Securing therapeutic justice through mediation: the challenge of medical treatment disputes

Abstract

This paper explores the use of mediation in medical treatment disputes through the lens of therapeutic justice (TJ), a concept developed in the 1990s to consider the therapeutic and anti-therapeutic effects of justice systems. The paper argues that mediation may be a mechanism for achieving therapeutic effects for people involved in medical treatment disputes. In doing so, the paper highlights the conflict that can often arise between healthcare professionals, family members and patients in medical treatment disputes and the related difficulties with using litigation to resolve this type of conflict. It has been suggested by judges, academics and policy-makers that mediation might be a better way of resolving conflict in these cases. While mediation and TJ have much in common, the paper explores the many tensions between them, considering ways in which mediation might need to be done differently to achieve therapeutic aims. Finally, the paper identifies six TJ features against which mediation can be tested to consider whether it can live up to the claims that it can be used to resolve medical treatment disputes more therapeutically.

Introduction

Medical treatment disputes are cases in which there is a disagreement about the provision of healthcare to a patient. They may involve a disagreement: between an adult or child patient and the healthcare professionals (HCPs) about what healthcare they wish to receive or refuse; between the adult and HCPs about whether the adult has capacity to make his or her own decisions; between the HCPs and adult or child's family; or between patients and family members themselves with HCPs being involved only on the periphery. The kernel of similarity is that there is a disagreement about the provision of healthcare to a patient, which could result in legal proceedings to enforce or deny treatment through judicial order. The backstop of litigation is an important dimension here; the potential for a judicial order can be viewed as enforcing rights or removing choice for those who disagree. A judgment offers the lure of finality and the reassurance of procedural safeguards. Litigation has its drawbacks, however, not only in terms of costs and time and associated stress, but in the way lay participants are often sidelined.¹ A court judgment offers finality but not necessarily resolution of the problem or restoration of relationships and trust.²

Conversely, when we look to mediation, as we do in this paper, we see a process which does not determine legal rights and wrongs but is intended to be voluntary, flexible, confidential and party-led.³ Mediation can take place alongside court proceedings but can also take place at an earlier point, before disagreement becomes entrenched, or following court determination. While not in complete opposition to litigation, mediation's distinct features make it a fundamentally different approach to resolving disputes. The mediator, an impartial third person, facilitates the dialogue, but cannot impose any solution on the parties.

This paper considers whether mediation may be a mechanism for achieving effects that can be ‘therapeutic’ for people in medical treatment disputes. We draw on the therapeutic jurisprudence literature,⁴ a movement which emerged from mental health courts and problem-solving criminal justice courts. The concept has developed globally with a focus on the emotional and psychological wellbeing of people involved in court proceedings.⁵ Theorised in the 1990s by Wexler and Winick, the approach explores ‘the extent to which substantive rules, legal procedures and the role of lawyers and judges produces therapeutic or antitherapeutic consequences’.⁶ We use the term ‘therapeutic justice’ (TJ) here rather than therapeutic jurisprudence, because ‘jurisprudence’ imports a narrower focus on courts and their decision-making processes. In contrast, we explore the role of mediation as a mechanism through which TJ might be achieved. We do not outline a complete theory of TJ in this paper; instead, we focus on the potential TJ features of medical mediation.

The paper starts with an introduction to the context of medical treatment disputes in England and Wales and the conflict that can arise in healthcare. Following this, we outline the meaning of TJ and consider its overlap with the literature on mediation. We then explore the synergies between TJ and mediation. Specifically, six TJ features are set out which are applied to mediation: promotes participant wellbeing, less adversarial, collaborative, flexible, voluntary, and participatory.⁷ Finally, we apply these features of mediation for medical treatment disputes to two reported judgments, highlighting the challenges of securing TJ through mediation as well as its therapeutic potential.

1. Medical treatment disputes

Medical treatment disputes have been the subject of numerous studies seeking to identify the factors that increase the risk of conflict and suggesting ways to prevent and resolve conflict. In this section we outline the legal framework and decisions in two high-profile examples of conflict, Archie Battersbee and Aintree v James,⁸ which show how difficult the litigation process can be for those involved. We then provide an overview of the literature on conflict in medical treatment disputes, highlighting some of the reasons why litigation has been criticised and mediation suggested.

The legal frameworks in England and Wales centre on the best interests test for adults (under the Mental Capacity Act 2005 (MCA 2005)) and children (drawing on the welfare of the child principle in the Children Act 1989 and the UN Convention on the Rights of the Child). While the legal frameworks differ, there is value in considering the cases synchronously, with each being governed by the best interests principle and an overlap in judicial personnel.⁹ The Court of Protection (CoP) can make medical treatment decisions on behalf of an incapacitated adult over the age of 16, if it is in their best interests. A 16- or 17-year-old cannot withhold consent, meaning that where someone else with authority to do so¹⁰ consents on their behalf, they can have treatment forced on them against their wishes. For children under the age of 16 the test of Gillick competence applies.¹¹ If they are found to be Gillick competent, then the child can consent to medical treatment. If they do not reach the threshold for Gillick competence, then others will have to make decisions about their healthcare on their behalf.¹² However, courts will not order HCPs to provide treatment against their clinical judgement; rather, they can determine whether available treatment, or withdrawal of it, is in the patient's best interests.¹³

