentional self-injury over time in non-clinical samples. By using longitudinal data from the whole population of the county of Scania, Sweden, aged 18 years and over (N = 936 449), we aim to identify risk factors for non-fatal diagnosed intentional self-injury. Groups at risk of repeat episodes of self-injury will be identified. METHODS: Information on hospital stays and outpatient specialized care visits registered as intentional self-injury was collected from the Region Skåne Healthcare database in 2007. These injuries were studied in relation to sociodemographic factors, previous disease, substance abuse and psychotropic drug treatment at baseline. RESULTS: There were increased odds of diagnosed intentional self-injury during follow-up in association with being single, of young or middle age, having low income and being born in the Nordic countries. Presence of neurological or psychiatric disease, substance abuse and previous assault-related injury were also strongly associated with future intentional self-injury. The use of psychotropic drugs showed a clear dose-response relationship with intentional self-injury during follow-up. Those diagnosed with self-injury in the 3-year period before baseline had more than 10 times increased odds of a new episode of intentional self-injury. The odds of repeated episodes of self-injury among subjects born in Europe, but outside Sweden, were less than half those seen for subjects born in Sweden. CONCLUSIONS: The present study, based on a total general population, expands the knowledge base regarding intentional self-injury in adults, repeat behaviour and its associations with sociodemographic variables, substance use and disease in both men and women.