Article Text
Abstract
This paper considers the ethical justification for the use of harm minimisation approaches with individuals who self-injure. While the general issues concerning harm minimisation have been widely debated, there has been only limited consideration of the ethical issues raised by allowing people to continue injuring themselves as part of an agreed therapeutic programme. I will argue that harm minimisation should be supported on the basis that it results in an overall reduction in harm when compared with more traditional ways of dealing with self-injurious behaviour. It will be argued that this is an example of a situation where healthcare professionals sometimes have a moral obligation to allow harm to come to their patients.
- Autonomy
- Clinical Ethics
- Psychiatry
- Paternalism
- Coercion
Statistics from Altmetric.com
A case history
Alison is 35 years old and has a long history of mental health problems. As a teenager, Alison started to cut herself and this has continued. In conversation Alison describes how she started to self-injure almost by accident and found that it made her feel better. Her self-injury follows a particular pattern and she becomes anxious and distressed if prevented from acting in this way. She describes wanting to stop and understands that there are better ways of coping but at the moment cutting is her preferred means of dealing with feelings of distress. Alison has spent long periods in mental health units and staff describe her behaviour as challenging. When in hospital her admissions are characterised by episodes of self-injury and attempts to prevent such behaviour. These attempts at preventing self-injury have not always been successful and on occasions she has been so desperate to injure herself that she has made use of more dangerous methods of self-injury such as ligatures. This is a fictional case and although not capturing all the complexities associated with self-injury it does point to some of the ethical issues that arise in many similar situations.
Introduction
Self-injury is a common occurrence in many mental health units and there is no agreed and empirically supported means of reducing its occurrence.1 Self-injury raises ethical and clinical challenges. There are moral questions regarding prevention of harm, especially as the behaviour often involves individuals who appear to understand the nature and consequences of their actions. There are also complex clinical issues regarding what interventions do and do not work.
I will first describe the standard clinical management of self-injury. This involves attempting to stop the person from performing self-injuring acts. I will illustrate the pitfalls of this approach and describe an alternative approach termed harm minimisation. I will argue that in certain circumstances harm minimisation is a viable and ethical alternative to more traditional interventions on the basis of a net reduction in harm.
Previous analyses of this issue have made similar proposals.2–4 Edwards and Hewitt,2 for example, have argued that among all options open to healthcare professionals, prevention is the least plausible and supervising self-injury the most appropriate form of intervention. While Gutridge3 also supports safer self-injury she argues that the individual who self-injures is not fully autonomous and she supports harm minimisation based on its therapeutic benefits.
I will add to their analysis by making a moral and clinical case for harm minimisation. This is based on the view that self-injury may in fact be an autonomous decision as it provides a means of coping with distress and therefore serves an important function for the individual. Furthermore, even if it were not autonomous, it does not follow that the individual lacks the capacity to participate in this type of therapeutic programme. This position promotes supporting autonomy and promoting independence in people, some of which are the most vulnerable in society. The clinical perspective is predicated on engagement in a psychotherapeutic process that aims to change the person's behaviour.
I will not argue that harm minimisation should be adopted routinely, and that in no circumstances individuals should be prevented from harming themselves. Harm minimisation and preventative approaches are not binary opposites and both may be appropriate in certain circumstances.
This implies that healthcare professionals sometimes have a moral obligation to allow harm. Although this appears to contravene the established ethical principle to first do no harm, I will argue that, in specific situations, this course of action may be clinically and morally justified.
First do no harm: preventing self-injury
It is estimated that among the general population the rate of self-injury is somewhere between 1% and 4%.5 In mental health units the rates are much higher.6 In hospital standard practice has been to try and stop self-injury occurring. Such a preventive strategy makes use of a range of interventions.7 These include searches of the individual and their possessions, removal of potentially harmful implements and the use of continuous observation. In more challenging cases, more intensive interventions such as seclusion, sedation and physical restraint may be used. Furthermore, patients are often detained under the Mental Health Act, which makes forced treatment and limitations of movement more likely. These measures are characterised by restrictions, attempts to increase control and on occasions the use of force. There are legal and ethical reasons to prevent harm and these will now be discussed.