The high-profile case of Archie Battersbee concerned a 12-year-old boy who had been found by his mother, suspended by his neck from the banisters in the family home. He was transferred to hospital, but never regained consciousness. The dispute was whether it was in Archie's best interests for mechanical ventilation to be withdrawn. The litigation became very hostile; for example, Archie's parents accused the hospital staff of falsifying medical documentation and starving their son.¹⁴ On the other side, the family said they had been very upset when three days after Archie's admission to hospital a consultant had raised the question of organ donation, as it had given the impression that the hospital was giving up on Archie.¹⁵ Ultimately the case culminated in a decision by Hayden J, upheld on appeal, that the treatment Archie was undergoing was ‘futile, compromises his dignity, deprives him of his autonomy, and becomes wholly inimical to his welfare. It serves only to protract his death, whilst being unable to prolong his life’.¹⁶ It was held that continuation of ventilation was not in Archie's best interests and it was withdrawn at hospital; Archie died on 6 August 2022.

We see similar disagreements about medical treatment in relation to adults.¹⁷ For example, in Aintree v James ¹⁸ the dispute was about the withdrawal of life-sustaining treatment for a critically ill man in his late sixties, with HCPs and family members disagreeing on the course of action best for him. Mr James depended on respiratory support and clinically assisted nutrition and hydration. Although he was diagnosed to be in a minimally conscious state, it was said that he was able to recognise people and enjoy their company.¹⁹ The NHS Trust sought a declaration that it was in Mr James’ best interests to withhold certain forms of life-sustaining treatment in case of a clinical deterioration, whereas Mr James’ family believed he could still enjoy the time spent with his family and friends. The CoP dismissed the NHS Trust's application, but the case reached the Supreme Court, who considered that Jackson J was right to decide that the treatment the Trust wanted to withhold could not be said to be futile or overly burdensome if weighed against the benefits of continued existence. Nevertheless, given the significant deterioration in Mr James’ condition, the Supreme Court also agreed with the approach taken by the Court of Appeal.

In light of the numerous challenges that litigation of medical treatment disputes entails, there has been sustained academic analysis considering issues such as the appropriate test for best interests, the role of religious beliefs and a move to a significant harm threshold. Much of this analysis has adopted a doctrinal perspective.²⁰ There have also been several studies that have explored the causes of conflict before the cases reach the courts.²¹ Moreton identifies three groups of causes in disagreements relating to children: internal causes; relational causes; and external causes.²² Internal causes include psychological reasons, differences in views based on religious or other beliefs and differing expectations of medical treatment. Relational causes refer to problems with communication, the behaviour of HCPs and parents’ perceptions of their role. Finally, the external causes point to the involvement of third parties and the role of social media. As Moreton notes, the causes may be ’complex and multi-faceted’,²³ with several factors present in the same case. Some have also looked at conflict in relation to adults, particularly those in intensive care settings.²⁴ The authors of these studies point to problems with communication, lack of trust²⁵ and differing views on best interests,²⁶ which are also reported in studies on children. In addition, lack of psychological support for HCPs,²⁷ doctors’ paternalism,²⁸ job strain and the control of pain and other symptoms²⁹ were reported as reasons for conflict in studies on adult intensive care.

Although there is an overlap, the causes of conflict identified by HCPs and other parties vary. HCPs identified communication breakdown, disagreements over treatment and parents’ or family's unrealistic expectations of clinical outcomes as the most common causes of conflicts.³⁰ Parents felt conflict arises when their role or expertise is challenged or disregarded and that stress, exhaustion and helplessness leave them more vulnerable to communication breakdown and subsequently conflict with professionals.³¹ They also referred to differing understanding of suffering and of best interests, as well as different approaches to decision making. An empirical study³² indicates that often the determination of a child's best interests with respect to withdrawal of treatment consists of HCPs convincing the parents that the medical view of the child's best interests is the right one. In the case of an impasse, HCPs tend to avoid litigation, and often the deterioration in the child's condition makes the parents change their view and agree with the medical advice. This may result, in some cases, in children being subjected to burdensome treatment for prolonged periods of time. Importantly, the study also found differences in parents’ and HCPs’ perception and approaches. For example, parents were unanimous in their focus on the wellbeing of the child, whereas HCPs said they tended to consider the interests of the family as a whole. HCPs also wanted to spare the parents the burden and eventual guilt of making the decision, whereas at least some parents wanted more independence.³³

