Social validation data on three methods of physical restraint:

Views of consumers, staff and students

 

Joanna Cunningham,

University of Portsmouth,

 

Andy McDonnell,

University of Birmingham

 

Simon Easton,

University of Portsmouth.

 

Peter Sturmey,

Department of Psychology and The Graduate Center

City University of New York

 

 

Correspondence: Peter Sturmey Ph.D.,

 

 

++Department of Psychology, City University of New York, 65-30 Kissena Boulevard, Flushing, NY 11368.

Acknowledgements: We would like to thank Endre Kadar, Paul Waby, Mark Turner, Studio Three Training the carers and service-users of Sherbourne House, Basingstoke; West Glamorgan Housing Consortium; and the Community Challenging Behaviour service, Birmingham who participated in this study.

 

Abstract

Twenty-four undergraduate students, 21 residential care staff and 18 service-users from community settings rated videotapes of three physical restraint procedures. Two of the methods involved restraining an individual on the floor and a third method involved restraining an individual in a chair. Participants answered two open-ended questions to rate the methods of restraint and rated the methods on a five-point scale of satisfaction (Flynn, 1986). Participants also rated the three restraint methods by a forced-choice comparison. Restraint was rated negatively by all participants. However, both the satisfaction ratings and the forced choice methods rated the chair method of restraint as most acceptable all three groups of participants. Consumers rated restraint more negatively than other groups.

Social validation data on three methods of physical restraint:

Views of consumers, staff and students

 

Harris (1996) conducted a comprehensive reviewed research on restraint in people with mental retardation. He concluded that restraint could be an effective method to reduce self-injury and aggression. Despite the possibility that such methods of intervention might be effective there remain many areas of concern about the use of restraint with persons with mental retardation. There are significant risks of consumer and staff safety (Hill & Spreat, 1987; Spreat & Balen-Potts, 1983), and risk of death and near-death experiences (General Accounting Office [GAO], 1999; White, 1998). Significant concerns have also been expressed over use of restraint for target behaviors that are not dangerous (Harris, 1996). Sturmey (1999) found that consumer challenging behavior was only weakly related to the use of restraint. This finding suggests that other factors such as staff training, supervision and oversight, and service policy might be more important in determining the use of restraint than consumer challenging behavior (Cf. GAO, 1999). Further concern has also been expressed over excessive use of restraint, inconsistent regulation, lack of oversight (GAO, 1999) and lack of local policy to regulate its use (Harris et al.1996). It is also possible that restraint may inadvertently reinforce and maintain maladaptive behaviors (Van Houten, 1999).

As institutions continue to close and downsize restraint may now take place in community settings such as special education classrooms, group homes, work and family settings (McDonnell & Sturmey, 1993). McDonnell, Dearden and Richens (1991) reported on the development of a course to prevent and reduce aggressive and violent behaviors in community settings. One part of that training course is the use of restraint with two staff and a chair. Previous research had reported that direct care staff and undergraduate students rate restraint in a chair is more acceptable than two methods of restraint on the floor (McDonnell et al, 1991; McDonnell & Sturmey, 1999).  Previous research had not reported the evaluations of consumers themselves. In this study we report the evaluations of three kinds of physical restraint rated by students, direct care staff and persons with borderline, mild and moderate mental retardation.

 

Method

Subjects

Three groups of participants took part in the study. Group 1 consisted of 10 men and 14 women undergraduate students. Their mean age was 20 years (SD = 2 years). Group 2 consisted of 13 men and 8 women who were residential care staff. They worked with people with mental retardation and challenging behavior in five different community residential settings. There were both qualified and unqualified staff in this group. All had used or witnessed the use of physical restraint. Their mean age was 39 years (SD = 10 years). Group 3 consisted of 18 men who were service-users living in four community settings. Seventeen had mild or moderate mental retardation and one had borderline mental retardation. Their mean age was 37 years (SD = 9 year). All had challenging behaviour and the majority also had additional psychiatric diagnoses. All of the service-users had experienced or personally witnessed physical restraint.