Context for preventative practices
Although the laws on professional obligations differ across jurisdictions, in most countries, as in England, healthcare professionals owe a duty of care to patients they are responsible for.8 If such a duty is breached, a legal liability is imposed upon the practitioner. This liability could be addressed through criminal or civil proceedings. A failure to comply could amount to clinical negligence, if it could be demonstrated that an individual experienced harm as a result of a breach of the duty of care on the part of the professional. In addition, in circumstances where there is a real and immediate risk of substantial harm, such as death, Article 2 of the European Convention on Human Rights imposes an operational duty to protect the life of a specific individual.9 This duty is established for patients detained under the Mental Health Act10 and more recently, it has been extended to individuals who are not subject to such detention.11
In this context it is reasonable for healthcare professionals to act cautiously particularly when dealing with self-injury. Although not all who self-injure are suicidal and self-injury is not normally life threatening, it is reasonable of healthcare professionals to be concerned that failing to prevent injury may constitute a potential breach of their duty of care. Organisational policies and protocols tend to support this stance and emphasise the prevention of harm.12 The difficulty, as will be shown below, is that attempts to prevent self-injury are not always effective and have important limitations.
Preventing self-injury: the paradox of preventative practice
The first problem with a preventative strategy is that it can exacerbate rather than contain the problem. Self-injury has a purpose and follows a fairly deliberate pattern. This is described in some detail in psychological models of self-injurious behaviour.13 Self-injury is used as a coping strategy to release emotional tension. An almost stereotypical pattern of behaviour is described, whereby the individual starts to experience negative feelings. These feelings become intolerable and injury reduces tension and increases control.14 The behaviour provides a positive relief from distressing feelings that threaten to overwhelm the person. Preventing such behaviour arguably deprives an individual of an important coping strategy and may increase his or her level of distress.
The second problem is moral in nature; preventative strategies involve taking control and attempting to prevent the individual acting as they wish. For example continuous observation involves a nurse being with the patient throughout the 24-hour period, hence the person’s most intimate acts are observed by another. This fails to respect the individual's autonomy and conveys the message that the individual's choices lack value. By failing to acknowledge that self-injury is something the individual does for deliberate and important reasons the healthcare professional is acting in a way that implies the individual lacks moral agency. The message given is that the person needs to be saved from themselves. Such a violation of a person's autonomy accompanied by restrictions on the individual's basic liberties, it can be argued, thwarts the person’s interests in exercising their autonomy and as such constitutes harm.15
It could be objected here, that self-injury is not an autonomous choice and therefore the decision to engage in such behaviour should not be respected, even less valued. It is not clear, however, that people who self-injure are not autonomous; such a judgement would depend on the notion of autonomy adopted.i Many individuals who self-injure are highly functioning in other ways and appear to understand the nature and consequences of their actions. Thus it is possible that the choice they make to self-injure is made autonomously in the same way as individuals choose to smoke, drink or engage in dangerous recreational activities. All these are harmful and yet regarded as legitimate choices. Furthermore, even if the choice to self injure were not autonomous, their still remain moral and clinical questions about the means used to prevent such behavior. Enforced interventions are often ineffective and are certainly perceived negatively.16
Significant infringements on basic freedoms are likely to produce a confrontational rather than therapeutic environment that increases levels of distress and reduces the chance of a positive outcome in the longer term.17 In such circumstances attempts to take away someone's ability to self-injure reduce their coping options and are likely to increase their distress or increase the risk of harm. For example, it must be noted that many individuals who self-injure have a history of abuse or trauma and preventative measures may increase their feelings of powerlessness and in extreme cases result in additional trauma and therapeutic alienation.18 This increases the risk that individuals will self-injure covertly, in more dangerous ways, or attempt suicide. For example, anecdotal evidence suggests the use of ligatures is increasingly prevalent in mental health units often among individuals whose preferred form of self-injury is to cut their skin. In some cases this can be fatal. This occurs in spite of high levels of observation.