Mediation is frequently proposed,³⁴ either as a tool among others, such as seeking a second opinion and the involvement of ethics committees, or as a preferable tool,³⁵ to prevent or resolve conflicts. For example, we have seen judicial statements regarding the use of mediation from Mr Justice Francis in the Charlie Gard case and in subsequent campaigns by Charlie Gard's family and others. ³⁶ In Gard, the HCPs considered that Charlie was beyond hope of any meaningful recovery, yet his parents disagreed, believing that alternative treatment provided him with some hope of recovery and invoking global media support for their cause. Charlie Gard's parents have become advocates for law and policy change, and making mediation available when disagreements arise is a key aspect of their proposal. In the Battersbee case mediation was suggested but was said to have been refused by the parents.³⁷ Arbuthnot J observed that she was ‘not convinced with the polarity of the Trust and the parents’ positions that it [mediation] would lead to a conclusion which was acceptable to both parties’.³⁸ This is interesting, as it suggests that mediation may only be suited to certain types of case or can only be attempted early in the dispute.³⁹ However, there is very little empirical evidence to support any of these calls for mediation.⁴⁰ Other alternatives to litigation have been considered, such as Clinical Ethics Committees (CECs),⁴¹ but they are quite distinct from mediation, being an internal hospital process without an independent, neutral third party. As a result, they may be seen by patients, and families in particular, as an extension of the hospital and more likely to favour HCPs.

Whereas a number of studies have explored how HCPs and parents⁴² experience the decision-making process and the arising disagreements, none have analysed their experience of mediation. Likewise, none relied on TJ as a theoretical framework, even though there is a clear demand for resolving disputes in a more therapeutic way for everyone involved.⁴³ It is this intersection between mediation, TJ and medical treatment disputes that this paper explores in light of the persistent calls for mediation's use.

2. Mediation and therapeutic justice: a meeting of minds?

TJ as an approach to law and its application by and to the court system started to emerge in the US in the 1970s as an attempt to better address the problems arising at the intersection of crime and mental health issues.⁴⁴ The novelty was to focus on a traditionally underestimated impact of law: ‘on [the] emotional life and psychological well-being’⁴⁵ of people involved in court proceedings. TJ can be defined as the ‘use of social sciences to study the extent to which a legal rule or practice promotes the psychological and physical well-being of people it affects’.⁴⁶ It is based on the premise that legal rules and procedures, as well as actions of legal actors, such as judges and lawyers, are social forces which can have both therapeutic and anti-therapeutic consequences.

TJ scholarship does not provide a clear-cut definition of ‘therapeutic’ (or, in fact, ‘justice’); this divides legal scholars, with TJ's conceptual fluidity being viewed in both positive and negative terms.⁴⁷ Wexler and Winick have left the concept of ‘therapeutic’ intentionally vague, to allow researchers to apply their intuition to interpreting its meaning in individual cases, allowing for responsiveness to the circumstances.⁴⁸ Winick underlines the importance of the viewpoint of the person at the centre of the process as to what is therapeutic to them, rather than relying on professional or clinical definitions that might reinforce power imbalances,⁴⁹ albeit some authors do refer to mental health and psychological wellbeing.⁵⁰ In fact, the term ‘therapeutic justice’ is not one that should be broken down into its parts – it is a concept that is explicitly about how justice can be therapeutically understood, in contrast with traditional views of justice in substantive or procedural terms. What a TJ process looks like is, we argue, more amenable to precision than defining the concept of ‘therapeutic’ because the justice process can be designed to draw out what is subjectively more therapeutic for those involved. In this paper, we have identified key features of TJ that can be applied to mediation, and tested through further research, to more carefully define TJ at the conceptual level.⁵¹

TJ is also not simply another term for procedural justice.⁵² TJ requires a substantive commitment to considering the wellbeing impact on participants. This requires looking beyond what is substantively or procedurally ‘just’ from a legal perspective, to looking at the social dimensions of the justice process – to consider how it makes people feel and the extent to which it impacts their lives, including their health, economic, social and cultural experience. That does not mean that TJ trumps questions of justice; rather, that it seeks to provide an understanding of a new dimension previously sidelined in traditional analyses of justice. According to Winick, ‘TJ does not suggest that therapeutic considerations should outweigh other normative values that law may properly seek to further. Rather, it calls for an awareness of these consequences and enables a more precise weighing of sometimes competing values.’⁵³ We agree that whether therapeutic aims ought to outweigh legal questions of, for example, best interests, in medical treatment disputes is a normative question and one which would need to be addressed were mediation to be expanded in this field. Moreover, we would need to better understand whether mediation does, in fact, pose any challenges to the application and interpretation of the legal principles here, particularly the best interests test, something considered below.⁵⁴ Whether TJ values ought to be prioritised over other considerations in medical treatment disputes is beyond the scope of this paper. Our claim is more limited: we seek to highlight possible ways in which mediation might be a mechanism for achieving TJ, in light of the many identified challenges involved in litigating medical conflict.

