Violence Risk Screening in an Adult Emergency Department

Abstract

Problem: Almost half of emergency physicians report being physically assaulted at work, while about 70 percent of nurses report being hit and kicked while on the job. Purpose: The purpose of this quality improvement project was to implement violence risk screening via the Broset Violence Checklist (BVC) during triage of all patients presenting to an Adult Emergency Department. Methods: A series of formal in-services were held over a one-week period at times covering both day and night shifts. Information regarding the checklist was included in daily shift huddle announcements and emailed as a PowerPoint presentation to staff. Retrospective chart reviews were conducted weekly during the 15-week implementation period to assess adherence to the tool. Data from monthly violent restraint audits were also collected. Results: Adherence to completion of the BVC at AED-1 was significantly lower than that of comparison AED-2 throughout the implementation period. Multiple logistic regression was performed to assess strength of association between restraint status and BVC score. The odds of restraints were less likely in patients who scored on the boisterous (OR 0.06, p = 0.026) and irritable (OR 0.016, p = <0.001) characteristics and more likely in patients who scored on the verbal threat characteristic (OR 68.26, p=0.016). Conclusion: It is feasible to implement violence risk screening at the time of triage in an adult emergency department. Adherence rates were lower in AED-1 due to barriers such as high burden of documentation for triage nurses and difficulty finding the checklist within the triage navigator. Verbal threats were associated with future violent restraint usage. Total BVC score was found to be not statistically significant correlated with future restraint use.