 

Materials

            The three methods of physical restraint selected were those used previously in McDonnell et al (1993) and McDonnell & Sturmey (1999). Method A was described by Harvey & Schepers’ (1977) as follows: “ … Two staff members, operating as a team, approach the resident from the sides, grasp the wrists on each hand at the joint and follow immediately by grasping the upper arm above the elbow. They push down on the wrist and pull up on the upper arm and place each of the resident's arms behind his back. Staff directions are maintain this hold throughout the entire procedure, do not switch hands or you will lose control of the resident. After the arms are secure behind the resident each staff member places one foot behind the resident's feet and pulls the resident off balance and to the floor. As soon as the resident is close to the floor one staff member steps over the resident and the other staff member comes behind the resident thus rolling the resident on his front maintaining the hold on the arms of the resident throughout these steps. The final step in securing the resident is for one member of staff to sit on the residents upper legs, just above the knees to prevent kicking.”. Method B was described by Lefensky, DePalma and Lociercero (1978) as follows: “ … Almost immediately after the second staff member touches the patient's wrist the first staff member takes the other arm... If the patient needs further restraint it is possible to lay the patient on the floor by placing a foot behind and walking or tripping the patient over … “  They go on to describe one member of staff sitting on the person's legs and the other member of staff sitting astride of the person holding both the service-user's arms. Method C was McDonnell, Dearden & Richens’ (1991) chair restraint. They describe the method as follows: “ … Firstly you should (if possible) place your hands in front of you at right angles to your body. This protects your hair and your eyes, as the client moves towards you twist your hands, taking care to keep your hands raised”. The restraint procedure, described in detail as follows, should only be used if the above fails. “  … You should then grab the clients forearm, and place your own forearm in a slightly upward movement into the clients armpit. Both staff and client are now facing in the same direction. At this point you should request help from a second member of staff, who should approach from the clients rear. Placing themselves in the same position on the opposite side of the client … For every step forward you make with the client you should take two steps backward. This procedure is carried out at a slow pace until the client can be sat in a chair. This can be achieved by backing the client into the chair, which gently buckles their knees  ...” The authors go on to describe how to restrain the service-user in a chair using nothing more than human body weight. “ … This can be achieved by bending your knees. You and your colleagues place the client's forearms on the arms of the chair, you then place your forearm in the crook of the client's elbow joint. Most crucially you must use your body weight when restraining the client and not brute arm strength (keeping your head up) and placing your forearm at a ninety degree angle into the bent arm of the client. You then lean forward onto the client's arm joint pushing your body weight into the back of the chair…”

The three procedures were presented by videotape and by freeze-frame presentation of the information. Three short videotaped sequences each contained a male aggressor who was restrained by two male care staff. The service-user resisted in all three sequences. The volume was muted during presentation of all three videotapes. Freeze-frames were made from the videotaped sequences, one for each of the three restraint methods. The freeze-frames were taken form the end of each videotaped sequence, enlarged and enhanced to make them clearer. The order of presentation of videotapes was randomized.

 

Procedure

The project was conducted in accordance with the British Psychological Society's ethical principles for conducting research with human participants and approved by the University of Portsmouth Ethics Committee. Agreement to approach both carers and service-users was first obtained from service-leaders. If service-leader had a concern then the participant was excluded. Interviews with Group 1 were conducted in a university psychology department. Interviews with Group 2 were conducted at their place of work. Interviews with Group 3 were conducted in their homes or day center. Interviews with Groups 1 and 2 were conducted individually. The first author, a female, 21-year-old undergraduate Psychology student not previously known to any of the participants, completed all interviews. A second rater independently rated audiotaped interview material. Interviews with Group 3 were conducted with a care staff present who did not participate in the interview process. For groups 1 and 2 the following information was read out to participants from Groups 1 and 2: “'Physical restraint procedures are used in the caring services for adults and children with learning disabilities [mental retardation] in order to prevent self-injury or injury to others. You are going to see three short video extracts of commonly used physical restraint procedures. I will then ask you some questions about the videos.” If participants were still happy to take part, written consent was obtained. An adapted form using simplified language was used for Group 3. The participants were free to withdraw at any time. If the care staff indicated that the consumer should withdraw they could also withdraw at that time also. If service-user and care staff both indicated that they were happy for the service user to take part in the written agreement was obtained from their representatives on their behalf.

Participants were shown the videotaped sequences of each of the three restraint methods in turn. After each sequence, the tape was stopped and the participants were presented with a freeze-frame of the procedure they had just seen. Participants were then asked, “How would you feel if you saw this happening?” and “How would you feel if this happened to you?” Questions were asked clearly and slowly. Participants were given as much time as they wished to respond. In the case of Group 3, up to a minute was allowed for a response. Responses to each question about each method of restraint were rated on a five- point satisfaction scale where ‘1’ indicated high satisfaction and 5 indicated dissatisfaction. (Flynn, 1986). Service-user responses were also rated on the Responsiveness and Communication Style Scale (Sigelman et al., 1991). Responses to the open-ended questions of five participants in each group were audio-recorded and rated independently by Rater 2. The ratings of satisfaction were highly reliable for all three groups (Group 1 -- r = 0.925, p <0.01; Group 2  -- r = 0.963, p < 0.01; and,  Group 3  -- r = 0.879, p < 0.01).