TJ is a suitable lens through which to study mediation, despite its traditional focus on courts. There has been some scholarly analysis of the relationship between the two,⁵⁵ albeit with surprisingly little overlap.⁵⁶ For example, according to Kupfer Schneider, because mediation was established with many TJ ideals, mediation can be tested for its ability to accomplish those ideals.⁵⁷ We agree that the ‘coupling of therapeutic jurisprudence and mediation could be a potential combination or even a potent one…’.⁵⁸ Historically, TJ developed as part of a larger comprehensive law movement to reform the justice system, and the proliferation of mediation was another vector of the same process.⁵⁹ Both TJ and mediation rely heavily on the psychology of procedural justice⁶⁰ and are driven by similar values, such as respect for participants’ dignity and autonomy, their voice, participation and trust.⁶¹ Mediation ‘could be described as conflict resolution in a “therapeutic key”’,⁶² which reinforces the fundamental link between TJ and mediation. Mediation is, in principle at least, considered ‘TJ-friendly’.⁶³ The proponents of TJ acknowledge that litigation, especially in the ‘traditional’ adversarial models, can be very stressful for participants. This is also one of the key arguments in favour of mediation in the medical treatment context, compared to mediation in other areas such as civil disputes, where damages are the primary remedy. In medical mediation disputes, the issue is not financial damages for past negligence, but immediate decision making on whether a particular healthcare treatment should be provided. In these contexts parties are said to experience litigation as exacerbating, rather than resolving, conflict, even where there is a judicially enforced outcome.⁶⁴

Facilitative mediation is the approach most practised in medical treatment disputes, and indeed most family and civil disputes, in England and Wales. A facilitative mediator guides discussion, prioritises parties’ self-determination and supports the parties to flexibly develop creative problem-solving and to reach agreement (or not) on their own terms, rather than having a decision imposed. Mediation may take place in a single meeting or over a period of time. Importantly, the mediator has a ‘relatively modest role’ in relation to outcome, enabling ‘the parties to choose the norms that are used for resolution of the dispute’.⁶⁵ These are some of the aspects of mediation which might provide therapeutic effects for parties; if parties can participate in the creation of solutions and feel that they have a say in the process of resolution, then they may feel an improvement in their wellbeing. Whether this does in fact happen requires empirical verification, research which we are currently carrying out, albeit there is evidence from other areas of mediation achieving these impacts.⁶⁶ However, a risk is that mediation for medical treatment disputes is not currently regulated in England and Wales, and there are no established training and knowledge requirements.⁶⁷ Mediation takes place with little or no external scrutiny and, where proceedings have not been issued, there is no judicial oversight of mediated outcomes. Notwithstanding moves towards the professionalisation of mediation practice in recent years and the development of sector-specific practice standards, there is a lack of oversight of and guidance for mediators in this sensitive area, which may limit the effectiveness of safeguards against anti-therapeutic mediation practice.⁶⁸

Facilitative mediation most closely reflects the values of TJ⁶⁹ as it has a normative element that presumes that parties derive benefit from being decision-makers, uninfluenced by the views of the mediator. This decision-making role for parties distinguishes mediation from litigation, in which parties hand over responsibility for decisions to a third party. This is a potentially key therapeutic benefit: voluntarily engaging in a mediation process enhances ‘the parties’ autonomy by encouraging their active participation’ and is ‘considered therapeutic because it contributes to individuals’ development and psychological well-being’.⁷⁰ There are several synergies between mediation and TJ, particularly for those mediation approaches that prioritise parties’ self-determination, wellbeing and participation and which, we argue, are the key aspects of mediation's therapeutic potential.

From our deductive analysis of the literature, we have identified six features of TJ (promotes participant wellbeing; less adversarial; collaborative; flexible; voluntary; participatory), which we apply to analyse mediation's effectiveness as a TJ mechanism, exploring both the potential and the risks of mediation. We agree with Wexler and Winick that TJ is a ‘new scale on which legal rules, procedures, and roles can be weighed’.⁷¹ However, this requires the identification of the features upon which the scale can be applied and then measured.

(a) Promotes participant wellbeing

It is one of the key tenets of TJ that law and legal processes have an impact on the wellbeing, feelings and self-esteem of those involved.⁷² Securing wellbeing, then, is arguably the most fundamental feature of a TJ approach. The meaning of participant wellbeing, much like the concept of ‘therapeutic’, is deliberately left vague by TJ scholars to allow for different benefits to accrue dependent on the nature of the dispute and the needs and circumstances of the parties. However, Winick and Wexler have argued that ‘TJ is the study of law's impact on psychological wellbeing’.⁷³ This incorporates a focus on the psycho-social aspects of law's rules and processes rather than, for example, the material or health effects of law.

What wellbeing entails will be different for participants depending on the context, and we suggest that empirical evidence is needed to consider what wellbeing might entail in medical treatment disputes. However, we surmise that wellbeing, as described by Wexler and Winick, is likely to correlate with participant experience of positive over negative emotions. Snedker, for example, acknowledges that emotions are an integral part of problem-solving jurisprudence⁷⁴ and therefore we must be alive to the impact of the justice process on participants’ emotions. It is not only the impact of the outcome, but the process itself that matters. Appearance in court can be stressful and on a TJ analysis, justice systems which produce trauma for participants are not therapeutically just. In the case of high-conflict medical treatment disputes, there is a strong argument for a different approach which prioritises participant wellbeing through the design of the process itself – for example, by incorporating practical approaches that make participants feel more at ease and experience fewer negative emotions and creating a safe and confidential space for discussion of difficult and sensitive issues. Moreover, being listened to and participating in the decision-making process may give some people a sense of empowerment and improved self-esteem,⁷⁵ which has positive impacts on wellbeing. By way of simple example, this might include familiarisation visits at the mediation venue in advance so that the participants know what to expect. Testing this through mediation would include asking participants how they felt before and after the mediation and to consider their emotional reflections on the mediation process.

