CHPR: Please tell us a little about yourself.

Dr. Knox: Previously I was the director of the psychiatric emergency setting at a large community health center, and later I was CMO of the Harris Center for Mental Health and IDD. At that time, I was also on the board of the American Association for Emergency Psychiatry and worked with Project BETA (Best Practices in the Evaluation and Treatment of Agitation) to produce articles on the best evidence for the treatment of agitation, including how to minimize the use of seclusion and restraints (S&R). At the time, Centers for Medicare and Medicaid Services (CMS) was citing many hospitals around S&R issues, so I co-wrote a paper to address this topic (Knox DK and Holloman GH Jr, West J Emerg Med 2012;13(1):35–40). Last year I joined the Department of Psychiatry at the University of Texas McGovern Medical School here in Houston.

CHPR: Why was CMS citing hospitals around S&R issues?

Dr. Knox: Hospitals that receive Medicare or Medicaid funds must follow CMS guidelines, and places get into trouble when CMS does a review and there’s no documentation, for example, to justify the need for restraints. CMS has extensive rules and regulations that need to be followed, all geared toward patient safety. Institutions can lose financial backing if CMS deems that a hospital hasn’t followed proper procedures (Editor’s note: CMS guidelines can be found at www.tinyurl.com/sm3adksw).

CHPR: How widespread is the use of physical restraint in psychiatric units and emergency departments (EDs)?

Dr. Knox: Any setting that treats acutely mentally ill patients is probably going to have incidents of S&R because these patients are often in very agitated states and display impaired judgment and aggressive behavior. The use of S&R is even more widespread in large urban areas where there’s a lot of drug use. Estimates of the use of restraints range widely, generally from about 4% to 20% in inpatient psychiatric wards and even higher, up to 30%, in EDs (Beghi M et al, Riv Psichiatr 2013;48(1):10–22; Downey LA et al, Gen Hosp Psychiatry 2007;29(6):470–474).

CHPR: I’ve noticed that the restraint data across hospitals vary widely, even when the patient populations are comparable. What factors are associated with higher use of S&R?

Dr. Knox: Understaffing, inadequately trained staff, and staff burnout are all linked with higher levels. And there’s less restraint use in environments where the staff get along well and have high levels of cooperation, compared with environments with high levels of hostility among staff (Vergallo GM and Gulino M, Psychiatr Clin Psychopharmacol2021;31(4):468–473). Patients are perceptive. If there’s a lot of bickering and disorganization among treatment staff, patients sense that and can feel more insecure and tense. Regular monitoring and debriefing of S&R incidents can often uncover staff morale as a root cause of rising trends.

CHPR: What types of patients are most likely to end up in S&R?

Dr. Knox: Patients with psychotic illnesses, like schizophrenia or bipolar disorder, are more likely. Young people ages 24–34 are twice as likely to be placed in restraints as other groups, particularly males; Black and Latino patients are also significantly more likely to be restrained (Carreras Tartak JA et al, Acad Emerg Med 2021;28(9):957–965). If you’re a large Black male, for instance, you may be perceived as more threatening, and staff might rush to use restraints rather than verbal de-escalation. It’s important for those of us who work in emergency psychiatry settings to be attuned to our cultural biases and ask ourselves whether we have an environment that’s inviting to people of diverse cultures.

CHPR: What are some risks from restraints?

Dr. Knox: These incidents are humiliating and traumatic. Patients still tell me, years after a restraint incident, of the trauma they experienced. Also, patients and staff can get injured, and in extreme cases, restraints can be deadly. There have been several deaths of patients in restraints, especially when they are in the prone position, during which staff can inadvertently restrict a patient’s breathing while applying physical restraints. And for staff who place the restraints, the experience can be emotionally scarring.

CHPR: Right. These are upsetting and difficult experiences for everyone involved.

Dr. Knox: And they interfere with the therapeutic relationship. If a patient is restrained in their first encounter with psychiatry, they’re more likely to be nonadherent to further mental health care. We must do whatever we can to make the encounter as nontraumatic and therapeutic as possible, especially when it’s a patient’s first encounter with psychiatry.

CHPR: What steps can hospitals take to reduce the use of S&R?

Dr. Knox: A key concept is that hospital staff, beginning with the leadership, must understand that S&R represents a treatment failure. It should only be implemented when everything else has been tried. Studies have shown that changes in the institutional culture, where S&R is recognized not as just another treatment modality but rather as a last-resort intervention, generally work. For example, a large study in Pennsylvania looked at the impact of changes to hospitals’ culture and attitudes about S&R. It found that the rate of seclusion dropped from 4.2 to 0.3 episodes per 1,000 patient days, the rate of restraint dropped from 3.5 to 1.2 episodes per 1,000 patient days, and the average length of time patients remained in restraints dropped from 12 hours to two hours (Smith GM et al, Psychiatr Serv 2005;56(9):1115–1122).

CHPR: It’s amazing how a change to a hospital’s culture can have such profound effects. Practically speaking, how do hospitals change their culture?

Dr. Knox: Hospital leadership must regularly communicate their nonrestraint values and monitor the use of S&R very closely. Once there’s an S&R incident, it’s important to debrief the incident with the staff involved and the patient. You want to see what went wrong, what went right, and how you could have intervened differently. I believe psychiatric units should have cameras so you can look at the footage. You want to review each incident with the people involved so you can learn from your mistakes and your successes. Other changes that help change a hospital’s culture include improving patient-to-staff ratios and increasing patient activities, like art therapy and exercise groups.

