The use of mechanical restraints in the treatment of violent or aggressive patients

Published: 2019/12/04 Number of words: 4512


The widespread use of physical restraints is an issue of increasing concern in today’s nursing practice. Physical restraints can be defined as any physical or mechanical device to involuntarily restrain the movement of the whole or a portion of a patient’s body as a means of controlling their physical activities (Howells and Hollin, 1989). The use of restraints is a contentious issue and there are both positive and negative outcomes. In light of the negative outcomes, a study in 2005 of the Pennsylvania state hospital system reported that all hospitals in the region planned to totally eliminate restraint interventions, positing that ‘these traumatising procedures produce significant barriers to recovery and have no clinical value’ (Smith, Davis and Bixler, 2005, p. 1116). This caused intense disputation across other states, however, with one author arguing that this aim ‘is laudable in idealism but lacking in clinical reality’ (p.576). One of the clinical realities proposed was that it would increase the use of PRO-RE-NATA medications and the trans-institutionalisation of aggressive persons with serious mental illness to prisons, where seclusion and restraint are much more frequent and damaging to patients than in hospitals (Liberman, 2006). Despite these two contrasting viewpoints on the management of violence and aggression, nurses must deliver care based on the best available evidence or practice (Nursing and Midwifery Council, 2008). Therefore, the aim of this paper is to critically analyse the evidence base for the safe and effective use of interventions when managing violence and aggression. Further, in this paper, I will be reflecting upon personal incidents: therefore, I feel it is appropriate that I write in the first person (Webb, 1992).

I have recently been involved in nursing a patient who was a serious suicide risk; she had made a very serious attempt by slashing her throat which needed 64 internal and external stitches. She was being nursed on level four enhanced observations with two members of staff constantly within an arm’s length reach of her. The reason for this was that she was an ongoing risk of suicide as she repeatedly attempted to reopen the wound by tearing it apart with her fingers. This led to her being physically restrained and, due to her continued resistance, usually being put on the floor and rapidly tranquillised. This would happen several times in the day and sometimes she would spend the whole day in restraints. The patient had a history of severe sexual abuse and would often scream that members of staff were trying to rape her while she was being restrained. Given her past abuse, I began to question the validity, appropriateness and efficacy of these interventions; in times of acute distress, the patient was being restrained by perhaps up to five members of staff, usually mostly males. I began to explore alternative treatment options that might be more appropriate, ethical and supportive for the client. One option that was posited by the multidisciplinary team was the use of mechanical restraints. However, the staff members were dubious about this option, as none of them had ever implemented such an intervention and they lacked knowledge of this area of clinical practice. Therefore, this paper will critically analyse the evidence base for the use of mechanical restraints for managing violence and aggression, particularly violence and aggression towards the self.

According to Dale (personal communication, 2009), the starting point when assessing the appropriateness of mechanical restraints is the Code of Practice of the Mental Health Act 1983. This states that ‘mechanical restraints are not a first-line response or standard means of managing disturbed or violent behaviour in acute mental health settings. Its use is exceptional. If any forms of mechanical restraint are to be employed a clear policy should be in place governing their use. Restraint which involves tying (whether by means of tape or by using a part of the patient’s garments) to some part of a building or its fixture should never be used’ (Department of Health, 2008, Para 15.31).

It is clear from this guidance that mechanical restraints are not standard practice and the code does not give any clear guidelines apart from the above. NICE (2005) have produced a clear policy declaration governing the use of physical interventions; however, it does not allude to mechanical restraint within these guidelines. Further, the Inquiry Report (2003) into the death of Mr. Bennett rejected any consideration of the use of mechanical restraint, although its own findings revealed that the lengthy use of physical restraints was a major factor in his death. Moreover, culturally, nurses in the United Kingdom (UK) struggle to substantiate mechanical restraint as an option to managing violence and aggression. The term ‘mechanical’ is loosely associated with ‘manacle’, both of which conjure up images of rattling chains against old stone walls in some dreary, dank, dungeon in medieval times. Further, ‘mechanical’ also gives an impression of cold steel machine parts operating without thought or feelings, merely going through the motions in a set and predetermined manner. However, as with any other area of mental health practice, advances have taken place since these times. Mechanical restraint is a widely used treatment option in some states of the United States of America (USA), with restraint devices usually custom-designed and made for each individual requirement (Wilder and Sorensen, 2001).