While wellbeing may be associated with positive emotional associations, it cannot be assumed that wellbeing is correlated with self-determination and decision-making power. Waldman, for example, has noted that some parties might obtain more therapeutic benefit from knowing that a third party trained in the law has determined the outcome of their dispute, one in accordance with accepted legal and social norms.⁷⁶ This may be important for those in civil disputes with a financial settlement (for example clinical negligence), as receiving professional advice on the outcome may be desirable. Similarly, family members who do not wish to make a decision about, or agree to, an action that would result in ending the life of their loved one, may derive a therapeutic benefit from a third party making a decision to, for example, withdraw treatment. Submitting to a court decision may enable them to continue ‘fighting’ for their loved one, an advocacy role that might contribute to family members’ wellbeing. Where a party finds decision making to be an unwelcome burden, it could be considered that not honouring his or her choice to hand over decision making to someone else potentially undermines their feeling of wellbeing. For some parties, the closure offered by a judicial determination brings therapeutic value in their wellbeing, whereas the uncertainty of closure in mediation and the need to be an active participant in making difficult decisions about a family member may be anti-therapeutic.

Mediation could, however, undermine wellbeing if it is seen as a tool of persuasion by HCPs. Research in children's cases suggests that doctors succeed in persuading parents, through more informal processes, that treatment should be withdrawn.⁷⁷ In one study, when parents initially disagreed with the proposed withdrawal or limitation of invasive treatment, ‘considering the best interest of the child’, allowing further time for the families, ‘ongoing multidisciplinary discussions’ and the involvement of religious leaders resulted in at least some parents accepting the recommended course of action.⁷⁸ This could be a result of knowledge gained by parents over the course of discussions with HCPs; equally, it could be a form of attrition in which one party accedes in the light of the other party's greater resources and power. With the backstop of court, and the understanding that most (but not all) judicial determinations find for HCPs in cases involving withdrawal of treatment, it is reasonable to see how this sense of mediation being a ‘friendlier’ form of getting the parents on side, through persuasion rather than judicial order, can arise.

A related concern is that HCPs are likely to have a stronger power base from which to negotiate in mediation. For example, they will have expert knowledge of the patient's condition and access to medical files, and they are likely to have the legal and financial backing of the NHS Trust and the authority to decide whether particular medical treatment is available. We know that deference to medical authority is prevalent⁷⁹ generally in healthcare law and so it plausibly operates in a similar way in mediation too. Conversely, parents and family members may feel overcome with the emotional burden of the situation, placing them at a disadvantage in a mediation. Therefore any incorporation of mediation on the basis that it may achieve TJ must be balanced against the risks of undermining the patient's ultimate wellbeing and/or creating a significant power imbalance in favour of HCPs. One way of overcoming these challenges could be to require mandatory reporting of mediated agreements. As we note in the final section of this paper, mediation is shrouded in secrecy, and even court judgments do not always accurately reflect when and if a case has been mediated. Mediation may secure participant wellbeing, particularly in cases where parties value taking an active role in the decision-making process, but empirical analysis of medical mediation is required to identify whether participants experience wellbeing benefits and, if so, what types of wellbeing benefits they are.

(b) Less adversarial

A key ingredient of the TJ concept is a less adversarial approach, which looks to resolving the problems for the participants. Adversarial systems are centred on opposition – two, or more, parties in dispute seek to ensure justice through an assessment of their opposing positions and evidence. The concept of TJ emerged as an attempt to reform the US criminal justice system, in light of the realisation that many defendants had underlying psychological, social and mental health problems, whereby the adversarial approach did not effectively secure positive outcomes for them or society. As long as these problems were not addressed, it was argued, there was a risk of reoffending. Problem-solving courts therefore developed along a model of rehabilitation, rather than retribution,⁸⁰ moving away from an adversarial approach.

What a ‘less adversarial’ approach looks like in practice can be seen where TJ has been applied, for example in problem-solving courts where the focus is distinctly on co-operation rather than assessing one argument against another to reach legal truths. Shapira has argued in relation to medical cases that ‘The adversarial representation creates an atmosphere of distrust between the patient and the medical staff and as a result the patient refuses to cooperate and does not receive the necessary treatment.’⁸¹ This can also impact HCPs’ and family members’ experience of the other in future encounters in different contexts. Adversarial proceedings can be stressful for participants, including for the party who leaves the court victorious, negatively affecting their wellbeing. This can be particularly harmful if the parties will need to remain in contact and collaborate in the future, for example parents following a separation or family and HCPs in case of a patient with long-term medical needs.

