Monday, February 20, 2012

Use of Restraint: Thoughts from a Behavior Analyst

In the Arms of the Behavior Analyst:
A brief review of behavior analytic methodology and research on restraint,
and the humble opinion of one such behavior analyst

by Lorien Quirk, M.Ed., BCBA

Let me start with my opinion on the topic of restraint. Having been in an educational system for 6 years (as both a teacher and an administrator) in which I was required to complete a certification course to be employed in my position, and currently serving as one of two trainers on crisis management and assaultive behavior for a district of 35,000 children, I feel I have a good grasp on the importance of this topic. I used to think restraint was absolutely necessary. In a previous position, I assisted with restraint every day. There are places where restraint is run-of-the-mill and standard procedure for treatment. It wasn’t until I became a Board Certified Behavior Analyst at the same time as becoming a Trainer for Professional Assault Crisis Training (Pro-ACT) that some serious questioning began.

One of the main questions I am faced with is, “What is the difference between prompting and restraint?” My best answer to this (which could be argued by staunch Pro-ACT advocates) is that a prompt enables the person to achieve the desired response in order to access reinforcement. A prompt facilitates learning. Sometimes, a prompt requires physically moving the person’s body and a strict interpretation of Pro-ACT might indicate that this is violating the person’s civil rights. In my experience, it is more of a violation to deny the person access to learning and reinforcement because of a fear of “restraint.” Restraint is about inhibiting a person’s movement to decrease their ability to commit aggravated assault, not about moving his/her body in a way that is instructional.

Restraint should be applied in circumstances of aggravated assault, in which the threatened or occurring behavior would result in serious injury meriting a hospital visit. The problem we have encountered as of late, particularly with the legislative contingent, is that restraint is being used too frequently and inappropriately with students with disabilities. Given my educated description of prompting and the potential restrictions that may be imposed upon school personnel regarding physical management of students, we will continue to seriously question the alternatives. Weiner and Wettstein (1993) describe the conflict: “For those who view these [restraint] practices as worthy, the restrictions in the form of procedural requirements represent troublesome interference with treatment. In contrast, for those who view restraints and seclusion as punitive, anything short of their prohibition is inadequate.”

There is of course the issue of restraint actually being misused in a way that causes trauma, or even death, in students with disabilities. There are between 50-150 documented occurrences of death during restraint each year (“Professional Crisis Management” website). These tragedies occur for a variety of reasons, including inappropriate decisions to use restraint and inappropriate or untrained use of restraint. In becoming a Pro-ACT Trainer, I have taken on the challenge and responsibility of ensuring the people in my school system are trained and applying the principles correctly. In cases in which restraint was improperly used, it is likely that lack of training or supervision contributed to the problem, not the urgency of the need to restrain the student.

The other issue is the damage that can be caused by a student engaging in assaultive behavior. At the moment a person begins aggravated assault, they have automatically given up the privilege of unconditional access to society. Look at the laws by which all citizens are expected to abide. If someone in public begins assaulting another person, they are arrested, hand-cuffed (a form of restraint), and possibly convicted and put in jail (another form of restraint/seclusion). In my opinion, people with disabilities should not be exempt from the societal consequences of such behavior, HOWEVER, we as a society have been charged with seeking alternatives to these restrictive procedures to hold in highest regard the best interest of these individuals and their exceptional needs. As with any procedure used in schools, we must constantly question and research the effectiveness of the procedures we are using.

There are several main issues with restraint, from a behavioral perspective. There is the possibility that by using manual restraint that the restrainers are inadvertently reinforcing the problem behavior with the physical sensation of restraint, or by the close and intense attention that inevitably occurs during restraint. On the other hand, it is also possible that a restraint would punish the problem behavior, which, according to behavior analysis, can have unwanted side-effects such as increased aggression or withdrawal. In addition to this type of punishment, students who have previously experienced physical trauma would most likely find being restrained to be extremely punishing to the point of causing psychological damage. From the behavioral perspective, reinforcing problem behavior is not desired, but neither is punishment when it has the potential side-effects that can be associated with restraint.