Indeed, for over a century, health professionals in the USA have differed from their British counterparts over the use of restraints. Many of the contemporary dilemmas involving physical restraint can be traced back to an earlier ‘restraint crisis’ that occurred during the mid-nineteenth century. Then, as now, Americans and Britons debated the advisability of the use of restraints; then, as now, American practices differed from the ‘nonrestraint’ philosophy that came to characterise UK hospitals (Strumpf and Tomes, 1993). However, it could now be argued that it has come full circle and the ‘nonrestraint’ philosophy is being advocated by the USA, considering the above discussion regarding the Pennsylvania state hospital system. The central question for nurses, therefore, is ‘which approach is least acceptable?’

There are indeed no right or wrong answers to this question. However, this could relate to what the the patient looks like. There is evidence to suggest that acceptability has more to do with the visual imagery produced than with any moral interpretation of the least restrictive practice, safety or dignity. In one paper by Maier and Van Rybroek (1990) about offensive images, the authors pointedly note that ‘managing aggression isn’t pretty’ (p.357). In this article, they debate which form of restraint is least unacceptable and deliberate as to whether using seclusion is an easier option than mechanical restraints because it is a case of ‘out of sight, out of mind’. They suggest that this is a form of denial as it is more difficult to engage with the problem when a person is in mechanical restraints, as the sight is visually more unpleasant. Moreover, they conclude: ‘we look forward to the day when images of an aggressive patient in ambulatory restraints will be as acceptable as the surgical patient attached to an IV pole or an adolescent wearing [dental] braces’ (p. 357). As unpleasant as someone who is in a mechanical restraint looks, we need to consider how unpleasant it looks when a patient is being physically restrained on the floor by up to five people. Therefore, as shocking as the above statement appears, put in the context of other forms of restraint, it may have some merit.

I searched the electronic bibliographic databases CINHAL, psycINFO, Academic Search Premier, ProQuest, Blackwell Syngery, MEDLINE, EBSCohost EJS and PubMed. The search terms that were used in various combinations to search publications from 1975 to the present day were ‘mechanical restraint’, ‘self-injury’ ‘self-harm’, ‘self-injurious behaviour’, ‘violence’ and ‘aggression’ ‘suicidal behaviour’. Twenty-one relevant articles were identified. Since the number was small, there was no need to limit the search. No empirical studies were identified that specifically investigated the use of mechanical restraints for suicidal behaviour in the UK. Therefore, assessing the evidence base for this has proved to be difficult. However, it could be posited that assessing the evidence base for physical restraint interventions in general could prove complex, as few of the NICE guidelines for the Short Term Management of Aggression and Violence in Psychiatric In-Patient Settings and Emergency Departments (2005) are actually evidence-based. The Cochrane review (Sailas and Fenton, 2002), undertaken to inform these NICE (2005) guidelines, noted the widespread and essential use of physical interventions in mental health settings but expressed concern about the lack of high-quality empirical evidence to underpin its use and, indeed, commented on the basic lack of information about the proper use and efficacy of such interventions.

Many of the articles identified related to self-injurious behaviour in patients with developmental disabilities (Singh, Dawson & Manning, 1981; Dorsey, Iwata, Reid, & Davies, 1982; Foxx & Dufrense, 1984; Neufeld & Fantuzzo, 1984; Emerson, 1992; Mazaleski, Iwata, Rodgers, Vollmer & Zarcone, 1994; Harris, 1996; Kahng, Abt & Wilder, 2001; Kahng, Leak, Vu & Mishler, 2008). The limited studies regarding the management of violence and aggression with mechanical restraints in mental health settings were all conducted in the USA (Whitman, Davidson, Sereika & Ruby, 2001; Tumeinski, 2005; Prescott, Madden, Denniss, Tisher & Wingate, 2007).