In contrast, the epistemological standpoint underpinning adversarial approaches is that they are more likely to secure truth. This creates a conflict between the common law and TJ values. In crude terms, this can be seen in the former prioritising truth and the latter prioritising participant experience. It is true that critical theorists and others have challenged whether the adversarial model is an effective way of discovering truth,⁸² and there is a wider debate over the extent to which an adversarial judicial system can, in fact, secure substantive justice, with critical scholars long having questioned the assumptions underpinning our legal system.⁸³ An enduring criticism of mediation, on the other hand, is that, as a non-adversarial process without judicial oversight, it permits participants to agree any outcome, even outcomes which do not secure the patient's ‘best interests’. In this criticism, mediation risks undermining substantive justice⁸⁴ by permitting parties to agree what is in their own interests, which in some instances may undermine substantive conceptions of justice.

Relatedly, it has been argued that concerns about mediation's ability to secure the best interests of adults in the CoP represents a positivistic understanding of justice, and is ‘fundamentally a concern about the ability of mediation to be a neutral mechanism for resolving disputes and the extent to which laypeople (participants and mediators) can be trusted to reach substantively fair agreements without judicial oversight’.⁸⁵ While this concern about mediation abounds in the literature, it is balanced against the possibility for participants in any mediation to refer the matter for judicial resolution if, for example, the agreement reached is perceived by any party not to be in the patient's best interests, which is the substantive justice question at the heart of these disputes.

In any event, in the family court and CoP, there is an argument that they already adopt a more inquisitorial approach in trying to get to the truth of what is in the best interests of the patient. The extent to which this is the approach in practice is questionable generally,⁸⁶ but also specifically in relation to medical treatment disputes, which are, we have seen, characterised by high conflict and a sense of opposition between HCPs, patients and family members.⁸⁷ In contrast, if mediation can provide a less adversarial approach, this might enable the participants to experience the dispute more therapeutically than litigation. Evidence of problem-solving and solution-oriented approaches in mediation is likely to be more TJ-compliant than adversarial design.

(c) Collaborative

One of the key features of TJ, at least as it has been articulated in practice through the predominant model of mental health courts, is its flatter hierarchy and more collaborative approach at a systems level.⁸⁸ This does not remove the judicial role where it is present (although it is not present in the mediation context) but ensures that before any judicial order is imposed a collaborative model is utilised. While systems modelled on TJ do have some hierarchy in that the judge can impose a decision, they also encourage collaboration across all participants and focus on problem-solving. A collaborative approach would include working with all parties to identify which support services are required, or which expertise is necessary at which point in the process. This is an important feature of inclusive court and tribunal design, even for those which do not explicitly adopt a TJ perspective.⁸⁹

Justice systems which are more collaborative tend to aim for greater equality of arms between participants. This is important from a TJ perspective because an imbalance of power, knowledge or resources can lead to a sense of unfairness and anti-therapeutic impacts. Collaboration works in TJ by valuing each participant's expertise and making everyone a problem-solver and a ‘co-designer’ of solutions. TJ requires those involved to work together towards the aim of improving the experience. If professionals and lay parties work together to respond in ways to improve participant wellbeing, then the environment must be established to enable free and frank discussion without hierarchical imposition.

We therefore argue that evidence of a collaborative approach within mediation is likely to indicate TJ. The nature of mediation itself, focusing as it does on consensual agreements rather than imposed decisions, is intended to be collaborative. Collaboration might not always be therapeutic; the same concerns about self-determination discussed previously apply, whereby some parties might experience anti-therapeutic effects of having to work collaboratively with parties with whom they are in fundamental and acrimonious disagreement. However, that risk is potentially ameliorated by the flexibility and responsiveness of the mediation process. Features of collaboration to explore through empirical analysis may include: no hierarchy of participants; inclusion and discussion of varied forms of expertise; facilitation of creative problem-solving; and equality of arms between participants (ie if the family do not have lawyers then it would be inappropriate for the HCPs to attend mediation with lawyers present).

(d) Flexible

A flexible process and outcome means being responsive to the needs of the participants and the particulars of the case, rather than rigidly adhering to pre-formulated rules or processes. As has been written in relation to the role of flexibility in procedural justice theory:⁹⁰

Flexibility enables procedures to be adaptable to the different needs that arise and to change in ways that are necessary for the effective administration of justice, which is context specific, contingent and responsive to individual need.

Looking to the development of TJ theory, judges try to adapt to the situation of each defendant in order to obtain the best outcome.⁹¹ Unlike court proceedings, though, a mediation process is not constrained by the same types of procedural rules, and so flexibility can be embedded from the start, even in the choice of mediator. Pre-discussions between parties and mediator take place at a time and place (often online) that suits the parties, and mediators ideally will adopt flexible techniques to ensure that the views and wishes of the person at the centre of the dispute are heard. This is particularly important when the person is an unwell patient and unable to attend mediation. The way the mediation is conducted should be adapted to the needs of the participants: not only the place, time and duration of the mediation, but the mode (online, or in person) and pace (one day, over a series of meetings, over an extended period of time).

Flexibility is inherent in decisions parties make about who is in the mediation room and whether parties meet jointly, or meet with the mediator separately, or a combination of both. Parties are also free to develop a solution they can agree on, indicating flexibility in outcome as well as process for mediation. Outcomes can be original and creative, not bound by precedent.⁹² Although flexibility may not always in itself confer therapeutic benefits on the parties (see, for example, the discussion above about differing views on whether self-determination is always therapeutic), the responsiveness to parties’ needs and circumstances that flexibility affords is likely to be therapeutic. One of the drawbacks of this is the possibility for agreements to be reached which are not in the patient's best interests. However, we have discussed above, and return to in our analysis of the reported judgments below, ways in which this concern may be ameliorated.

