By Gary Evans, Medical Writer

The widely reported1 reduction in the number of the nation’s mental health facilities has led to an increase in agitated and potentially violent patients in EDs.

Coinciding with this increase, healthcare facilities are also dealing with violence associated with the opioid epidemic. Dealing with these challenges and others, healthcare workers may find themselves having to restrain mental health patients, which creates an ethical dilemma, explains Ambrose Wong, MD, MSEd, an assistant professor of emergency medicine at the Yale School of Medicine in New Haven, CT.

“We are restraining, on average, four or five patients a day — which is a lot,” he says. “Anecdotally, there are other health systems that are doing this even more frequently.”

The traditional practice of using sedation and restraints for agitated patients can have long-term consequences for the mentally ill, who may be homeless, intoxicated, suffering from delirium, or dealing with dementia. The immediate threats of restraint include chest trauma, aspiration, respiratory depression, asphyxiation, and cardiac arrest, Wong and research colleagues note.2 A secondary effect is that these patients may avoid subsequent care after being restrained for treatment.

“Physical restraints can have lasting psychological implications,” he says. “We are learning that from both talking to patients who have [been restrained] and sometimes seeing it [in the data] — they are much more reluctant to come back to the ED or any psychiatric service.”

Healthcare workers are nothing if not compassionate, making it all the more difficult to put patients through something that may increase harm, he says.

“They come to us looking for help, so the last thing we want to do is make them feel we were holding them down against their will,” Wong says.

“It’s all linked to having less access to psychiatrists and mental health services, and the fact that the burden of mental health is increasing overall just in terms of people needing access and being sicker.”

The Paradox

Trying to help these agitated patients, who still must be assessed for injuries or other conditions, can be emotionally and mentally challenging for healthcare workers.

“The staff really feel sometimes that the only way to make sure that they and other patients are safe is to do these restraints,” he says. “But what we found when we interviewed the staff is something called the ‘care paradox,’ which is that they almost have a conflict of interest that they have to choose between their own safety and the patient’s safety.”

The result can hamper the clinical teamwork needed to safely treat this marginalized population. “Hierarchy and professional silos hinder coordinated care between healthcare professionals,” the researchers report.3 “Barriers to effective teamwork appear to exacerbate safety threats when patients present with acute agitation.”

The care paradox can contribute to feelings of burnout, an increasing problem in the demanding world of healthcare. “Experts have suggested strategies to promote collaboration between healthcare professionals in behavioral emergency care, including the implementation of a structured team approach in the form of a ‘rapid response team’ and formal delineation of roles and responsibilities,” Wong and colleagues report.

In general, clinicians try to avoid these coercive measures with agitated paitents if at all possible. For some patients, it is unavoidable.

“Sometimes, patients come into the ED so acutely agitated that we cannot de-escalate them,” Wong says. “It’s a combination of the fact that they may be psychotic or very intoxicated, and there are certain substances that are more agitating than others, such as PCP or ‘angel dust.’”

In terms of maintaining workflow and protecting other patients, it sometimes is necessary to restrain and sedate these agitated patients.

“They become calm and you can move on to other tasks in the ED,” he says. “But the tradeoff is restraining these patients potentially has lasting consequences. The staff know that, but you have to make that decision very quickly.”

In reviewing cases, Wong says there are some patients who are restrained too early and others where a delay results in a staff injury.

“It is not an easy line to draw,” he says. “In order protect yourself, you might have to sacrifice the patient’s safety a little bit — and that can lead to burnout.”

Mental health experts — like those in Project BETA (Best practices in Evaluation and Treatment of Agitation)4 — are trying to assist in early recognition of agitated mental health patients and improve de-escalation.

“Many facilities now use techniques such as intervention teams, which are paged instantly when there is an agitated patient, or ‘management of assaultive behavior’ protocols that seek to engage patients into voluntarily accepting treatment,” the Project BETA authors report. “However, far too many agencies still treat all episodes of agitation in a fashion that might best be described as ‘restrain and sedate.’”

The Project BETA guidelines and approaches to the problems are based on some of the fundamental tenets of emergency psychiatry, including:

  • rule out any other medical causes of agitation symptoms;
  • rapidly stabilize the patient;
  • do not coerce patients into complying;
  • find the least-restrictive setting for treatment;
  • create therapeutic alliance.

“They did a really good job of listing those best practices, but the adoption of it is variable,” Wong says.

Patient or Criminal?

Given the daily demeaning attacks and horrific assaults now frequently reported in healthcare, it can be tempting to view violent patients as criminals who should be punished, he says.

“Sometimes, in the world of occupational health and employee health, I hear experts talk about strategies that could potentially sacrifice patient wellness,” Wong says. “For example, there are some folks who talk about prosecuting these individuals, locking them up, passing state and federal laws that persecute patients that assault healthcare workers.”