This brief and limited review highlights that for almost every type of developmental disability or disorder, there is some form of apparatus that can be innovatively adopted in an attempt to control the behaviour of the disturbed person. There is no doubt that restraints can be very effective in reducing self-injury whilst in use (Harris, 1996). However, mechanical restraints can also have a number of detrimental side effects. A major concern is that, for those with developmental disabilities, the social attention received during the application of restraints can serve to positively reinforce the behaviour and therefore increase rates of self-harm in individuals whose behaviour is maintained by this contingency (Harris, 1996; Kahng, Abt & Wilder, 2001; Kahng, Leak, Vu & Mishler, 2008), therefore making the person dependant on wearing the restraints for long periods of time (Foxx & Dufrense, 1984). Mechanical restraint has also been shown to lead to the development of other self-harm behaviours in addition to the one the device is trying to prevent; controlling one form of self-injury (e.g. hand biting) by means of restraint may simply result in other forms (e.g. hitting head against a wall: Singh, Dawson & Manning, 1981). Muscular atrophy, demineralisation of bones, shortening of tendons, arrested motor development and disuse of limbs may occur because of long-term restriction (Emerson, 1992).

Mechanical restraints are also associated with the disruption or prevention of opportunities to engage in activities associated with daily living, education and leisure and reduced levels of interaction with carers (Emerson, 1992). There are also reports of ‘incontinence, loss of ability to walk, depression, fear, panic and accentuation of cognitive impairment occur frequently. Elimination problems and cardiac stress have also been observed’ (Archea, McNeely, Martino-Saltzman, Hennessy, Whittington, Myers, 1993, p.6). Clearly, these are serious disadvantages in the use of mechanical restraints, and can be seen to be the result of bad nursing of patients in such apparatus. Deaths have also been reported as a result of the use of mechanical restraint, with one study claiming that more than fifty deaths were attributed to mechanical restraints between 1984 and 1992 (Weick, 1992). The age range of those that died was between 10 and 90 years: this displays the extent of the use of restraints in the USA. The main reasons why patients died were because of using the restraints incorrectly, the patient resisting, the patient not being monitored, the patient not being regularly freed and the wrong choice of restraint apparatus. Despite the limited evidence base on the subject, there appears to be some constraints to the use of mechanical restraint. Are there any benefits?

The advantages of utilising mechanical restraint include the fact that the patient is not left alone and has a nurse in constant attendance, allowing for interaction and communication with positive feedback, which is not readily available either when the patient is secluded or when someone is tranquilised with medication (Mason and Chandley, 1999). Mechanical restraints allow the person to vent his or her anger but keeps the person safe from damaging property or other people (Crichton, 1995). The patient is usually free to eat, smoke and use the toilet independently and therefore has a degree of autonomy; this is not the case with chemical restraints and seclusion (Howells and Hollin, 1989). Another advantage of mechanical restraints is the multidisciplinary approach that is used. Whereas the use of rapid tranquilisation and seclusion tend to leave the patient marginalised and isolated once the crisis is under control, the use of mechanical restraint is high-profile and needs constant nursing input as well as multidisciplinary reviews (Mason and Chandley, 1999).

Since there is such a paucity of empirical literature on this subject, I feel it is imperative to examine the legal and ethical implications of the use of mechanical restraint to manage violence and aggression. There is no formalised legislation on the use of mechanical restraint in adult psychiatry (Harris, 1996) besides the aforementioned Code of Practice Mental Health Act 1983. The principles underlying USA law are: first, a person should be deprived of liberty only when presenting a serious risk of harm to self or others; second, the extent of the deprivation of liberty should be that necessary to achieve safety and no more; third, the patient should choose the restraints wherever possible; and fourth, infringement on patients’ liberty should be the absolute minimum required (Blumenreich & Lewis, 1993).

If liberty is a central tenet of American society, then reasonableness is the British equivalent. This includes reasonableness in terms of both how feasible it is for professionals to predict violence and foresee harm to the patient or others and how reasonable is the restraint to prevent such injury. Diamond (1995) claims that ‘reasonableness means firstly that the force used should be no more than is necessary to accomplish the object for which it is allowed (so retaliation, revenge and punishment are not permitted), and secondly, that the reaction must be in proportion to the harm which is threatened’ (p.200).

