The use of leverage in community mental health: Ethical guidance for practitioners

Introduction

A range of pressures are commonly applied within community mental health care to increase patients’ adherence to treatment (Szmukler, 2009, 2010; Szmukler & Appelbaum, 2008). In this article, we focus on the use of leverage. Leverage has been introduced recently as a new concept to describe a particular type of pressure that is distinct from persuasion and compulsion (Monahan et al., 2005; Szmukler & Appelbaum, 2008), and that can be defined by three components (Canvin, Rugkåsa, Sinclair, & Burns, 2013).

First, leverage involves the use of a specific, identifiable lever. Surveys have measured some such levers, including access to housing, receiving welfare payments and personal money, having contact with children and having contact with the criminal justice system (Appelbaum & Redlich, 2006; Burns et al., 2011; Monahan et al., 2005; Redlich, Steadman, Robbins, & Swanson, 2006). Over 50% of patients in the United States and 30% of patients in the United Kingdom report experiencing one or more of these levers when receiving outpatient services. Second, leverage is conditional. This means that the lever is applied by a practitioner who makes an explicit proposal to a patient that attaches conditions – positive or negative – to the possible courses of action between which the patient may choose. Finally, this proposal must be directly applied by people who are perceived to have the power to act upon the conditions that they impose. If the patient does not believe that these individuals are able to apply such conditions, then this proposal should not be seen as an instance of leverage.

This threefold definition of leverage does not require the proposal to successfully influence a patient’s choices. However, the ethical acceptability of the proposal will depend on an assessment of the consequences that might result by imposing conditions on patients’ choices. Case studies 1–3 in Table 1 illustrate how leverage might be used in practice. These cases were jointly developed by an experienced qualitative researcher (K.J.C.) and a consultant psychiatrist (J.M.A.S.) from examples of leverage in data collected in a recent qualitative study of treatment pressures in community psychiatric care in the United Kingdom (Burns et al., 2011; Canvin et al., 2013). The cases cover a range of different community mental health service settings and are intended to illustrate the potential dilemmas and alternative interpretations that can arise when leverage is used in practice.

OPEN IN VIEWER

Table 1.

Leverage in practice.

Case 1
Jack, 24, lives at home with his parents. Jack has a diagnosis of schizophrenia, but has consistently refused to accept that he has a mental disorder. When unwell, Jack increases his cannabis use, which exacerbates his symptoms. Jack’s parents find it difficult to cope with his aggressive and threatening behaviour. They have sought advice from a support group and have told services that they want more support for themselves and their son. The Assertive Outreach Team’s main concern is Jack’s well-being and his parents’ safety. Members of the Team have suggested to Jack that if he stops using cannabis, they will help facilitate an unsupported housing placement for him. Jack’s parents support this proposal.
Case 2
Ann, 35, is a single mother of three children. Ann has recurrent severe depression which has involved several serious suicide attempts. Ann repeatedly stops taking the medication, asserting that she feels well enough to manage without it. When she does so, her mental state and ability to care for herself and her children gradually deteriorate, resulting in long periods of hospitalisation. Ann’s situation generates multiple practical and financial support needs. The Community Mental Health Team have concerns about the well-being of Ann’s children in this unstable environment and have encouraged her to consider what might happen to her children if she is hospitalised again. They have told her that next time she has a severe depressive episode it is likely that social services will conduct an assessment, potentially leading to her children being placed on the at-risk register. Ann is extremely worried that she may lose custody of her children.
Case 3
Craig is 28. He has misused various substances since his teens and is unable to hold down a job. He currently receives a daily supervised dose of methadone as part of a Drug Treatment Order imposed by the court in lieu of imprisonment. The order stipulates that he must reside with his parents and is enforced with an electronic tag. Craig continues to take heroin on top of his methadone and has been told by his drug worker that if he wants to pick up his methadone on a daily basis to take unsupervised, he will have to return heroin-free urine tests.

