Rethinking Hospital Restraints

It was 2 a.m., but the intensive care unit (ICU) where I was rotating as a medical student was wide awake. In one room, a man with liver failure was moaning as his medical team attempted to insert a central line into his jugular vein. In the next room, a woman who had been taken off her ventilator earlier that day had awoken and was screaming “Get me out of here!” At the end of the unit, a man recovering from pneumonia was trying to get out of bed despite being hooked to several intravenous lines.

Three elderly patients. Three different conditions. But on that busy night in the ICU, the medical team responded the same way to all three—with physical restraints.

Most of us who have been hospitalized have never seen physical restraints, as they are rarely used outside the ICU. Examples include wrist and ankle belts, vests, mitts, and full-length side rails attached to the bed. According to Medicare guidelines, restraints should only be used to ensure the safety of patients and staff and should be removed as early as possible. There are only a handful of situations where Medicare and other physician groups recommend using restraints, including patient violence towards himself or others and a threat of a patient disrupting his or her life-saving therapy, such as a breathing tube.

As a supposed measure of last resort, however, restraints are surprisingly common. A 2007 study in the Journal of Nursing Scholarship estimated that 27,000 patients are restrained every day in U.S. hospitals—an average of five patients per hospital per day.

In some situations, restraints may be ineffective and even harmful. Doctors and nurses often employ restraints when a patient is at risk for falling or delirious. However, evidence suggests that restraints do not reduce one’s risk of falling.  Likewise, a study in the Journal of the American Medical Association suggested that restraints increase the risk of delirium in the hospital by 4-fold, possibly by increasing patients’ levels of anxiety and stress due to involuntary immobilization.  Physical restraints and the resulting immobilization they cause are also associated with increased rates of pressure ulcers, respiratory complications—and even death via strangulation and aspiration. Even more disturbingly, a 2006 report by the U.S. Department of Health and Human Services concluded that hospitals failed to report more than 40 percent of deaths related to restraints to The Centers for Medicare and Medicaid Services (CMS).

Though these statistics may make the ICU seem like an asylum, it is not. The majority of patients I cared for in the ICU never required restraints. And for the others who were confused or agitated, our team could usually avoid restraints by adjusting or prescribing medication.

However, as one of my senior residents confided in me, when caring for the sickest patients in the hospital, restraining a delirious patient might be the only way to devote time to other seriously ill patients. On that busy night during my ICU rotation (which happened before I came to Brigham and Women's Hospital, where I work now), there were other patients suffering from life-threatening infections, failure of multiple organs, or other conditions that put them on the cusp of death. They needed our attention. And yet watching those three patients fighting against their restraints, becoming more delirious with the very devices used to manage their delirium, was one of the most troubling things I witnessed as a medical student.

At its annual meeting this year, the American Geriatrics Society (AGS) released a recommendation that healthcare providers avoid using physical restraints to manage behavioral symptoms of hospitalized delirious adults. Dr. Caroline Vitale, associate professor of medicine at the University of Michigan, presented the recommendation at the AGS Annual Meeting, where she highlighted innovative efforts by some groups to reduce restraint use. Some physicians are using devices, such as shields around IV or central line sites, to protect medical interventions without restricting patient movement. Another hospital is redesigning an entire ward to be “restraint-free” through steps like providing mobility aids in all rooms so that patients can move around safely when they want to.

These interventions can only help. Ultimately, though, reducing the use of restraints doesn’t even require a large investment of time and money. I remember another patient from that same ICU rotation who became acutely delirious and was trying to leave the floor. He could have easily been restrained, just as the three other patients were. However, the medical care team took the time to re-orient him, reminding him why he was in the unit and what day and time it was. The team offered him medications to deal with the agitation and anxiety, some of which he accepted. One of the nurses called the patient’s wife and put her on speakerphone to provide a familiar, soothing voice. Taking the time to explore these other options worked: The patient eventually calmed down without the need for restraints or other aggressive interventions, and the next morning his mental status improved. I hope that stories like his can become the standard, not the exception, in the ICU.