Of the types of restraint used in psychiatry, chemical restraints would clearly qualify as treatment, whereas mechanical and physical restraints are not so much treatments for mental disorder itself as mechanisms to deal with the behaviour resulting from mental disorder. On the other hand, if the adverse behaviour is part of the mental disorder, then the use of restraint to prevent the behaviour could be seen to be part of the treatment. Whilst there is no consensus as to whether seclusion is a treatment technique, it does nonetheless fall within the definition of medical treatment and the same might be said to apply to physical and mechanical restraints. Over recent years, British courts have taken an increasingly broad view of what constitutes medical treatment for mental disorders, though their verdicts have been controversial. Tube-feeding as part of the treatment of anorexia nervosa (Whitefield, 1997) and the enforcement of caesarean section under the Mental Health Act 1983 have been of contemporary interest (Bewley, 1997). Could there be an argument for classifying mechanical restraint as a treatment for mental disorder?

Ethically, scholars emphasise affording the patient as much choice as possible regarding the types of restraint that may be used in an aggressive incident (Blumenreich and Lewis, 1993; Harris, 1996). However, where some form of restraint is necessary, it is difficult to ascertain what constitutes the least restrictive option – chemical, geographical, physical or mechanical (Howells and Hollin, 1989). Therefore, the Mental Health Act Commission and the Ethics Committee of the Nursing and Midwifery Council (NMC, 2008) accentuate the ethical need to address what is in the patient’s best interests and especially to seek to benefit the ill, retain confidentiality and avoid intentional injustice (Musto, 1991).

The Code of Practice of the Mental Health Act 1983 states; ‘If any forms of mechanical restraint are to be employed a clear policy should be in place governing their use’ (Department of Health, 2008, Para 15.31). However, there is very limited guidance and this is restricted to patients with developmental disabilities (Paley, 2008) and mental health patients in the USA (Blumenreich and Lewis, 1993). Paley (2008) states that the use of mechanical restraint as part of a protective and preventative behavioural approach is applied within the three following approaches: Level 1 – advanced planning as part of a behavioural support strategy; Level 2 – as part of a short-term therapeutic intervention in a reactive context; Level 3 – to reduce risk to individuals from their environment as a result of their behaviour which is judged to be a risk to themselves in cases where the person appears to have no control over that behaviour.

Blumenreich and Lewis (1993) go on to state that patients in mechanical restraints need very close monitoring. The restraints should be applied with care. Restraining a patient should be the equivalent of an intensive-care situation. The patient, at this point, requires one-to-one observation. The airway, circulation, skin condition and mental status must be monitored. Such basic needs such as nutrition, elimination and safety must be considered. Further documentation of these matters is pertinent, both to safeguard good practice and to protect against litigation. As is the case with seclusion, a written observation should be charted every 15 minutes (Mason and Chandley, 1999). Mechanical restraint should never be used for punishment. However, the punitive use of restraints may be covert. Careful rationalisations and skilfully worded progress notes can make their use seem appropriate. Therefore, personal biases and prejudices should not be allowed to influence decisions concerning patient care (Wilder and Sorensen, 2001). Further, clinical assessments should be made not only of the need for mechanical restraint, but also of the risk to patients. This would also include continual assessment and review in relation to the appropriateness and efficacy of the intervention. Such review intervals need to be agreed and set out in organisations’ policies (Mason and Chandley, 1999).

In this paper, we have seen that the use of mechanical restraint has become much more refined in recent years. This refinement takes the form of customised restraints developed specifically for each individual patient’s particular needs. It is clear that mechanical restraint has benefits and constraints. However, it could be argued that any form of intervention for managing the violent and aggressive person will have both good and bad points, especially for exceptionally violent persons. There are ethical, legal, organisational and professional dimensions. Ethically, practitioners need to address issues of liberty, reasonableness and the least restrictive option. Legal concerns would mean the involvement of the Mental Health Act 1983, the patient’s responsible clinician, the multidisciplinary team, the patient, his or her advocate, relatives and legal representatives where necessary. Clinical staff also need to be alert to their own feelings and beliefs about violent patients so that these are identified and not acted on.