In this article, our primary objective is to examine whether practitioners act ethically when they use leverage in the ways described in Table 1. More broadly, we are concerned with helping to shape the development of the concept of leverage in community mental health and in sharpening up the different ethical considerations relevant to understanding the use of treatment pressures.

The complex nature of therapeutic encounters in community mental health services means that there are no straightforward answers to the question of whether practitioners act ethically when they use leverage (Dunn et al., 2012). Indeed, as the three cases in Table 1 illustrate, it can also be difficult to ascertain whether leverage is actually being used, especially in clinical encounters in which practitioners are required to provide information or where requirements for treatment adherence are enforced using legal frameworks. In the face of such complexity, we show how different duties can guide practitioners’ judgements about whether and when it is acceptable to influence patients’ treatment adherence using leverage in a given situation. The four duties that we discuss below are those which can guide ethical practice in community mental health, irrespective of social and cultural context (e.g. Beauchamp, 2009). These duties are as follows:

The duty to benefit the individual patient

That health-care practitioners have a duty to benefit their patients is not contestable. How practitioners should act to uphold this duty in practice ought to be determined by the unique needs of each patient. For this reason, enacting the duty to benefit the patient (and therefore acting ethically) depends on context-specific factors. It ought to be obvious, for example, that practitioners should provide different services to a patient experiencing an acute phase of a psychotic illness compared to a patient receiving long-term support to manage an enduring psychotic illness. For the acutely ill patient, the scope of the duty to benefit her is likely to be focused – at least in the short term – on providing immediate psychiatric input to manage her distressing symptoms and keeping her safe. For the chronically ill patient, the scope of the duty will extend to include the facilitation of social activities, employment opportunities and contact with family.

Commonly, interpreting the duty to benefit a patient will need to take into account more nuanced differences between the needs of comparable patients within a service. For example, how this duty ought to be interpreted in decisions about antipsychotic medications will be shaped on a case-by-case basis by the associated side effects for each drug and a patient’s views about weight gain, opportunity for exercise or risk of developing metabolic syndrome.

The duty to benefit the patient may manifest itself differently depending on the therapeutic model adopted by a service. In Case 3, the duty to reduce harm caused to Craig by his illicit drug use through prescribing substitute medication will be judged to take primacy over providing assistance to Craig in meeting a duty to assist him in meeting additional social goals (National Institute of Health and Clinical Excellence (NICE), 2007), particularly when taking methadone under supervision is a requirement of a Drug Treatment Order invoked to support Craig’s rehabilitation. In contrast, practitioners working in the Community Mental Health Team (CMHT) in Case 2 are likely to interpret their duties to benefit Ann in terms of prioritising a broader set of obligations concerning positive social, employment or educational outcomes.

How the duty to benefit the patient is understood in light of the patient’s individual needs and the therapeutic culture of the service has implications for judging whether a specific act of leverage for a patient is ethically permissible. Moreover, because the application of this duty is context dependent, the same proposal might be ethical when presented to one patient, but unethical when presented to another patient. It is also conceivable that the same proposal might be ethical when presented to the same patient depending on circumstances at particular points in time. In Case 1, Jack’s team propose to provide him with a housing placement if he stops using cannabis. It is arguable that the team have a duty to benefit Jack by facilitating his access to a housing placement, because of their concerns about his well-being and his parents’ safety, irrespective of his use of illicit substances. If so, making this proposal would not be ethically acceptable: they should facilitate his housing placement anyway. Alternatively, if it were the case that Jack posed no risk to his parents, and he had not expressed any preference about his housing situation, the team might claim that their duty to benefit Jack does not extend to making alternative housing available to him. As such, they should conclude – from the standpoint of this duty – that they have no reason to make this proposal to Jack.