In these circumstances, at best what restrictive measures achieve is a reduction in the number of incidents but bring with them the danger of more serious harm. When applied routinely they encounter the classic objections to paternalism. They do not always work and the utilitarian argument whereby the constraints on fundamental liberties are justified in terms of a reduction in the overall level of harm fails. The actions taken to prevent the behaviour can be more harmful in their consequences than the behaviour itself.
If we add to this the fact that enforced intervention affects key personal interests such as, ‘autonomy, bodily integrity, privacy, property and liberty’,19 it becomes clear that routine prevention is likely to lead to a net increase in harm. Let us now return to the vignette described above to start to articulate an alternative position.
An alternative approach: harm minimisation
Rather than trying to stop Alison cutting herself the clinical team has agreed that she be able to access clean razors for her own use and that staff should work with her to help her understand how to injure herself more safely. In return Alison has agreed to participate in psychological therapy, where she will explore the meaning and function of her behaviour as she tries to reduce and ultimately stop it.
Alison's story illustrates an alternative based on the principles of harm minimisation. These are interventions that aim to reduce the immediate harm associated with self-injury. In practical terms the approach may include providing access to sterile cutting implements for personal use and education regarding how to injure more safety.20 For example, explaining basic anatomy and physiology and ensuring the individual understands concepts such as sepsis and the implications of infection. The development of first aid skills and an understanding of the possible consequences of self-injury may support this. The development of problem-solving skills and plans of what to do if a crisis occurs are integral to the approach. These interventions form part of a longer-term strategy to reduce the likelihood that the person resorts to self-injury. Access to psychological therapies designed to support individuals explore the meaning and function of their behavior and help promote change must also occur; this is an essential component of harm minimisation. Self-injury is being allowed, in order to maintain its role as a coping mechanism, based on the understanding that this occurs safety. Without the access to psychological therapies designed to facilitate change the arguments supporting harm minimisation are weakened significantly.
Harm minimisation is a concept drawn from the discipline of public health, with wide application in areas such as substance misuse where the approach is associated with needle exchange programmes and supervised injection sites.21 The aim is to prevent harm by reducing the potentially harmful consequences of engaging in high-risk behaviours. For example, in substance misuse, harm reduction provides an alternative to abstinence. The approach has been criticised. For example Christie et al22 note that it encourages drug use, it sends a mixed message and it fails to get people off drugs. Whether it is cost-effective and its validity as an appropriate treatment have also been questioned.23 Critics of harm reduction, particularly in the context of substance misuse, tend to focus on the moral issues involved, for example, heathcare professionals condoning illegal activity. In the context of self-injury it has to be accepted that there is limited evidence of its effectiveness but the high incidence of self-injury found in mental health units suggest that traditional approaches to dealing with self-injury are not effective. It is argued here that there is a strong moral reason to consider alternatives and harm minimisation provides an alternative to traditional ways of working. Although evidence is weak or not available, proponents suggest it is a more realistic and pragmatic response to a complex health and social issue.
As we have seen there are accounts of using harm minimisation to support people who self-injure in mental health units in both the philosophical and clinical literature. Its use has been a pragmatic and direct response to the needs of people who self-injure. Inkle,24 for example, argues that listening to people who use self-injury as their main coping strategy invites professionals to consider the value of prohibiting such a strategy; prohibition, in her words, is a form of control rather than support.
