Risk factors for filicide and homicide: 36-year national matched cohort study.

OBJECTIVE: To identify specific risk factors for filicide, parental killing of one's own children.

METHOD: A nationwide matched cohort study of filicides, including filicide-suicide, was conducted in Sweden 1973-2008 (N = 151). We linked longitudinal national registry data on offender psychiatric and neurologic morbidity (ICD codes), suicide attempts, traumatic injury, and previous criminality. We compared filicide offenders and their children to age- and sex-matched general population controls and their children. To compare homicide perpetrator subtypes, parallel analyses were made with nonfilicide homicide offenders (N = 3,979) and matched controls.

RESULTS: History of major psychiatric disorder, defined as psychotic, affective, or personality disorder, was more common among filicide perpetrators even after adjusting for socioeconomic factors and other potential perpetrator and child confounders (adjusted odds ratio [AOR] = 8.6; 95% CI, 3.7-20.0). Substance misuse was not an independent risk factor, whereas prior suicide attempt conferred a marked increase of filicide risk (AOR = 11.6; 95% CI, 4.0-33.3). Multiple birth was the only child risk factor for filicide that was significant (AOR = 4.8; 95% CI, 1.3-17.6) in multivariate analyses. Overall, nonfilicide homicide offenders exhibited similar risk factors as filicide perpetrators; additionally, substance misuse and violent crime were significantly more frequent and suicide attempt less common (nonoverlapping 95% CIs) than in filicide offenders.

CONCLUSIONS: Major psychiatric disorder and previous suicide attempt were strong independent risk factors of filicide; however, substance misuse and prior violent crime were less common than among nonfilicide homicide offenders. Filicide prevention attempts should focus major psychiatric disorder and suicidal behavior in particularly strained parents, instead of more general risk factors for interpersonal violence such as substance misuse.