The duty to benefit the patient extends to ensuring that the patient is not harmed, or left worse off, by the provision of care. There are two ways that leverage might lead to negative outcomes for the patient. First leverage might lead the patient to become increasingly dependent on the service, and unlikely to obtain control over – or take responsibility for – her own recovery. In Case 1, the concern might be that leveraging Jack with the offer of a housing placement might create the expectation that such things will be provided for him, undermining his responsibility for making major life decisions in the future. Consequently, evidence about short-term risks and benefits will need to be balanced against foreseen long-term harms.

Second, patients might perceive themselves to be worse off if an instance of leverage places conditions on a course of action that they would value. In Case 2, for example, Ann’s quality of life may decrease when the proposal is made to have a children’s services assessment if she becomes unwell. Judgements about the acceptability of leverage here will depend on how patient benefit is conceptualised, and from whose perspective.

Patients’ subjective well-being may be reduced in circumstances where they feel coerced or threatened (McNeil, Gormley, & Binder, 2013; Newton-Howes & Mullen, 2011; Olofsson & Norberg, 2001; Van Dorn et al., 2006). Again, careful thought must be given to judging how the harm potentially arising out of experiences of coercion is made sense of, and how it is balanced against other ways of conceiving of good and bad outcomes for the patient.

The duty to benefit other individuals

Mental health-care practitioners have duties to other people in addition to the patient at whom leverage is directed. These duties extend primarily to other patients in the service, the patient’s carers, to children and to members of the public:

The practitioner has a duty to benefit other patients receiving care within the service. How might the requirements that follow from this duty shape ethical judgements about leverage?

If, for example, other patients supported by the Assertive Outreach Team in Case 1 learn that the team is prepared to facilitate a housing placement for Jack if he stops using cannabis, it is conceivable that other patients might expect to be made a similar offer and that failure to do so would impede their engagement with the team and adherence to treatment. Such actions could lead to deterioration in their mental state. Alternatively, as evidence from studies of financial inducements in mental health suggest, compliance among all patients might be enhanced by the use of such inducements (e.g. Giuffrida & Torgerson, 1997).

It must also be acknowledged that mental health services, particularly community-based services, have obligations to patients’ families. In many jurisdictions, practitioners must take into account carers’ needs in developing a care and treatment plan for a patient. An act of leverage by a practitioner may result in significant burdens for a patient’s carers, such that leverage equates with the practitioner acting inconsistently with this duty. In Case 1, the proposal to offer Jack a housing placement on condition that he stops using cannabis is aligned with the team’s identified duty to reduce the risk of harm to Jack’s parents. Such a proposal – if accepted – would also meet Jack’s parents’ request for additional services for their son. Moreover, it could be argued that the effect of Jack’s behaviour on his parents is such that it would be ethically unacceptable not to make this proposal.

Additionally, leveraging a patient by making an offer to her might reduce the stress experienced by carers who are required to support the patient. In Case 3, Craig’s parents are supportive of the proposal to allow unsupervised access to methadone treatment if Craig returns drug-free urine tests, and have additionally agreed to increase his weekly allowance if he remains drug free. Here, leverage would be consistent with the drug worker’s duty towards Jack’s family carers.

Professional duties extend to children and vulnerable adults who are dependent upon the patient. Commonly such duties are inscribed in regulations across different jurisdictions, with obligations on practitioners to safeguard the welfare of these individuals. Again, practitioners need to consider the likely consequences should the patient decide to accept, or not accept, the conditions placed upon their choices. In Case 2, Ann has significant caring responsibilities as a single mother of three children. The CMHT are aware that a failure to take her medication leads Ann to become unwell and unable to provide care for her children. It would be important here to assess whether using Ann’s children as a lever would itself increase the risk of Ann’s condition deteriorating such that her children are harmed or neglected. Equally, if the CMHT decide that they have an obligation to take immediate steps to safeguard Ann’s children, this duty ought to be enacted, and making this course of action conditional on Ann’s choices would be unacceptable.