Following a review of the evidence base, do I consider that mechanical restraint would be appropriate for the aforementioned suicidal patient? Given the paucity of literature on the subject, my decision would be tentative. However, the full range of available medical and psychological interventions has failed to produce changes in her behaviour. It is therefore argued that in such circumstances, the use of mechanical restraint would be reasonable and could prove beneficial in reducing the extent of physical and chemical restraint. Reduction of these interventions remains a key therapeutic objective, since its long-term use in this case may be perpetuating that patient’s psychopathology and abnormal behaviours.

Therefore, I take the view that mechanical restraint in such a case may be ethically, legally and clinically appropriate. However, this would need to be agreed with the abovementioned persons and would need constant review although I would argue that the prevailing ethos of care in Britain renders clinicians reluctant to support such a proposal. Mechanical restraint produces visual images that prick consciences and are unpleasant in an ideal world. Concerns could also relate, in part, to the perception that if applied to one patient, this procedure could be used and possibly misused elsewhere; thoughts of the patients being tied to beds or in straitjackets may prevail, although the device referred to for this patient is a wrist–waist restraint. Finally, given the lack of empirical data on the use of mechanical restraint, before-and-after studies, surveys, cross-sectional studies and cohort studies are urgently required to investigate the effectiveness of mechanical restraints, the ethical and legal aspects, their role within the range of physical interventions taught to staff and staff and patients’ perceptions of such restraints for the short-term management of violent behaviour in psychiatric inpatient settings.


