Abstract Background Healthcare workers face workplace violence (WPV) at rates nearly five times higher than other industries. Traditional hospital responses to WPV, such as security involvement or zero‐tolerance policies, may escalate patient distress and contribute to racialized harm. Behavioral Emergency Response Teams (BERTs) have emerged as an alternative, yet their implementation and outcomes in nonpsychiatric, inpatient settings remain poorly characterized. Objective This scoping review synthesizes the literature on BERT implementation in adult and pediatric nonpsychiatric inpatient and emergency department settings, with attention to team design, training, implementation, evaluation, and equity considerations. Methods This review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses extension for Scoping Reviews (PRISMA‐ScR) guidelines. Five databases and gray literature were searched for studies on BERTs through May 2024. Data extraction and critical appraisal were conducted independently by multiple reviewers using validated tools. Results Ultimately, 28 studies were selected for inclusion. Most described BERTs in academic settings ( n = 19) with diverse team compositions, often including security personnel ( n = 22). Over half of the studies reported staff training in de‐escalation ( n = 16), though few detailed curricula. Overall, BERT implementation was associated with decreased WPV, restraint use, and security calls. Staff reported increased perceptions of safety and institutional support. However, reporting on equity metrics was limited; only eight studies tracked race, and some suggested disproportionate BERT interventions and restraint use among Black patients. Conclusion BERTs are a promising intervention to manage behavioral emergencies, reduce WPV, and reduce security calls. Standardized guidelines, training, and equity‐focused evaluation will be essential to optimize BERT effectiveness and ensure equitable implementation.
Mendelsohn et al. (Thu,) studied this question.