It is conceivable that leverage could be used in ways that expose members of the public to risk as a by-product of providing benefits within the practitioner–patient–carer relationship. In Case 1, the behavioural problems that Jack has displayed towards his parents might be evidence of heightened risk to the public if the proposal to provide him with an unsupported housing placement is made. Alternatively, this risk may be no greater than if the proposal was not made, or the heightened level of risk can be managed such that the proposal would not infringe upon this duty.

There may also be limited obligations towards potential or future patients that fall within the scope of the duty to benefit other individuals. Such patients would include those individuals with mental health problems who have not sought help (or whose mental health problems have not yet developed), but who meet (or will meet) the criteria for treatment in the service. It is possible that practitioners’ actions could undermine the delivery of sustainable and beneficial services to all patients. There is evidence, for example, that people from black and ethnic minority groups perceive mental health services to be unresponsive to their cultural norms and values (e.g. Keating & Robertson, 2004; Leong & Lau, 2001; Moffat, Sass, McKenzie, & Bhui, 2009; Snowden, 2001), and that these perceptions have resulted in individuals not seeking help from mainstream services (Morgan & Hutchinson, 2010).

Similarly, in cases of leverage, public perceptions of practitioners making conditional proposals to patients could give rise to the view that such proposals involve manipulating disempowered, vulnerable patients into treatment that they would not otherwise agree to. This perception might, in turn, lead potential patients to turn away from mental health services, potentially with negative consequences for their mental health and well-being. Alternatively, widespread understanding that community mental health practitioners offer inducements to patients might encourage more individuals to make contact with these services.

The scope of practitioners’ duties to provide a beneficial service to the patient and other individuals is broad, but not unlimited. It should be recognised that practitioners may act in ways that are designed to bring about additional benefits for themselves, their colleagues or others. For example, a practitioner might use leverage to meet targets for reducing involuntary hospital admissions or increasing the number of patients discharged to general practitioners, set by the employing organisation. Or, a practitioner might justify an act of leverage on the grounds that she wishes to avoid the reputational damage that might result from a non-compliant patient harming herself or others: so-called defensive practice.

Such actions can only be justified if they are also judged likely to benefit the patient or the other relevant individuals to whom this duty applies. There will be situations in which the benefits accrued through the use of leverage will extend beyond the patient and other individuals to include benefits to the practitioner issuing the proposal. In such situations, these additional benefits should be seen as positive by-products from any duty-based justification for leverage, but should not motivate practitioners’ decisions to use leverage.

The duty to treat patients fairly

In addition to the obligation to deliver services that provide benefits to patients and other individuals, community mental health practitioners are obliged to treat all patients fairly. This is a well-established ethical principle endorsed across all health-care settings. If leverage is restricted to those patients who are not treatment adherent, patients will be managed in ways that are not equivalent. In particular, when an offer is made to influence a patient’s choice to accept treatment, those patients who have adhered to their treatment regimes lose out on an option. Does this, as Szmukler (2009) suggests, imply that patients are being treated unfairly?

Most arguments concerning the fair distribution of resources require people to be treated equitably, but not necessarily equally. Considerations relating to need and responsibility will be relevant here, but ethical arguments about the allocation of health resources start from facts about comparable cost effectiveness of the interventions in question (Sheehan & Hope, 2012; Williams, 1992). All acts of leverage have an opportunity cost associated with them, and community practitioners need to be attuned to the fact that different acts of leverage will lead to outcomes for patients at markedly different costs to the service.

The cost effectiveness of a specific act of leverage should be compared against other acts of leverage, other interventions provided by the service and the costs of not using leverage such as those associated with medication wasted as a result of non-adherence or increased hospitalisation. In Case 1, the team is proposing to facilitate Jack’s access to an unsupported housing placement on the condition that he agrees to stop using cannabis. The opportunity costs associated with Jack’s acceptance of this proposal should be determined. This placement will involve considerable expenditure and will require a substantial amount of the team’s time to arrange. The expected clinical benefit associated with the patient meeting the conditions of the proposal, and the likelihood of this benefit accruing, would also need to be assessed. Imagine instead that the team had proposed to improve Jack’s well-being by expanding his social circle by arranging a package of social activities for him. This alternative proposal would cost less and may involve a less substantial time commitment. Both proposals would also need to be compared against other ways of using resources within the service. Is either proposal defensible when the team could devote their time to spending the day undertaking activities for several patients in the service simultaneously?

