Violence and homicidal behaviors in psychiatric disorders.

Our review evaluating the relationship between violent/homicidal behaviors and mental illness/psychiatric disorders used many different data including that assessing the prevalence of violent/homicidal behaviors in former psychiatric inpatients (just before hospitalization, during hospitalization, and after discharge) as outpatients and in community samples as well as evaluating the prevalence rate of psychiatric disorders in people who actually engaged in violent/homicidal disorders (jail detainees, prison inmates, and community samples). Irrespective of which line of investigation, there was convincing evidence that violent/ homicidal behavior was associated significantly with mental illness. Although earlier investigations failed to control for important variables, such as age and sociodemographics, most studies reviewed in this article did control for these items, further underlining the association of violence and mental illness. The question of whether specific psychiatric diagnostic categories are associated with violent/homicidal behavior is less definite across the various studies reviewed. The presence of substance abuse and dependence and alcohol abuse and dependence as well as antisocial personality disorder are particularly associated with an increased risk of violent/homicidal behaviors. The risk for these latter behaviors in schizophrenia, mood disorders, and anxiety disorders may appear somewhat greater than that for a general population but are not of the same magnitude of that for substance abuse or antisocial personality disorder. Interestingly, our outpatient study found that homicidal behaviors were not associated with any specific psychiatric diagnosis. Although understanding whether specific psychiatric diagnostic categories are more prone to violent behaviors may be of importance, most studies have been shortsighted regarding this evaluation. All the studies presented in this article except the ECA project, presented diagnostic data where either the presence of one psychiatric disorder did not preclude the diagnosis of another or assigned subjects/patients into the severest disorder of a predetermined hierarchy of diagnoses or only selected their principal/primary diagnosis. Thus, the effect of having a solitary psychiatric disorder (only one disorder present) as well as the effect of comorbidity per se on the relationship of psychiatric disorders and violent/homicidal behaviors were unexplored. Only the ECA study by Swanson and colleagues reported on the effect of comorbidity. As reviewed earlier in the article, Swanson et al found that comorbidity of psychiatric diagnostic categories further increased the risk of violent/ homicidal behaviors. In most cases, it was many more times than simply adding the rates of either diagnosis alone. Because more than 54% of respondents of the National Comorbidity Survey study who had one DSM-III-R diagnosis also had at least a second Axis I diagnosis, the association of violent/homicidal behaviors to mental illness may even be stronger than originally believed. Within the relationship of violent/homicidal behaviors and mental illness, this article suggests a number of particular risk factors. As just reviewed, substance/alcohol abuse and antisocial personality disorder as well as the presence of comorbid psychiatric disorders are significant risk factors. Which particular comorbid illness increases the risk still needs further elaboration. Studies must continue to try to define and understand the relationship of violent/homicidal behaviors in mental illness. Although mental disorders per se are significantly associated with violent/homicidal behaviors, it is reasonable to believe that targeting certain subgroups of patients should be helpful. Probably the presence of psychotic symptoms is a significant risk factor in violent/ homicidal behaviors in the mentally ill. Only one of the studies reviewed in this article evaluated this issue. (ABSTRACT TRUNCATED)