Cognitive predictors of violence in schizophrenia: a meta-analytic review

Abstract Background Aggression committed by patients with schizophrenia and other serious and persistent mental illnesses represents a major public health concern affecting patients, their families, treating clinicians as well as the community at large. Cortical dysfunction has been implicated as an anatomical correlate of acts of aggression as well as a fundamental feature associated with individuals with schizophrenia (SZ). As a result, examination of neurocognitive deficits may serves as a natural experiment to explore the relationship between cognition and aggression committed by SZ patients. Past studies, however, have yielded inconsistent and complex results regarding the relevance of cognitive impairment to aggressive behavior. Objective Despite a fair number of studies in the literature, there have been no statistical reviews conducted to date examining the association between cognitive deficits and aggression in SZ. The present meta-analytic study examined the relationship between cognitive impairment and SZ acts of aggression. Methods Electronic databases were searched up to April 2013 using the words and word stems "aggress*, psychotic, risk, cognit*, neurocognit*, and neurobiological." The search resulted in 29 studies with independent samples. Information was extracted regarding study sample and methodological characteristics in addition to aggression prediction, and comprehensive meta-analytic procedures were performed. Inter-rater reliability for coding was good to excellent. Results The meta-analysis (4764 participants) demonstrated heterogeneous results, leading to follow-up comparisons. Results revealed that SZ cognitive impairment exerted a significant risk for aggression, across studies with differing methodologies. Global cognitive impairment and lack of insight emerged as significant risk indicators for aggression, accounting for 2% of the variance. Conclusions It was concluded that measurement of patients' global cognitive ability adds incremental variance in the comprehensive assessment and prediction of SZ violence risk.