Abstract
BACKGROUND: Childhood traumas and low theory of mind abilities have been shown to be associated with violence, crime and schizophrenia. Understanding the factors that predispose to violent behaviour by patients with schizophrenia is important for treatment and safety. AIMS: To investigate relationships between childhood trauma, theory of mind and violent behaviour in patients with schizophrenia and in a healthy comparison sample. Our hypothesis was that patients with schizophrenia who had been violent would be more likely to have a history of childhood trauma and show impairments in theory of mind impairments than either non-violent patients with schizophrenia or healthy non-violent people. METHODS: In a cross-sectional design, we recruited 30 patients with schizophrenia who had a history of violent crime, 50 patients with schizophrenia but no history of violence and 50 healthy people without history of violence. Each participant completed the positive and negative syndrome scale, childhood trauma questionnaire, the reading the mind in the eyes test and the hinting task. RESULTS: Patients were significantly more likely to be male, without employment and on low income than healthy controls with a suggestion of the violent patient group being worst off. Childhood trauma history and theory of mind tests distinguished the schizophrenia groups from healthy controls but not from each other. Logistic regression analysis, comparing the schizophrenia groups only, confirmed that male sex and number of prior hospitalisations were the only two characteristics that independently distinguished the violent from the non-violent groups with schizophrenia. CONCLUSIONS: Childhood trauma histories and theory of mind test results differed only between people with schizophrenia and healthy prosocial adults of similar age, but did not distinguish between the violent and non-violent people with schizophrenia. Whereas a pathway to violence in the context of schizophrenia from early trauma through impaired reading of others' emotions seems plausible, it still lacks evidence. Our findings suggest good reason to assess and treat impairments of emotional perception and processing in people with schizophrenia, but that the need is unlikely to be specific to those who become violent. This needs further research with larger samples.