In the forced choice methodology participants were then shown freeze-frames of two procedures. They were asked to point to the picture they thought was the worst out of the two. This was repeated until combinations of procedures had been presented. The procedures were then ranked 1 through 3 from least to most preferred method. A final comparison between all three procedures was used to check for consistency. Participants were then debriefed.

The quality of the responses was evaluated using the criteria of Sigelman et al (1991). The codings were: (a) no response; (b) unintelligible response; (c) irrelevant response; (d)  'don't know', 'don't remember', 'not sure'; (e) inadequate response (vague or uncodeable); (f) request for clarification of question; (g) refusal to answer; (h) minimally appropriate response (response which meets the formal demands of the question); and, (h) expanded response (response which provides qualifying or additional information beyond the formal demands of the question). Ratings are made of the response to the initial question, not to the sum of responses concerning a topic. In total, 82.4% of responses to the open-ended questions were rated as adequate responses, 57.4% of these were minimally appropriate responses, and 25% were expanded responses. These results suggest that the majority of service-users interviewed were able to respond adequately to open-ended questions. Seventeen of the 18 service-users were able to provide an adequate response to at least one of the two open-ended questions for each method of restraint. One service-user did not provide adequate answers to either of the open-ended questions in two of the three conditions. Data from this participant was excluded.

 

Results

Open-ended questions

Satisfaction ratings  A two way ANOVA was conducted using groups and videotapes as the two factors. Post hoc tests were performed using the Least Difference Tests using an alpha of .01. Mauchly’s test of sphericity was significant (p = 0.022) indicating a significant difference in the variances of the three within-subjects conditions. Therefore, Greenhouse-Geisser’s more conservative F value was used, which corrects for problems with the sphericity assumption by setting the degrees of freedom for the interaction at their minimum possible. A two-way ANOVA yielded a highly significant two-way interaction between restraint method and group (F (4, 105) = 5.550, p <  0.001). A significant main effect for method of restraint (F (2,  105) = 153.86,  p << 0.001). Pair-wise comparisons revealed that Method C (Chair) was preferred over both other methods (p’s  < 001). Method B (Harvey & Shepers, 1977) was preferred over Method A (Lefensky et al, 1978) (p < .001). A highly significant main effect for group (F (2, 59) = 8.305, p < 0.001) was found. Pair-wise comparisons revealed that ratings from the service users were more negative than the other two groups (both p’s < .005). The interaction effect was accounted for by the fact that service user’s rated the Chair method more negatively than the other groups (both p’s < .001).

_____________________

Insert Figure 1 about here

_____________________

 

Forced choice comparison The Friedman ANOVAs yielded a very highly significant difference in mean ranks between restraint methods for all three groups (all p’s < 0.001). Multiple comparisons revealed that Method C was rated as significantly more acceptable than Method A in all three groups. In groups 1 and 2 Method C was rated as more acceptable than Method B (all p’s <. 05). All three groups viewed Method A as significantly worse than method C. Only groups 1 and 2 viewed method B as significantly worse than method C (all p’s < .01). There were no significant differences in the ranks between the three groups for any method. This indicates that there were no differences between the three groups on the rankings of the restraint methods.

 

_____________________

Insert Figure 2 about here

_____________________

 

Discussion

The chair method of restraint was rated as more acceptable than both methods of restraint on the floor. This was true for all three groups of raters. A different methodology was employed in the present study, however the results obtained still provide an indication of the acceptability of the three restraint methods to participants. The results of this study replicate and extend the findings of our two earlier studies (McDonnell et al., 1993; McDonnell & Sturmey, 1999). These findings are important as they demonstrate that consumers with borderline to moderate mental retardation can participate in the evaluation of alternate forms of restraint. The results also show a significant difference between the satisfaction ratings of the two methods of restraint on the floor. One of the floor methods (Lefensky et al., 1978) was viewed less negatively than the other floor method. McDonnell & Sturmey (1999) revealed similar findings. The authors found that the same floor method (Lefensky et al., 1978) was rated as more acceptable than the other floor method by the high school students. In the present study however, the finding occurred across all three groups.  