In cases such as Alison's, harm minimisation emerges as a therapeutic option for a number of reasons. Self-injury may range from superficial cuts to potentially life-threatening injury and intervention should reflect the risks involved. Where the risks of serious injury are low limitations on basic freedoms are more difficult to justify. Furthermore, where self-injury is used as a way of coping, harm minimisation provides a means by which healthcare professionals can engage therapeutically with individuals, as they work with the individual as an equal and are less likely to be seen as someone who exerts power and control. This is clinically important, as there is evidence that a positive therapeutic relationship leads to more favourable outcomes across a range of diagnoses and treatment settings.16
As part of the therapeutic process, self-injury is accepted as necessary during a period when different coping strategies are developed. For many individuals this acceptance of risk may prove beneficial and provides the flexibility to work with the individual in a way that aims to contain rather than control risk based on an understanding of what works for the individual. It does not promote harm but aims to reduce it. Pembroke notes that the majority of health risks associated with self-injury relate to the dangers of permanent injury or infection. For example, the risk of infection due to using dirty cutting instruments, or sharing such implements or cutting into areas that may risk serious injury.25
Harm minimisation is designed to reduce risks of this sort. The provision of education and access to sterile equipment reduces the danger of infection, haemorrhage, permanent injury or death. These strategies are also accompanied by advice around dress, make-up and camouflage that support individuals in managing longer-term harms such as scarification. Incidents of self-injury are accepted but the harmful nature of the self-injurious act is reduced and the patient benefits through participation in a therapeutic process.
Finally harm minimisation respects and enhances autonomy. Self-injury is supported as an autonomous choice and participation in the programme is based on consent. It provides a means whereby individuals are able to express what they are feeling inside and by doing so make a clear statement about both their identity and agency. The approach is controversial and I will now briefly address three possible objections to harm minimisation.
Harm minimisation: objections
The first objection concerns consent.26 As stated above harm minimisation requires the individual to consent to participation in the process, it cannot be enforced. In the context of drug misuse, Charland asks whether it is possible for individuals to consent to a programme of intervention that involves harm, when it is not clear that the individual can control their behaviour. Whether the act of self-injury is autonomous raises philosophical questions that go beyond the remit of this paper. However, most people who self-injure appreciate the nature and consequences of their choices. But even if they are unable to control their actions, that does not mean they are unable to consent to a certain form of intervention.
Even if it is accepted that self-injury is symptomatic of a mental disorder, it is to be noted that many people with mental health problems still have some degree of autonomy and few lack autonomy completely. In English law, adults are assumed to have capacity unless demonstrated otherwise, and it is accepted that mental health problems do not necessarily jeopardise the capacity to make decisions about healthcare, even in relation to the mental health problem itself.27 For these reasons it is reasonable to assume that at least some individuals who self-injure may be able to make autonomous and capacitous decisions about their therapy even if they struggle to control their behaviour. At least some individuals can therefore make competent decisions about participation in a harm minimisation programme.
The second objection concerns allowing the person to self-injure. It may be argued that supporting someone to injure him or herself encourages the problem, thereby undermining treatment efforts. The concern may be that such an approach suggests that it is acceptable to injure yourself and to risk the harm that may result from your actions, therefore supporting a maladaptive coping strategy. This could lead to an increase in harm and in the worst case scenario be fatal. For these reasons some psychological therapists insist that the cessation of self-injury is a precondition for therapy and continuation of the behaviour may result in the therapy being withdrawn.28
The problem with this objection is that individuals, for whom harm minimisation is a possibility, have already accepted the use of self-injury as a routine and necessary part of their life. The risks associated with the behaviour do not constitute a deterrent. Although not necessarily addictive, the behaviour has some similarities with addictive behaviour and is difficult to stop. Allowing some degree of self-injury is a realistic and pragmatic approach to addressing the issue and in some cases may be the only option the individual is willing to accept. The patient may in fact see insisting on stopping the behaviour as indicative of a poor prognosis and this may lead to an increased sense of helplessness. Furthermore, as already indicated earlier, insisting on cessation may be interpreted in a pejorative way by the person who self-injures, perceived as implying that self-injury is wrong, and this may inhibit the development of an effective therapeutic alliance as moral and clinical issues become confused.29
The final objection concerns allowing harm to occur within a clinical setting. As we have seen, healthcare professionals have an obligation to do what their profession requires, and in healthcare the principle of ‘first do no harm’ has a long tradition.30 Allowing self-injury thus appears, at least at first sight, to go against the professional's duty of care, as reasonable care must be taken to avoid acts or omissions, which are likely to cause harm.
