Behavioral strategies to mitigate violent behavior among inpatients with severe mental illness: A literature review

Prior studies have shown that while most people with mental illness are not violent, having a mental illness does increase a person’s risk of violent behavior [1]. During inpatient psychiatric hospitalizations, the prevalence of aggressive behavior is much higher, ranging from 8–44% [2]. Several studies have focused on pharmacotherapy strategies to reduce aggressive or violent behavior (e.g., valproic acid and clozapine) [3], but hospital administrators could benefit from evidence-based suggestions to guide violence risk reduction program development to complement medications while serving high-risk and acutely ill psychiatric patients on inpatient units.

To explore the evidence base of behavioral interventions to reduce violence for people with a history of violence and severe mental illness, a systematic literature review was performed using PubMed/MEDLINE. Inclusion criteria included English-language publication, adults hospitalized or recently discharged from an inpatient psychiatric unit or state forensic hospital, and an emphasis on behavioral strategies that targeted violent behavior. Titles, and then abstracts, were examined to determine if they met inclusion/exclusion criteria. We searched citations of review articles meeting criteria for missing references. We supplemented our literature review by interviewing forensic psychiatry experts in violence risk assessment.

Our search identified only 13 articles. Of these articles reviewed, 31% (4/13) focused exclusively on inpatient psychiatry (see online Supplemental Table 1). While several authors recommended techniques such as containment strategies (e.g., locked units, verbal redirection, behavioral contracts) or behavioral methods involving positive or negative reinforcement, there were only two randomized controlled trials (RCT) examining behavioral interventions to reduce violence in psychotic patients [4]–[5]. Both of these RCT interventions were based on CBT principles and did show efficacy, but unfortunately had multiple limitations including small sample sizes, selection bias, heterogeneous samples which could limit the applicability of research findings, an inability to control for contributing personality factors, and potential treatment contamination at study sites. In addition, all studies provided limited guidance on whether certain behavioral strategies may be more effective in people with specific psychiatric diagnoses.

Despite the limitation of these findings, the publications and expert consultants emphasized the importance of a formal violence risk assessment in developing an effective plan to manage violent behavior among inpatients with severe mental illness. They also highlighted the importance of medication adherence and reducing co-morbid substance use.

Our review revealed a dearth of high-quality interventional studies to examine the efficacy of behavioral techniques to reduce violence among psychiatric inpatients. There was also limited information on whether specific behavioral strategies may be more effective in patients with a specific diagnosis or etiology of mental illness. Large randomized trials are desperately needed to compare the effectiveness of different behavioral techniques. Given this lack of evidence and the prevalence of violence on inpatient psychiatry units despite medication management, these RCTs could help guide hospital administrators in the development of programs to better serve this patient population and protect them—and the staff serving them—from violence.