Association with Aggression. Consistently reported, modest, but significant, associations have been found between schizophrenia-spectrum conditions and aggression. Interpreting this relationship can be complicated because of the different aggressive outcomes used, ranging from criminal conviction for aggressive offences to aggressive acts committed in the community or prison/hospital services (see the Aggression in Inpatient Mental Health Services Below for a review). Two meta-analyses reported that the odds of being convicted for an aggressive crime (Xxxxx et al., 2009a) and/or behaving aggressively in the community or institutional settings (Xxxxxxx et al., 2009) were up to seven time higher among those with schizophrenia- spectrum conditions compared to general population controls. Base rates of aggression were relatively low at 1.6% however, indicating that aggression is still uncommon even with the additional risk conferred by schizophrenia-spectrum conditions. There was also significant heterogeneity in the reported effect sizes in these studies. Xxxxx et al. (2009a) reported that the increased risk of aggression appeared to be predicted by substance abuse, with no significant differences in risk of aggression observed between those with substance abuse alone versus those with comorbid schizophrenia-spectrum conditions and substance abuse. Xxxxxxx et al. (2009) also concluded that that the increased risk arising from schizophrenia- spectrum conditions was like that of other established risk factors, such as previous aggression. Similar findings have been reported in longitudinal investigations using national registry data. After adjusting for substance abuse, schizophrenia-spectrum conditions were associated with a three- (men) and seven-fold (women) increased risk of conviction for an aggressive offence (Xxxxxxx et al., 2015). Xxxxx et al. (2009b) also reported that schizophrenia-spectrum conditions were associated with a significantly increased likelihood of conviction for an aggressive crime, compared to non-affected individuals, with comorbid substance abused conferring significantly greater risk. When controlling for genetic and/or shared environmental factors through comparisons with unaffected siblings, the relationship between substance abuse and aggression was significantly reduced. Another investigation using the Swedish national registry data (Xxxxx et al., 2014) indicated that, within five years of receiving a schizophrenia-spectrum diagnosis, 11% of men and 3% of wo...
Association with Aggression. Dissocial personality disorder is overrepresented among forensic populations. A meta- analysis of over 23,000 prisoners reported that 47% of men and 21% of women met diagnostic criteria for DPD, a 10-fold increase compared to rates in the general population (Xxxxx & Xxxxxx, 2002). High rates of DPD have also been reported in forensic mental health services. For example, Coid (2003) reported a prevalence of approximately 55% in high- secure services in the UK. Dissocial personality disorder is also strongly associated with community aggression and crime. In England and Wales over a 10 year period, 16% of homicides were committed by an individual with DPD (National Confidential Enquiry for Suicide and Homicide, 2014). It was estimated from a household survey of over 8,000 UK individuals that 24% of the risk of community aggression was attributable to DPD (Coid et al. 2006). Prospective studies have also shown dissocial personality disorder to be predictive of future aggression in inpatient settings. A retrospective review of over 500 service users in a forensic mental health service reported that the risk of engaging in aggressive behaviour in the previous year was double that for individuals diagnosed with DPD, compared to non- affected individuals (Xxxxxxx et al., 2009). Multiple studies have also reported that rates of aggressive reoffending following discharge from forensic mental health services are significantly higher, and occur over a shorter period of time among those with DPD (Xxxx et al., 2007; Xxxxxx et al., 2013).
