Absolute and relative rates of both violent victimisation and perpetration amongst people with mental illness in a whole of population Danish national register-based study

Dean K^1,2, Marr C¹

¹University Of New South Wales, ²Justice Health and Forensic Mental Health Network

Biography:

Prof. Kimberlie Dean is Chair of Forensic Mental Health at UNSW where she is also Head of the Discipline of Psychiatry and Mental Health. She leads the Forensic Mental Health Research program at Justice Health and Forensic Mental Health Network in NSW, where she also works clinically as a Forensic Psychiatrist.

Risks of both violent victimisation and perpetration are known to be elevated amongst people with mental illness, compared to the general population. It is often claimed that the risk of victimisation is greater than perpetration but evidence to support this is limited. There are few studies that examine both outcomes in the same sample and also includes individuals with mental illness and those from the general population without illness. In a population-based study using data from Danish national registers, police-recorded victimisation and perpetration was examined in relation to mental health service contact across the full diagnostic spectrum. The relative risk of police-reported victimisation was higher amongst those with mental health service contact compared to those without contact but the association found was not greater than that seen for perpetration. Those with mental health service contact for substance use problems and personality disorders had the highest relative risks of both victimisation and perpetration. Women had particularly high relative risks of both outcomes and their absolute risks (cumulative incidence) of victimisation were actually higher than for perpetration (not so for men). Estimates of risks of violent victimisation and perpetration amongst those with mental illness compared to the general population are likely to depend on factors such as the source of outcome data. While police-recorded events can be examined in large population-based samples, avoiding information biases associated with interview-based studies, some occurrences of victimisation, and of perpetration, are likely to remain unreported to police, perhaps particularly for those with mental illness. Both violent victimisation and perpetration represent very poor outcomes for people with mental illness and remain priorities for evidence-based intervention.