Another fundamental behavioral issue with the use of restraint as treatment and/or crisis management is that it does not teach any new skills. As a behavior analyst, we are interested in determining the function of a behavior and teaching a new skill that may be more efficient or socially appropriate than the unwanted behavior. When a person is in restraint, they are not learning anything but how to behave to get restrained. This defeats a primary function of behavior analysis.

One research article in the Journal of Applied Behavior Analysis that touches on the subject of restraint in a classroom setting described the use of a basket-hold restraint to treat a student in a classroom who engaged in aggressive behavior. The results indicated that the use of restraint may have “either maintained or evoked” the target behavior in the student. The point of the article is that using any type of physical restraint without a full functional analysis is unwise given that it may inadvertently reinforce or punish the problem behavior (Magee & Ellis, 2001).

A similar result was found when Favell, et al. (1978) used physical restraint to treat severe self-injury (SIB) by performing a functional analysis that showed that physical restraint actually functioned as a reinforcer to the client. Though the behavior-change techniques they used in this article would most likely not be as acceptable today, they did use physical restraint as a reinforcer in a DRO (differential reinforcement of other) procedure in which the restraint was given contingent upon time without SIB. This experiment showed that physical restraint can function as a reinforcer.

Wallace, et al. (1999) explored the use of restraint with SIB and found that the flaw with their strategy of using physical restraint in their treatment was that it neglected to teach the student new adaptive skills related to the function of her problem behavior. Not only does restraint not teach a new skill, but time in restraint is time that could be spent teaching new skills instead.

Grace, et al. (1994) explored a case in which differential use of restraint was implemented contingent upon differential levels of the problem behavior in a school setting. In the study, the use of restraint was viewed as a punishment procedure to the student, and given the school’s restrictions on the use of punishment, the authors tested whether the differential use of this type of punishment would have any effect on the overall rate of problem behavior. The authors found inconsistent results which would indicate that just as with any other behavioral intervention, it must be applied consistently to have the desired effect.
As the (limited) research shows us, the use of restraint in behavior analytic practice varies greatly. There is not a reliable or predictable way to “prescribe” restraint as a treatment unless full functional analysis is conducted, which is seldom done in public school settings. According to Pro-ACT, restraint should not be used as treatment at all, but only ever in a situation of aggravated assault. The question remains, with all we know about the bad things about restraint, and all we know about the bad things that can happen if we don’t restrain, how do we proceed as a field? There is not a clear answer here, but as long as people working in the field continue to access training, supervision, and research on the topic, at least we should be able to trust our humble opinions.


Favell, J. E., McGimsey, J. F., & Jones, M. I. (1978). The use of physical restraint in the treatment of self-injury and as positive reinforcement. Journal of Applied Behavior Analysis, 11, 225-241.

Grace, N. C., Kahng, S., & Fisher, W. W. (1994). Balancing social acceptability with treatment effectiveness of an intrusive procedure: A case report. Journal of Applied Behavior Analysis, 27, 171-172.

Wallace, M. D., Iwata, B. A., Zhou, L., & Goff, G. A. (1999). Rapid assessment of the effects of restraint on self-injury and adaptive behavior. Journal of Applied Behavior Analysis, 32, 525-528.

Parents and legal guardians can learn more, or contact us, by going to

Weiner, B. A., & Wettstein, R. M. (1993). Legal issues in mental health care. New York, NY: Plenum Press.

1 comment:

  1. Thanks for this great post!- This provides good insight. You might also be interested to know more about generating more leads and getting the right intelligence to engage prospects. E-Health Care Lists E-Health Care Lists implements new lead gen ideas and strategies for generating more leads and targeting the right leads and accounts.E-Health Care Lists is one of the global suppliers of healthcare mailing list & email list.Marketing to the healthcare industry, reaching the doctors and other healthcare decision makers is often an impossibletask.Behavior Analyst Mailing list & Email list