CHPR: We installed cameras a few years ago to review and learn from incidents on the unit. They’ve also been helpful when a patient claims a staff member mistreated or injured them, so we review those incidents also.

Dr. Knox: Yes, absolutely. We had a lot of pushback from some staff initially who said, “Oh, you’re spying on us.” I said, “No, they’re to monitor the treatment area and make sure everyone is safe, and also to help you when a patient makes an accusation about staff.” We’ve had patients who make accusations when they’re delusional, and the cameras help staff and administration see what actually occurred, often exonerating the accused staff.

CHPR: What else can we do to reduce the use of restraints?

Dr. Knox: We need to make sure we provide a healing environment. I’ve visited a number of psychiatric emergency settings and the spaces are often very cramped, with bright lights and garish colors. Psychiatric assessment areas are often shoehorned into the leftover smaller spaces in the ED. Ideally the environment should have natural lighting, soothing colors, and lots of space so patients can move around. Another important point is to try to establish trust and rapport during the psychiatric intake process. Sometimes this process can feel like the patient is being booked into jail—it’s important to reassure patients that we’re here to help.

CHPR: We recently spoke with Dr. Kim Nordstrom about the EmPATH units that Dr. Scott Zeller created and that are expanding around the US. EmPATH units focus on having a calm, open milieu with lots of natural light, similar to what you’re proposing (Editor’s note: See EmPATH unit Q&A in this issue).

Dr. Knox: Yes, these therapeutic environments can help de-escalate agitated patients, but it’s also helpful for the healing process to start even before the patient arrives. Here in the Houston area, law enforcement sometimes picks people up from the street while those people are highly agitated. We’re fortunate to have highly trained law enforcement—all officers in the Houston Police Department are required to take crisis intervention training (CIT), which informs them about mental illness, de-escalation techniques, and the most appropriate resources for psychiatric care. Thus, when law enforcement officers bring patients to the ED or psychiatry emergency service (PES) department, the patients are often calm because they’ve already established rapport with the officers. Many times, individuals who had been aggressive and threatening when the officers picked them up are no longer in handcuffs by the time they arrive in the ED/PES. Harris County, where Houston is located, has about 60 law enforcement entities. Not all of them have taken CIT, and you can tell the difference. Many times, patients have arrived in restraints that could have been avoided if the law enforcement officers had gone through CIT.

CHPR: And once the patients have arrived, maybe we can assign staff with better people skills to work with the more agitated patients.

Dr. Knox: Absolutely. We try to have the staff person who has the most rapport with the patient talk to them. I tell people, too, that not everybody is cut out to work in mental health, especially crisis settings. If you’re uncomfortable with people who talk, look, and behave strangely and who don’t have the best hygiene, then you probably shouldn’t work here; it’s not everybody’s cup of tea. But you also want to watch out for burnout among staff if only the same staff members are tasked with dealing with agitated patients. Ideally, all staff should be trained and able to work with challenging patients.

CHPR: Do you worry that by reducing the use of restraints, staff or other patients might face greater risks for injury?

Dr. Knox: When S&R reduction efforts are first initiated, there may be an uptick in staff injuries as staff may be reluctant to use these interventions even when they are indicated. But overall, programs that have successfully reduced their rates of S&R have not reported increased risks to staff or other patients. In fact, they have reported lower rates of injuries (Busch AB and Shore MF, Harv Rev Psychiatry 2000;8(5):261–270).

CHPR: We’ve talked mostly about the use of restraints. Could we use seclusion more often, in place of restraints?

Dr. Knox: Absolutely. Part of the de-escalation is providing choices and asking the patient what they prefer. Seclusion is often a sufficient intervention for patients—even open-door seclusion has a place. And similarly, you can often give patients a choice of oral medication as opposed to IM.

CHPR: What tips can you give us about medications?

Dr. Knox: Unmedicated and actively psychotic patients pose a greater risk for assault, so we can reduce the rate of S&R by ensuring patients receive proper medications (Editor’s note: For more on this topic, see “Medications to Rapidly Treat Psychotic Agitation” in CHPR Oct/Nov/Dec 2021). Be sure to adjust medications based on patients’ responses and modify doses if they are receiving frequent PRN medications. Also, it’s helpful to submit court petitions for involuntary medications sooner rather than later for agitated patients who need but refuse medications.

CHPR: Is there anything else we should know?

Dr. Knox: We sometimes see incidents increase when there’s a shift change. When staff are coming in and other staff are leaving, they can become distracted and less vigilant, so we have to make sure that someone is paying attention to the patients. Dimming the lights and encouraging quiet time on the unit during these transitions can help.

CHPR: I’m glad you brought that up; it’s definitely the case at my hospital. The other seemingly high-risk time is on weekends when there are fewer activities to keep patients occupied and engaged, so it’s important to make sure patients don’t start feeling bored and restless whenever possible.

Dr. Knox: Right. And again, I want to emphasize that the institution must start from the top down. To be effective in reducing S&R incidents, you need to have strong nursing and medical leadership, and you must have quality improvement processes in place to measure success. Once it’s clear that leadership prioritizes the reduction of these incidents, they will diminish.

CHPR: Thank you for your time, Dr. Knox.