(e) Voluntary

The founders of TJ theory believed that choice has a ‘therapeutic value’⁹³ and have argued that research in psychology suggests that when the defendant experiences the choice to accept drug treatment as a ‘voluntary and non-coerced choice’,⁹⁴ the chances of the treatment's success increase. This is backed by decades of literature which confirms that people are more willing to accept decisions that they feel they have had a say in.⁹⁵ However, ‘voluntary’ is different from ‘participatory’, to the extent that voluntariness is related to the participant's choice to enter into the process at all and, furthermore, to continue to engage with it - in other words, a key feature of TJ is the freedom to withdraw at any point.

One of the central criticisms of TJ has been its potential for coercion, particularly from critical disability perspectives.⁹⁶ Specifically, the critique of TJ is that it engages in a form of ‘mission drift’, whereby TJ processes move beyond their original aims, or have unclear aims, and end up engaging more widely in individual lives. For example, professionals may be given powers to make decisions about a person's life or share information in the name of collaboration without the subject's consent.⁹⁷ This critique is a particular feature of the history of how TJ developed, stemming primarily from the use of mental health courts as a better way to deal with criminal justice problems which are, by their very nature, not voluntarily engaged with. However, such a criticism is less applicable beyond the criminal law. For example, in medical treatment disputes, the state already has legal jurisdiction to intervene in the best interests of the incapacitated adult or child patient, and developing a TJ response to these disputes is unlikely to lead to greater state control; it could, in contrast, lead to greater party involvement.

We argue that any application of mediation for medical treatment disputes should be non-coercive, unlike analysis of other TJ models which have been criticised by Arstein-Kerslake and Black.⁹⁸ Applying our TJ analysis to disabled people in the context of medical treatment disputes may also help to address this particular criticism of the way in which TJ justice systems have developed elsewhere. It is clear, however, that voluntariness is in conflict with any movement towards mandatory mediation.⁹⁹ While that does not mean that mandatory mediation could not be a therapeutically just approach overall, it would have to be counterbalanced by other features of mediation which may secure TJ.

(f) Participatory

A process which is participatory includes ‘individuals … having the right, the means, the space, the opportunity and, where necessary, the support to freely express their views, to be heard and to contribute to decision making on matters affecting them’.¹⁰⁰ According to Ronner, ‘If the litigant feels that the tribunal has genuinely listened to, heard, and taken seriously the litigant's story, the litigant feels a sense of validation.’¹⁰¹ The benefits of participation are well established, with being heard and participating in making decisions being repeatedly shown to have a positive impact on the acceptance of the decision made as well as positive impacts for compliance with the decision.¹⁰² If parties in mediation have some degree of involvement in the process, this can be evidence of a participatory approach.

Traditionally, TJ focuses on whether the justice system is therapeutic for its subject and so must be designed in a participatory way for the subject of the case. In criminal cases that is the person on trial, and in mental health courts it is the individual with the mental health difficulty. Therefore, in courts where TJ theory has taken hold, the subject is easy to identify, notwithstanding a persuasive critique from a victims’ rights perspective that a victim's position is ignored in TJ. In medical treatment disputes, however, the patient is not often considered at the centre of the process and rarely participates directly.¹⁰³ This is partly because they are often very unwell or young, and as a result they have limited participation in these disputes. If the patient is not present in the mediation, then they cannot directly benefit from therapeutic impacts of the process, such as choosing to take part, being heard, and collaborating on decision making. In the context of incapacitated adults and children, there could be a tendency to focus on the therapeutic benefits to others involved, including family members and HCPs, rather than the individual herself. While there is value in considering the experiences of all participants, it is also important to analyse the TJ elements of the use of mediation for the patient, even where they are unable to participate directly. One way of addressing this challenge might be to ensure that the patient's representatives, either the child and family court advisory service (Cafcass) or the Official Solicitor for adults, participates in the mediation process to ensure the child's wellbeing is considered. However, we do not know the extent to which these organisations are involved in mediation, particularly mediations which take place at an early stage pre-proceedings, highlighting a gap in our knowledge base.

What a participatory approach would look like in individual cases may vary, including depending on the needs of the participants. For example, in the case of medical treatment disputes, securing the participation of the patient would be age and disability specific, as well as requiring flexibility. In fact, Article 13 of the Convention on the Rights of Person with Disabilities (CRPD) requires states to provide ‘procedural and age-appropriate accommodations, in order to facilitate their effective role as direct and indirect participants … in all legal proceedings’. Participation can be direct or indirect, with indirect methods including advocacy, letters and video, and even drawings that participants make to convey views and preferences. Participation need not require decision-making capacity, and one important task for any therapeutic justice or mediation practitioner is to ensure that a participant who may have support needs is supported to participate in the way they choose. A participatory approach in mediation is essential to securing TJ, but it must be linked with voluntary participation, in which the participant chooses to give voice and feels heard, in order to have positive wellbeing effects and reduce anti-therapeutic consequences.