A weakness with this argument is that although the maxim to do no harm has such credence, if applied precisely it would mean that many healthcare interventions would not be provided. Many medical interventions do harm and yet are provided routinely and are seen as ethical and lawful. Painful surgery or other invasive procedures are obvious examples.31 When the benefits outweigh the harm, the intervention is normally justified. For example, antipsychotic medications have significant side effects such as weight gain and sedation, yet are generally accepted due to the control of distressing symptoms that allows the individual to function independently.
On a day-to-day basis healthcare professionals working in mental health facilities are constantly weighing the risks of harm that interventions may entail against the likely benefits for patients. Harm minimisation may lead to an increase in the number of actual incidents, but the approach is justified by a reduction in the harmful nature of the injuries sustained. More restrictive approaches may reduce the actual number of incidents but the level of harm may increase. On this view, we are permitting harm to the person, as on balance the approach results in a net reduction in harm.
Conclusions
There are good reasons to try and prevent harm and these usually lead to attempts to prevent self-injury within a mental health setting. This paper has challenged this perspective and shown how there are problems associated with a preventative strategy. It has been argued that an appeal to a prevention of harm cannot be sufficient reason to support the routine use of restrictive practices. This is not to say that restrictive measures should never be used, as there will be situations where self-injury poses such immediate and serious risks that the only option is to try to prevent it. This paper has argued that harm minimisation provides a viable, ethically and clinically sound therapeutic option in dealing with self-injury, one that combines a reduction in overall harm with fewer restrictions on autonomy. It does not involve a blanket permission to self-injure with staff allowing significant harm to occur. Harm minimisation is a sophisticated approach to intervention and in adopting the approach the healthcare professional allows the infliction of a lesser harm to prevent a more serious harm and as such the harm allowed serves a legitimate function.
There are without doubt practical difficulties in implementing such an approach, as organisations will struggle with the legal and ethical implications, while many healthcare professionals will struggle with the idea of supporting harm in the context of a therapeutic relationship. However, it has been argued that healthcare professionals may sometimes have good reasons to allow harm, in fact they routinely do so; allowing harm is not necessarily contrary to the professionals’ duty of care, and in fact it may be required if the benefits are significant and likely to outweigh such harm. Harm minimisation provides a means of working with an individual in a way that recognises their autonomy and accepts that they have a different way of coping with distress. By trying to prevent their injury we harm them, we may fail to help them.32 I conclude that healthcare professionals sometimes have an obligation to allow harm.
Acknowledgments
The author thanks Nicola Glover Thomas and Simona Giordano for their support and for providing detailed comments and suggested amendments to earlier drafts of this paper.
References
Footnotes
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
↵i I am indebted to Simona Giordano for this point and the main argument outlined in this paragraph.
Other content recommended for you
- Balancing costs and benefits: a clinical perspective does not support a harm minimisation approach for self-injury outside of community settings
- Harm minimisation for self-harm: a cross-sectional survey of British clinicians’ perspectives and practises
- Retrospective analysis of reported suicide deaths and attempts on veterans health administration campuses and inpatient units
- Threats and offers in community mental healthcare
- David Oliver: The risks of discharging patients early against doctors’ judgment
- Self-harm in immigration detention: political, not (just) medical
- Hospital admissions for self harm after discharge from psychiatric inpatient care: cohort study
- Longer term management of self harm: summary of NICE guidance
- Antidepressant treatment and the risk of fatal and non-fatal self harm in first episode depression: nested case-control study
- Medical involvement in torture today?