April 29, 2024
HOMERuN Collaborative: Impact of EHR Behavioral Alerts on Healthcare Workplace Safety and Patient Safety
The Hospital Medicine Reengineering Network (HOMERuN) is a rapidly growing collaborative made up of more than 50 Hospital Medicine groups from academic and non-academic hospitals across the United States.
Organizers and Facilitators: Himali Weerahandi, Zoƫ Kopp, Andrew Auerbach, Marisha Burden, Angela Keniston, Katie E. Raffel, Jeff Schnipper

Background: Violence against healthcare workers is a significant problem. In 2022, the Joint Commission established new workplace violence prevention requirements for hospitals. Electronic health record (EHR) behavioral alerts are a common approach that healthcare organizations use to flag patients at risk for violent behavior. However, concerns about behavioral alerts are emerging. Behavioral alert placement is often subjective, and alerts are placed differentially based on patient identity. The governance around behavioral alerts (i.e., who monitors them, specific indications for placement, what their content should include, and when the flags should sunset) is also poorly understood. Behavioral alerts may also cause harm. They often contain stigmatizing language and clinician exposure to stigmatizing language about a patient can negatively influence care.

Little is known about how behavioral alerts impact hospitalized patients. To address this knowledge gap, we conducted focus groups of hospitalists to understand how EHR behavioral alerts are applied across healthcare workplaces, clinician perspectives on EHR behavioral alerts on preventing workplace violence, and variation in programs to prevent workplace violence in healthcare settings.
How Behavioral Alerts Are Used
While many institutions use Epic as their EHR, the use of EHR behavioral alerts varies greatly between institutions. Some participants noted there were no alerts at all; others noted a sidebar alert or flag that needed to be clicked on; others noted pop-ups that appeared every time the chart was opened and required a response from the end user.
"It's not really clear who puts them in or what the triggers are. It seems more like some people know how to put them in and some people don't."
Impact of Behavioral Alerts
  • The effectiveness of behavioral alerts in preventing future violence is unclear.
  • Alerts that were placed years ago that influence care now (i.e., if additional security is being considered or denial of post-acute care) can cause patient distress if they become aware of these circumstances.
  • Some clinicians state that the alerts do not impact their care. Others comment that the alerts may prompt them to check in with the bedside team about their comfort level with the patient. Others observed that they may receive more questions from nurses about patients with behavioral alerts. Some clinicians note the alert may alter their behavior during the patient encounter, for example, they make sure they are in between the door and the patient when interviewing the patient or ask for other staff to accompany them during the physical exam.
  • Some clinicians are aware of the biases that may come from these alerts, and the alert prompts them to check their biases.
"I don't really know how useful they are. I take care of a lot of patients with substance use disorders and I find that they have many, many behavioral alerts, probably put in when patients were experiencing withdrawal and were not adequately treated. So, I typically ignore them quite honestly."
  • Largely impact minoritized populations and already stigmatized populations, and these alerts may further stigmatize them.
  • Once placed, significant inertia to remove them: no one "owns" the alerts, no one wants to remove or disable them, even if they are no longer relevant.
  • Lack of guidance or governance on alert placement can lead to alerts being placed that are more venting about a challenging encounter than guidance on how to manage a patient's future behaviors. 
Workplace Violence Prevention and Solutions
  • There seems to be a double-edged sword with approaches to workplace violence prevention and solutions: the need to protect both workers and patients.
  • Need for triaging different types of events (i.e., if a weapon is involved, an alert should never sunset).
  • Behavioral health teams who see patients with challenging behaviors are helpful, especially when the team leaves actionable items in the chart on how to engage a patient if the encounter is not going well, how to de-escalate, or when to leave and come back later.
  • Complex intervention units which cohort disruptive patients and have a psychiatrist on the unit co-managing is another approach some hospitals are using, but also may be stigmatizing.
"Right now it's deferring more to the patient rather than protecting the staff. But I think people are struggling with the difficulties we're struggling with too, which is you want to provide patient-centered care. You want to provide care in a thoughtful way, in an evidence-based way. And we don't have a lot of evidence on how to predict and how to appropriately manage."
Next Steps
  1. Develop a conceptual model on how behavioral alerts and workplace violence prevention programs impact patients and the healthcare workforce.
  2. Map out potential data sources to investigate the impact of behavioral alerts and workplace violence prevention programs on outcomes relevant to patients and workers.
Our next meeting will be on May 10, 2024.
Image Attributions: Vector images from vecteezy.com
Check out the HOMERuN website for more information.
If you would like to join the HOMERuN Collaborative calls, please reach out to Tiffany.Lee@ucsf.edu.