It would be incorrect to interpret the findings of the present study as an endorsement of the use of physical restraint, including the chair method of restraint. The satisfaction ratings for all three restraint methods in all three groups were negative and the service users found the chair method of restraint more negative than students and staff.  This finding supports other social validation studies which report that physical restraint is consistently viewed as restrictive (Morgan, 1989). The results of this study clearly indicate that while physical restraint was viewed negatively across groups, the degree to which it was viewed negatively was dependent on the type of restraint method used. The chair method was viewed by all groups of raters, including service-users themselves, as the most acceptable of the restraint procedures.

In the two previous acceptability studies (McDonnell et al., 1993; McDonnell & Sturmey, 1999) the restraint procedures were presented to participants as videotaped sequences rather than written descriptions. As McDonnell & Sturmey (1999) point out, this material is still an analogue method of presenting information. However, we argue that it is closer to reality than written description. In the present study, the three restraint procedures were presented in two iconic formats. Iconic presentation is particularly useful as a research tool when participants have learning disabilities because it does not require reading and memory skills needed for written or verbal descriptions of restraint procedures. This study extended earlier research by demonstrating that consumers can rate the acceptability of restraint methods using both open-ended questions as well as forced-choice methodology.

 The current concerns over restraint relate to consumer safety rather than treatment acceptability (GAO, 1999; White, 1998). The safety of the chair method of restraint has not yet been evaluated empirically. Future research should expand on these preliminary findings to evaluate the safety of the chair method of restraint both for staff and consumers.

References

Flynn, M.C. (1986). Adults who are mentally handicapped as consumers: Issues and guidelines. Journal of Mental Deficiency Research, 30, 369 – 377.

General Accounting Office (1999). Improper restraint or seclusion use places people at risk. Washington, DC: General Accounting Offices.

Harris, J. (1996). Physical restraint procedures for managing challenging behaviours. Presented by mentally retarded adults and children. Research in Developmental Disabilities, 17, 99-134.

Harvey, E. R. & Schepers, J. (1977).  Physical control techniques and defensive holds for use with aggressive retarded.  Mental Retardation, 13, 29-31. Hill, J.,

Spreat, S. (1987).  Staff injury rates associated with the implementation of contingent restraint.  Mental Retardation, 25, 141-145.

Lefensky, B., De  Palma, B. T. & Lociercero, D. (1978). Management of violent behaviors. Perspectives in Psychiatric Care, 16, 212 – 217. 

McDonnell, A. A., Dearden, R.L. and Richens, A.  (1991).  Staff training in the management of violence and aggression. 1 - Setting up a training system. Mental Handicap , 19, 73-76.

McDonnell, A. A. & Sturmey, P. (1993).  Managing Violent and Aggressive Behaviour: Towards Better Practice. (pp.  148 - 171). In R.S.P. Jones & C.B. Eayrs (Eds. ). Challenging Behaviour and Intellectual Disability: A Psychological Perspective.  Avon: BILD.

McDonnell, A & Sturmey, P. (1999). The social validation of physical restraint procedures with people with developmental disabilities: A comparison of young people and professional groups. Research in Developmental Disabilities, 21, 84 – 92.

McDonnell, A. A. & Sturmey, P. & Dearden, B. (1993).  The acceptability of physical restraint procedures for people with a learning difficulty. Behavioural and Cognitive Psychotherapy, 21, 255-264.

Morgan, R. (1989).  Judgements of restrictiveness, social acceptability and usage: Review of research on procedures to decrease behavior.  American Journal of Mental Retardation, 94, 121-133.

Sigelman, C. K., Budd, E. C., Spahel, C. L . & Schoenrock, C. J. (1981). Asking questions of retarded persons: a comparison of yes-no and either-or formats. Applied Research in  Mental  Retardation, 2, 347-57.

Spreat, S. & Baker-Potts, J. C. (1983).  Patterns of injury in institutionalized mentally retarded residents.  Mental Retardation, 21, 23-29.

Sturmey, P. (1999). Correlates of restraint use in an institutional population. Research in Developmental Disabilities, 20, 339 – 346.

Van Houten. R. (1999). The role of physical restraint in treatment. Paper presented to the American Association on Mental Retardation, Chicago, October, 1999.

White (1998). Deadly restraints. Hartford Courant, October 11 – 15, 1998.

Figure captions

Figure 1. Mean satisfaction ratings by students, carers and consumers of methods A, method B and the chair restraint.

Figure 2. Mean rankings by students, carers and consumers of methods A, method B and the chair restraint.

 

Figure 1.

 

 

 

 

Figure 2.