Archea, C., McNeely, E., Martino-Saltzman, D., Hennessy, C., Whittington, F., Myers, D. (1993) Restraints in long term care. Physical and Occupational Therapy in Geriatrics, 11 (2), pp. 3-23
Bewley, S. (1997) Commentary: Bad Medicine and Bad Law. British Medical Journal, 314, pp. 1184-1185
Blumenreich, P.E. & Lewis, S. (1993) Managing the Violent Patient: a clinician’s guide. New York: Brunner Mazel
Crichton, J. (1995) Psychiatric Patient Violence Risk and Response. London: Duckworth
Department of Health (2008) Mental Health Act Code of Practice. London: HMSO
Diamond, B. (1995) Legal Aspects of Nursing. 2nd ed. London: Prentice Hall
Dorsey, M.J., Iwata, B.A., Reid, D.H., & Davies, P.A. (1982) Protective equipment: continuous and contingent application in the treatment of self-injurious behaviour. Journal of Applied Behaviour Analysis, 15, pp. 217-230
Emerson, E. (2002) The prevention of use of reactive management strategies in community-based services in the UK. In. Allen, D. (ed) Ethical Approaches to Physical Intervention. Kidderminster: BILD
Foxx, R.M. & Dufrense, D. (1984) ‘Harry’: The use of physical restraint as a reinforcer, time-out from restraint, and fading restraint in treating a self-injurious man. Analysis and Intervention in Developmental Disabilities, 4, pp. 1-7.
Harris, J. (1996) Physical restraint procedures for managing challenging behaviours presented by mentally retarded adults and children. Research in Developmental Disabilities, 17 (2), pp. 99-134
Howells, K & Hollin, C. R. (1989) Clinical Approaches to Violence. Chichester: Wiley and Sons Ltd.
Injury Report (2003) Independent Inquiry into the death of David Bennett [online]. SHA: Norfolk, Suffolk and Cambridge. Available from: [accessed 25 May 2009]
Jones, E., Allen, D., Moore, K., Phillips, B., Lowe, K. (2007) Restraint and self-injury in people with intellectual disabilities. Journal of Intellectual Disabilities, 11 (1), 105-118
Kahng, S., Leak, J.M., Vu, C. & Mishler, B. (2008) Mechanical restraints as positive reinforcers for aggression. Behavioural Interventions, 23, pp. 137-142
Kahng, S., Abt, A.K., Wilder, D.A. (2001) Treatment of self-injury correlated with mechanical restraints. Behavioural Interventions, 16, pp. 105-110.
Liberman, R. P. (2006) Elimination of Seclusion and Restraint: A reasonable Goal? Psychiatric Services, 57 (4), pp. 576
Maier, G. J. & Van Rybroek G. J. (1990) Offensive Images: managing violence and aggression isn’t pretty. Hospital and Community Psychiatry, 41 (4), pp. 357
Mason, T., Chandley, M. (1999) Managing Violence and Aggression a Manual for Nurses and Health Care Workers. New York: Churchill Livingstone
Mattson, J.L. & Keyes, J. (1988) Contingent reinforcement and Contingent restraint to treat severe aggression and self-injury in mentally retarded and autistic adults. Journal of the Mentally Handicapped Person, 1 (2), pp. 141-153
Mazaleski, J.L., Iwata, B.A., Rodgers, T.A., Vollmer, T.R., Zarcone, J.R. (1994) Protective equipment as treatment for stereotypic hand mouthing: sensory extinction or punishment effects? Journal of Applied Behaviour Analysis, 27 (2), pp. 345-355
Musto, D. (1991) An historical perspective. In Bloch, S. & Chodoff, P. (eds) Psychiatric Ethics, 2nd ed. Oxford: Oxford University Press.
National Institute of Clinical Excellence (2005) Clinical Practice Guidelines for Violence: The short-term management of disturbed/violent behaviour in psychiatric inpatient settings and emergency departments. London: NICE
Neufeld, A., Fantuzzo, J.W. (1984) Contingent application of a protective device to treat the severe self-biting behaviour of a disturbed autistic child. Journal of Behaviour Therapy and Experimental Psychiatry, 15, pp. 79-83
Nursing and Midwifery Council (2008) Standards of conduct, performance and ethics for nurses and midwives [online]. London: Nursing and Midwifery Council. Available from: [accessed 26 May 2009]
Paley, S. (2008) Use of Mechanical Devices: restrictive physical intervention principles for practice [online]. London: British Institute of Learning Disabilities. Available from: [accessed 29 May 2009].
Prescott, D.L., Madden, L.M., Dennis, M., Tisher, P. & Wingate, C. (2007) Reducing Mechanical Restraints in Acute Psychiatric Care Settings Using Rapid Response Teams. The Journal of Behavioural Health Services & Research, 34 (1), pp. 96-105
Sailas, E. & Fenton, M. (2002) Seclusion and restraint for people with serious mental health illnesses (Cochrane Review). The Cochrane Library. Issue 1. Oxford: Update Software.
Singh, N. N., Dawson, M.J., Manning, P.J. (1981) The effect of physical restraint on self-injurious behaviour. Journal of Mental Deficiency Research, 25, pp. 207-216
Smith, G.M., Davis, R.H., & Bixler, E.O. (2005) Pennsylvania State Hospital Systems Seclusion and restraint Reduction Programme. Psychiatric Services, 56, p. 1115-1122
Strumpf, N. E. and Tomes, N. (1993) Restraining the Troublesome Patient: A historical perspective on a contemporary debate. The Historical Mirror, 14, pp. 3-24
Tumeinski, M. (2005) Problems associated with the use of physical restraints in contemporary human services. Mental Retardation, 43 (1), pp. 43-47
Webb, C. (1992) Communication Skills. London: Macmillan
Weick, M.D. (1992) Physical Restraints: an FDA update. American Journal of Nursing, 92, pp. 74-80
Whitefield, A. (1997) Commentary: a decision that stretches the law too far. British Medical Journal, 314, pp. 1185-1186
Whitman, G.R., Davidson, L. J., Sereika, S.M., Ruby, E.B. (2001) Staffing and pattern of mechanical restraint use across a multiple hospital system. Nursing, 50 (6), pp. 356-362

Wilder, S. & Sorensen, C. (2001) Essentials of Aggression Management in Health Care. New Jersey: Prentice Hall.

Cite this page

Choose cite format:
Online Chat Messenger Email
+44 800 520 0055