This is a short-term fix for an admittedly horrible problem, Wong says, and one that does not really address the bigger question: Why is this happening?

“They really need our help to get the mental health service they need,” he says. “Throwing them in prison is not necessarily the right answer. It reinforces the stigma that these patients already have. It may make them feel that the healthcare system isn’t necessarily where they want to go — but they really need our help.”

One problem with developing violence prevention methods in healthcare in general is the lack of rigorous research and studies on patient safety, Wong says.

Randomized Trials Examine Violence

One of the few randomized controlled interventions to reduce violence against healthcare workers showed the efficacy of thorough data collection and the innovation of inviting individual units to solve perceived hazards.5

As federal lawmakers return to the issue of requiring violence prevention programs in healthcare, the 2017 study is being revisited, most recently in a webinar by The Joint Commision.6

The lead author of the study is Judy Arnetz, PhD, MPH, PT, professor and associate chair for research in the Department of Family Medicine at Michigan State University’s College of Human Medicine. “The importance of this study is that we did have some control sites,” she tells Hospital Employee Health. “Randomized controlled studies are still the gold standard in research.”

Arnetz and colleagues randomized 41 units into intervention (21) and control (20) groups. The intervention units received unit-level reports of violence and data to use to develop interventions. The control units did not receive the incident data.

“Six months post-intervention, incident rate ratios of violent events were significantly lower on intervention units compared to controls,” Arnetz reported in the paper. “At 24 months, the risk for violence-related injury was lower on intervention units, compared to controls. This data-driven, worksite-based intervention was effective in decreasing risks of patient-to-worker violence and related injury.”

A survey of emergency physicians last year found that half have been assaulted while at work in the ED, while more than 70% have witnessed another assault. Only 10% experienced neither. (See HEH, April 2019.)

“Emergency departments are definitely known for being at increased risk,” Arnetz says. “One of the problems is it is so common and expected that it becomes a barrier to reporting.”

Using a hazard-risk matrix as a tool for identifying increased risk and collecting data on reported incidents, the researchers found that four of the five EDs in the study met high-risk criteria.

“The [outlying] emergency department was in an inner-city hospital in a tough area,” she says. “But they were in the categories of low probability of violence and medium severity. Staff were definitely getting injured, but the reason it was low probability was because they were not reporting.”

Data collected from reported incidents eventually were shared with the individual units, who were asked to address the findings.

“One emergency department requested increased lighting in the parking lot surrounding the ED,” Arnetz says. “That was an environmental strategy. One of the issues was that patients who were discharged, or family members, sometimes came looking for staff. They were angry or upset.”

Another ED looked at its data and decided to balance work schedules better to avoid vulnerable periods when few staff were on the unit, she adds.

“Additional ED strategies included mandatory team-building classes, active shooter training, and customer service training,” she says. “That was training of staff in a way that could help them become more sensitive to their patients’ needs and to be more aware.”

Overall, Arnetz stresses the importance of a violent incident reporting system, as data will drive and pinpoint prevention measures.

“I emphasize the importance of a central reporting system,” she says. “Hospitals can at least develop that — or if that is not a realistic goal, they can try to collate data from the various sources. For example, security usually collects their data and occupational health services collect data on staff that come to them with violence-related injuries.”

In addition to developing a good reporting system, there must be stakeholder engagement — both management and workers — to prevent violence in the facility.

“Healthcare workers need to know that there is a belief in developing a culture of safety that is supported by management action,” Arnetz says. “By involving employees in the process, especially with improvements at the unit level, you increase the chances of developing effective interventions.”


  1. Raphelson, S. How The Loss Of U.S. Psychiatric Hospitals Led To A Mental Health Crisis. National Public Radio, Nov. 30, 2017. Available at:
  2. Wong AH, Ray JM, Lennaco JD, et al. Workplace Violence in Health Care and Agitation Management: Safety for Patients and Health Care Professionals Are Two Sides of the Same Coin. Jt Comm J Qual Saf 2019; 45:71–73.
  3. Wong AH, Combellick J, Wispelwey BA, et al. The Patient Care Paradox: An Interprofessional Qualitative Study of Agitated Patient Care in the Emergency Department. Acad Emerg Med. 2017;24(2):226-235. doi: 10.1111/acem.13117.
  4. Holloman GW, Zeller SL. Overview of Project BETA: Best practices in Evaluation and Treatment of Agitation. West J Emerg Med 2012;13(1):1–2.
  5. Arnetz JE, Hamblin L, Russell J, et al. Preventing Patient-to-Worker Violence in Hospitals: Outcome of a Randomized Controlled Intervention. J Occup Environ Med 2017;59(1):18–27.
  6. The Joint Commission. Workplace Violence Prevention: Implementing Strategies for Safer Healthcare Organizations. July 31, 2018. Available at: