Emergency Room Workplace Violence: Electronic Behavioral Alerts May Be Prone to Bias

‐

By Christopher Cheney

Patient behavioral alerts in electronic medical records designed to mitigate workplace violence may perpetuate systemic inequities, according to a recent research article.

In a poll funded by the American College of Emergency Physicians, two-thirds of emergency physicians and 70% of nurses said they had been physically assaulted at work in the prior year. Patients were perpetrators in 97% of the workplace violence incidences in the poll. The poll found hitting, spitting, and punching were the most common kinds of physical assaults.

The recent research article, which was published in Annals of Emergency Medicine, features data collected from nearly 3 million emergency department visits at 10 EDs from 2013 to 2022.

The study includes several key data points:

  • Out of the 2,932,870 ED visits, 6,775 (0.2%) generated electronic behavioral alerts for 789 patients
  • Out of the ED visits with electronic behavioral alerts, 5,945 (88%) were determined to have a safety concern involving 653 patient perpetrators
  • Among patients with safety-related electronic behavioral alerts, the median age was 44, 66% were men, and 37% were Black
  • In subsequent ED visits, patients with safety-related behavioral alerts had higher rates of discontinuance of care (7.8% versus 1.5% for patients with no alert) such as leaving without being seen
  • The most common incidents that prompted electronic behavior alerts were physical abuse (41%) or verbal abuse (36%)
  • Black non-Hispanic patients were more likely to be the subject of an electronic behavioral alert than White non-Hispanic patients (odds ratio 2.60)
  • Patients younger than 45 were more likely to be the subject of an electronic behavioral alert than patients 45 to 64 years old (odds ratio 1.41)
  • Male patients were more likely to be the subject of an electronic behavioral alert than female patients (odds ratio 2.09)
  • Publicly insured patients were more likely to be the subject of an electronic behavioral alert than commercially insured patients (Medicaid, odds ratio 6.18)

The data indicates electronic behavioral alerts may perpetuate bias against historically marginalized groups, the study’s co-authors wrote. “In our analysis, younger, Black non-Hispanic, publicly insured, and male patients were at a higher risk of having an ED electronic behavioral alert. Although our study is not designed to reflect causality, electronic behavioral alerts may disproportionately affect care delivery and medical decisions for historically marginalized populations presenting to the ED, contribute to structural racism, and perpetuate systemic inequities.”

Interpreting the Data

ED clinicians may make biased decisions about agitated patients, the study’s co-authors wrote.

“When agitation occurs, clinicians are required to rapidly diagnose potential causes and intervene to minimize any physical danger, apply verbal and behavioral techniques to deescalate behavior, and assess the need for any coercive measures such as physical restraints. … This combination of physical danger with a need for quick decision making may lead to reinforcement of biases during agitation events. Indeed, emergency clinicians have expressed frustration and negative attitudes toward individuals with substance use and mental illness, and racial and ethnic minorities are particularly vulnerable to negative outcomes in the ED,” they wrote.

Electronic behavioral alerts may increase bias and discrimination, the study’s co-authors wrote. “Electronic behavioral alerts with frequent notifications may only exacerbate biases given that we found their disproportionate application to sociodemographic minorities. Moreover, the fact that patients with electronic behavioral alerts have higher rates of care discontinuance suggests the possibility that these patients are treated differently or more quickly dismissed.”

The data shows that workplace violence is a serious issue, but the best interventions to address specific patients are unclear, the study’s co-authors wrote. “What remains unknown is what form and timing these interventions should have, which appropriately balance the risks to safety and mitigate propagating disparities. These interventions may take the form of focused meetings with patients outside of the acute encounter, auditing of electronic behavioral alert placement to ensure accuracy and fairness, continuing review to assess further need of the electronic behavioral alert as the risk of violence decreases, and removal of potentially biased or stigmatizing language.”

Christopher Cheney is the senior clinical care​ editor at HealthLeaders.