Restraints: All Holds Barred?


Dear Director: My hospital has just changed its restraints policy, wanting us to minimize any restraints and also prohibiting chemical restraints. What’s going on? Some patients just need restraints.

For those of us old enough to remember the TV show M*A*S*H, they had a magical IM injection that could make a crazy patient, or Major Burns, collapse to the ground unconscious in a matter of seconds. Depending on where you work, you may wish you had a drug like this at your disposal on a daily basis.

Definitions of Restraints
Let’s just start with the basics. There are three types of restraints—physical, mechanical, and chemical. Physical restraints are a manual hold that restricts the patient’s freedom of movement—such as holding someone down for an injection. Mechanical restraints are when a device or material is attached to a person’s body that restricts freedom of movement or restricts normal access to his/her body. These are made of tough, neoprene fabric that have a Velcro lock so that patients cannot remove them. Your hospital probably got rid of the leather restraints and vests years ago because these were considered the most dangerous and were the ones associated with more adverse events. Chemical restraints are a bit more complicated. These are medications that are not part of a patient’s standard or PRN meds (remember these rules apply to all healthcare facilities such as acute care hospitals, psychiatric hospitals, and nursing homes), which are used to manage behavior or restrict a patient’s freedom of movement. This could be the magic injection in M*A*S*H, often the way we used droperidol to sedate an obnoxious drunk patient 20 years ago or giving the Haldol/Ativan cocktail to a combative patient that the police bring in. Police often bring handcuffed patients into the ED. These are not restraints from our point of view (and associated with the necessary documentation) but handcuffed patients still require monitoring like other patients within our care.


Restraint Indications
There are clear indications for restraints and certainly in many EDs, restraints are needed on a regular basis for patient and staff safety. Federal regulations states that seclusion and physical restraint may only be used to ensure physical safety of the patient, other patients, or staff members [1]. However restraints may not be used as punishment or for the convenience of staff members. These are the kinds of things that got healthcare in trouble in the first place — ask yourself how many F-yous it took from a drunk patient in the past before you sedated them, restrained them, or intubated them? Laws also vary from state to state, but restraint use generally requires a patient who has failed less restrictive interventions and whose behavior presents an immediate danger of harm to self or others.

So Why are Hospitals Getting Away From Restraints
The main reason that groups are pushing for a decrease in restraints is because restraints are considered “violations against human rights.” While every regulatory organization completely understands that restraints are sometimes necessary (e.g. violent or combative psychiatric patients or patients impaired by alcohol or other substances), they are highly regulated because many organizations used them for convenience (think intubated patients, pulling at lines, low staffing). A second reason is safety. Many patients have died in restraints. Confused elderly patients sometimes fall between the bed and rails and have been found strangled by restraints. This is not a new issue. The discussion between patient rights, restraints, and injury patterns goes back to the 1990s. An early study showed that a substantial decrease in restraint use in a nursing home population occurred without an increase in serious injuries when a comprehensive patient assessment was done and restraint alternatives were used [2]. 

Another reason that hospitals are getting away from restraints is that CMS makes the documentation and reporting so difficult that most organizations have tried to adopt a “no restraint” usage policy. Another chair said to me once that it seems that the documentation necessary to restrain someone is an example of “government regulations run amok.” (true quote) That may be your opinion, too. My personal philosophy has always been to understand the rules of the game and then learn how to play by them, whether I agree with them or not. The CMS requirements go back to at least 2008 (that’s the date on the copy I’m reading) [1]. I can remember completing restraint documentation 15 years ago where we could check boxes for considering soft lighting and music, massage, and giving the patient distracting tasks to perform. But for the two nursing home dementia patients I remember who collated blank ED charts as a way to modify their environment and distract them because they were screaming and agitated when they were just laying in bed, we had numerous patients who were combative, dangerous, and/or intoxicated and required physical and/or chemical restraints.