The duty to respect the patient’s autonomy

Another well-recognised ethical duty in health-care settings is the duty to respect patients’ autonomy. Specific concerns relating to autonomy have been identified as central to making ethical judgements about treatment in mental health (e.g. Radoilska, 2012). Szmukler and Appelbaum (2008), for example, argue that the coercive force applied in leveraging patients’ decisions about treatment undermines the voluntariness of those patients’ ability to act autonomously.

This argument is problematic. Leverage involves attempting to influence a patient’s behaviour. The patient is not acting on the basis of the decision she would have made had no conditions been placed upon her decision-making. Crucially, however, the patient remains able to decide between the (conditional) choices she is presented with (Dunn et al., 2012). Importantly, it is precisely when a patient’s choices are manipulated in this way that she is forced to evaluate her values and commitments in life. In Case 2, proposing to Ann that her children may be placed on the at-risk register if she does not engage with services requires her to assess the importance of having continual access to her children against the value she places on living her life without the involvement of support. In this sense, leveraging Ann requires her to be proactive in ascertaining what it is that she wants, what she values and what is important to her. Such considerations are those that are typically seen to be relevant to the exercise of personal autonomy.

To see the imposition of conditions on decision-making as a problem of voluntariness is to conflate autonomy and liberty. While autonomy concerns the value of a person being able to exercise her free will, liberty concerns the value of a person being free to act without third-party interference (Coggon & Miola, 2011). Leverage involves manipulating the range of options from which a patient can choose, and therefore raises concerns about whether this action involves an unjustified infringement on the patient’s liberty. The imposition of such constraints does not, however, raise concerns about patient autonomy because leverage does not impact on the exercise of free will. A competent patient is able to make a decision, one way or the other, regardless of what conditions are placed upon her decision-making.

There will, of course, be acts of leverage that are problematic on the grounds that that they interfere with the patient’s ability to exercise her right to basic freedoms. Instead of the case as it is presented, imagine that a practitioner sought to leverage Ann by proposing to place her children in local authority care if she fails to adhere to keep taking her medication. Such a proposal would infringe upon Ann’s right to be free to raise her children and to forge a family. Teasing out whether leverage involves justified or unjustified infringements on liberty can be ascertained, we believe, by examining whether the proposal under consideration would lead to the mental health practitioner acting in a way that is proportionate to the benefit that is likely to be accrued in making the proposal. Thus, it would be ethically justified to leverage Ann in this way if the expected consequences of doing so were positive for either her or her children. This way of articulating the justification for threatening to restrict patients’ freedoms is, of course, the same justification that underpins other ways of justifying restrictions of personal liberty in the mental health context, most notably the use of compulsory powers of detention and treatment.

In community mental health services, patient autonomy is respected by adopting models of shared decision-making for treatment planning with a view to achieving a therapeutic alliance (McCabe & Priebe, 2004) or through the use of mechanisms for advanced treatment decision-making (Dresser, 1984; Rosenson & Kasten, 1991; Srebnik & La Fond, 1999). In circumstances where a patient has previously agreed to a comprehensive treatment plan, the use of leverage in a way that has not been endorsed by the patient in the development of this plan would be ethically problematic on autonomy grounds.

