Risk Factors for Violent Crime in Schizophrenia: A National Cohort Study of 13,806 Patients

Objective: To determine risk factors for and prevalence of violent crime in patients with schizophrenia, and in particular, to explore the contribution of familial risk factors.

Method: We designed a cohort study that followed up patients with 2 or more hospitalizations for schizophrenia (ICD-8, ICD-9, and ICD-10 criteria) and investigated the risk for a violent conviction using Cox proportional hazards models. All 13,806 patients with 2 hospital discharge diagnoses of schizophrenia from January 1, 1973, through December 31, 2004, in Sweden were followed until violent conviction, emigration, death, or end of follow-up (December 31, 2004), and associations with sociodemographic, individual (substance abuse comorbidity, and previous violence), and familial (parental violent crime and parental alcohol abuse) factors were examined.

Results: Over an average follow-up period of 12 years, 17.1% (N = 1519) of the men and 5.6% (N = 273) of the women with 2 or more hospitalizations for schizophrenia had a violent conviction after discharge from hospital. Familial risk factors had moderate effects, increasing the risk for violent convictions by 50% to 150%. After adjustment for sociodemographic and individual risk factors, the associations between parental violent crime and risk of violent convictions remained in men (adjusted hazard ratio [HR] = 1.65, 95% CI = 1.33 to 2.04) and in women (adjusted HR =1.83, 95% CI =1.11 to 3.01), whereas parental alcohol abuse was no longer significantly associated with violent crime.

Conclusion: Parental violent crime had moderate associations with violent crime in male and female offspring with at least 2 hospitalizations for schizophrenia, which were mostly stronger than the better documented sociodemographic risk factors. This suggests that familial (genetic or early environmental) risk factors have an important role in the etiology of violent offending among individuals with schizophrenia and should be considered in violence risk assessment.