Chemical Restraints
The use of chemical restraints adds an extra layer of complexity for hospitals. Although the term and practice is considered standard among emergency physicians and is still found in references and taught in residency, there appears to be a disconnect to what many community hospitals are doing—establishing policies that prohibit chemical restraints. Medically speaking, the literature suggests that chemical restraints may be safer for the patient than mechanical restraints in certain circumstances, such as preventing hyperthermia or rhabdomyolsis from a patient that continues to fight and struggle in mechanical restraints. Of course, there is also a risk of injury due to the chemical restraints (e.g. aspiration, apnea, arrhythmia). It’s always going to be a judgment call as to when it’s best to use chemical restraints. And there is the real issue of staff safety as well. But the argument against chemical restraints is that just like physical restraints, there is possibly a human rights violation because of the restriction of behavior and movement. When we chemically restrain patients, they are unable to freely move, etc. While this may be necessary in some time limited situations because of a patient’s behavior, it can cause a cycle of not getting to the root cause of the patient’s actual medical issues. When chemical restraints are used, it also may become difficult to understand the patient’s underlying medical issues. The observation and documentation that is typically performed by a tech or a sitter for restrained patients becomes that much more complex as well. There is now a sedated patient, perhaps completely unconscious, and it can be challenging for a sitter to make a medical observation as to their airway or respiratory status. And while physical restraints can be removed when the patient is calm and not a threat to self or others, it can be challenging to know when the chemically restrained patient is truly awake, alert, and competent for decision making or discharge. While we shouldn’t be handcuffed by regulations, I do think that in view of today’s regulatory climate treating physicians should be prepared to answer questions regarding a restraint during a CMS audit. Picture how you would answer the question about knowing as to how you knew whether the recent patient who was chemically restrained due to has violent behavior from PCP intoxication is now medically okay to sign out AMA?

Best Practices
There are clear best practices when it comes to restraint usage. Above all else, we need to take care of the patient and protect the staff. But in order to avoid any issues with CMS, we need to document, document, and document. The documentation may be considered onerous. But first, I would advise you to review your hospital’s restraint policy. The newest and perhaps biggest issue to sort out is whether your hospital allows use of chemical restraints and if so, what special documentation requirements are needed for these patients. Don’t fret if your hospital restraint policy does not allow chemical restraints. The good news is that we can treat symptoms without the use of chemical restraints. Developing individual patient plans of care has become a key initiative by regulatory agencies. Medicating patients to control symptoms so that they can be participants in their plan of care is an important aspect. For instance, we frequently give meds like Ativan and Haldol (often together for chemical restraints) for agitation, but using=each medication individually (even in high doses) may be necessary to adequately treat the patient’s symptoms so they can better participate in their care plan and this would note qualify as a chemical restraint. Documentation does matter so when we use these meds, we should be including a medical decision note indicating that we are giving them for agitation, psychosis, or control of other symptoms. If you document that you’re giving meds for chemical restraints (and keep in mind that the FDA has not approved any medication for chemical restraints) you and the ED will be held to the documentation standards for restraints. However, by preferably using only one medication, we can truthfully document that we’re treating the patient’s symptoms. Some severely agitated or psychotic patients may be given high doses of meds but again, we need to document that such a dose is required for symptom control. The intent of the medication is to treat the symptoms as part of the plan of care and thus to allow symptom control of the agitated, combative, or psychotic patients so they can be more participatory in their care. While sedation may be a side affect of the meds, it shouldn’t be the intent of the medication.

The documentation necessary for restraints is very amenable to the development of EMR phrases or scripts. Documentation should include what the thought process was that led to the necessity of restraints. Also document whether less restrictive methods were attempted. Use of restraints requires ongoing observation, assessment, and documentation that are typically done by the nursing staff. Keep in mind that there are regulations about the length of time that patients can be in restraints before the order expires — 1 hour for patients under 9 years; 2 hours for patients from 9 to 17 years, and 4 hours for patients 18 years of age and older.

More and more hospitals are changing their restraint policies or trying to go restraint free. The patient population in the ED often times has patients who will continue to require some sort of restraint to protect the patient and/or the staff. The goals of the provider are to diagnose and treat the patient’s underlying medical problem which may require effectively using medications to help control the patient’s symptoms so that they can be more participatory in their plan of care. Accurate and thorough documentation on these patients is critical.



  1. State Operations Manual Appendix A – Survey Protocol, Regulations and Interpretive Guidelines for Hospitals.
  2. Neufeld RR, Libow LS, Foley WJ, Dunbar JM, Cohen C, Breuer B. Restraint reduction reduces serious injuries among nursing home residents. J Am Geriatr Soc 1999; 47:1202-1207.


EXECUTIVE EDITOR Dr. Silverman is Chairman of Emergency Medicine at the Virginia Hospital Center. He also serves as the Director of the Alteon-Mid Atlantic Leadership Academy. Dr. Silverman’s practical wisdom is available in an easy-to-use reference guide, available on Amazon. Follow on Twitter @drmikesilverman


  1. David C Crutchfield on

    So we’re not restraining, we’re “treating symptoms”, got it! Yet another interference in the practice of medicine on the part of government bureaucracy. Sorry, I will still be “treating symptoms” all night long. What is CMS going to mandate when they can’t find enough physicians willing to put up with “symptoms” all night long, every night, with the distractions taking away from the rest of the busy, often seriously ill ED patient load. They wonder why physician burn out, particularly in EM is so high. Great move CMS, JCAHO and the rest from your ivory tower. The next combative drunk that comes in, I’m leaving int the administrator’s office for morning. /rantoff

Leave A Reply