The manipulation involved in imposing conditions on patients’ choices also means that it is important to ensure that the patient is aware of exactly what will happen if she decides or refuses to accept the proposal. There is evidence to suggest that patients experience leverage (i.e. thinking that their choices are conditional) even when practitioners have left open how they will act if the patient does – or does not – act in a particular way (Canvin et al., 2013). Given the complexity in how treatment relationships and patients’ attitudes vary over time, there may often be uncertainty in how patients interpret the management of their care, particularly with regard to coercive interventions. In Case 2, Ann is told that the next time she becomes unwell and requires hospitalisation, ‘it is likely’ that social services will become involved and assess her children’s welfare. This may be an act of leverage, or, if no conditions are being placed on Ann’s behaviour, it might be an instance of attempted persuasion by the team, or a ‘simple prediction’ (Canvin et al., 2013) outlining what will happen to Ann’s children if she continues to act in the way she has previously. It is equally important, therefore, that practitioners are able to recognise when they are using leverage, and that this is made clear to patients; one practical requirement that follows from the duty to respect patient autonomy is clarifying any misunderstanding that patients might have within treatment relationships about the pressures they are exposed to.

Balancing duties and making judgements

We have outlined four ethical duties underpinning community mental health care relevant to making ethical judgements about the use of leverage. The practical requirements that follow from these duties for the acceptability of specific acts of leverage have also been considered. Practitioners must determine whether the use of leverage will involve care being provided in ways that are consistent with the requirements of these duties, regardless of whether the patient accepts or rejects the terms of the proposal. While leverage is widely used within community mental health care, the complexity of clinical practice means that a judgement about whether a particular instance of leverage is right or wrong will not be straightforward. Instead, leverage, as a mechanism for attempting to influence patients’ behaviour, can be considered to be ethical only once the practitioners have reasoned through the four duties, in light of the patient’s individual circumstances.

None of these four duties can be determined a priori as overriding, and all ought to be given due consideration. It is helpful to conceive this process of making ethical judgements not as an algorithm for decision-making but as a process akin to that of conducting risk assessments. While leverage could be used unethically, because these decisions are in clear violation of one or more of the relevant duties, it is also possible that more than one course of action may be ethically defensible.

Notwithstanding this observation, a number of difficulties associated with translating the four duties into the decision to use leverage remain. First, practitioners will need to ascertain how the specific requirements of each duty relate to each decision about leverage that they make. This is particularly challenging because we have identified competing accounts of the relevant considerations required to act in line with each duty. For example, we have shown how the ethical duty to benefit a patient differs depending on how patient benefit is conceptualised. Practitioners also need to consider the weight they give different kinds of outcomes. In particular, the comparative importance of subjective quality of life outcomes, and objective social and biomedical outcomes such as expected reductions in symptoms and increased involvement in employment or community activities, will need to be weighed up in the decision-making process.

Second, practitioners will have to make judgements between duties when these pull in opposing directions. It is possible that an ethical assessment of a particular instance of leverage concludes that it will likely lead to significant improvements in the patient’s health and well-being, but also result in other patients receiving fewer resources and extra burden being imposed on the patient’s carers. Alternatively, the identified clash might be between clinical evidence suggesting that leverage will bring benefits to the patient and others, but that such an action will fail to respect the patient’s autonomy. Should leverage be used in such situations? There are no straightforward answers to such questions, and practitioners will need to weigh up the comparative importance of particular duties in their professional setting, given the specific details of the situation in question.

Finally, there is a tension between conceiving leverage as a single, isolated proposal for the purpose of ethical assessment and the fact that such a proposal will most often be part of a multifaceted and ever-evolving ‘package of care’. While we do not advocate dissecting every care management decision made for patients within practice settings, it is important for practitioners to be attuned to how changes to patients’ care and treatment plans might amount to attempts to leverage patients in ways that are not ethically justified. Ethical analyses of acts of leverage may be undertaken beforehand or as part of a post hoc review of a case.

Clearly, the process of making ethical judgements that we have outlined here should be done in a flexible and pragmatic manner as a patient’s circumstances change or new evidence about the wider consequences of the use of leverage comes to light. Careful thought must be given to how the assessment of leverage will take place in practice and how this assessment can be incorporated into service audit. We believe that the case conference or team meeting is an appropriate setting in which such judgements can be made productively by practitioners.