3. Resolving medical treatment disputes therapeutically: examples from the courts

In the next section we consider two reported judgments and apply our above analysis of TJ and mediation. We have selected reported cases due to the lack of publicly available data on mediations. Mediation is typically confidential and, despite calls for greater transparency, there is almost no publicly available data about real mediations in the healthcare context.¹⁰⁴ These cases were identified following a case law analysis of reported judgments on BAILII and Westlaw relating to healthcare disputes in England and Wales from January 2007 to January 2023, falling under the MCA 2005 or children in the Family Court.¹⁰⁵ GOSH v MX & FX & X (Re X) ¹⁰⁶ is selected as it is one of only eight reported cases where it is possible to identify that the case was mediated before the court judgment, and one of only two where both mediation and a CEC are mentioned.¹⁰⁷ A NHS Trust v G & Others ¹⁰⁸ is selected as it provides a paradigmatic example of protracted conflict in healthcare disputes in the CoP.¹⁰⁹

(a) Resolving conflict in a case concerning a critically ill child

Re X is a rare medical treatment case which was decided, at least partially, in favour of the parents. The case concerned a 9-year-old child with haemolytic uraemic syndrome affecting kidney function, renal disease, chronic lung disease, and intestinal failure. The Trust made an application to the court to withhold interventions and provide only palliative care and to require that there would be no readmission to a paediatric intensive care unit (PICU) for intensive care support. The child's parents opposed the application initially, although following mediation they agreed that it was not in the child's best interests to be provided with a range of interventions, but the dispute remained regarding the provision of oxygen and admission to the PICU.

The judgment indicates that mediation took place over two separate dates while proceedings were adjourned for a short period. One concern with this use of mediation might be delay which, in itself, risks undermining the patient's best interests, particularly where the mediation has not been successful in resolving the dispute or the agreement is one which subsequently breaks down.¹¹⁰ This was also considered in Newcastle Upon Tyne Hospitals NHS Foundation Trust v H [2022] EWFC 14, a case in which the judge praised the use of mediation, which took place over 8 weeks, but only where it was in the interests of the child ‘taking care to identify the point where they might diverge from the timescales of the parents or family’.¹¹¹ However, delay appeared not to be a concern in Re X, which is a case where mediation is discussed in positive terms. The Court ‘had to insist that mediation was imperative, both to assist in narrowing the issues to be decided, and to reduce the levels of distress for this family’.¹¹² More specifically, there was evidence of a participatory approach as the mediation was attended by X's treating clinicians and her parents and ‘the meetings allowed for an exchange of information about X's condition and her parents were able to voice their concerns and although no agreement was reached it paved the way for [the parents] to read and listen to the evidence of the two independent experts in paediatric intensive care …’¹¹³ and ultimately to agree that elective surgery was not in the patient's best interests.

The background to the case is important because it involved not only mediation but also a CEC, which had not involved the parents, something described as ‘unacceptable’ by Russell J.¹¹⁴ If parents are not informed or asked to provide input into pre-court processes such as CECs, then that potentially does not prioritise their wellbeing as participants in the process. They may feel excluded or deceived about what is taking place, which would be distinctly anti-therapeutic. It is, however, not clear to what extent the child's own perspective was considered here, particularly as she was 9 years old so may have been able to participate indirectly. Of course, one of the consequences of X's illness was a severe brain injury and it is possibly for this reason that X's views were not explicitly considered in the best interests analysis. However, it can be seen from the judgment that the court looked at her ability to interact and communicate and analysed her capacity to experience pleasure and benefit from the company of her family; this discussion itself indicates a focus on X's own wellbeing.

On the limited information available, there appear to be several features of TJ present in relation to the mediation process here. First, the fact the judge specifically mentions the parents voicing their concerns at mediation suggests this was an important outlet for enabling the parents to move forward in a more positive way for their own wellbeing. The dispute process appeared to facilitate improvement in the parents’ wellbeing because their view was given some weight and, furthermore, their child was given more comfort and some chance at a form of recovery, the latter potentially an indicator of an improvement in the patient's wellbeing too. Secondly, the outcome of the mediation highlights its flexibility and suggests a collaborative approach may have been used. This is a rare judgment where the court departed, partially, from the views of the Trust and HCPs and came up with a more flexible, and arguably creative, approach, with declarations being made regarding the way forward for X but also agreeing with the parents that X should be admitted to PICU and provided with oxygen if necessary. This indicates a collaborative approach by the parties as they were able to voluntarily agree to certain aspects of care being withdrawn, which we must also assume did not undermine X's best interests as this was not overturned by the court. Re X is noteworthy as it is one of the only cases where mediation is mentioned in a reported judgment and there is a judgment partially in favour of the parents. There are several features of TJ present in the way that mediation is presented in the judgment, albeit further evidence is required to make any stronger claims. Discussion of mediation in judgments is essential if we are to have more transparent understanding of mediation's use, albeit that can lead to skewed data because court cases are more likely to occur where mediation has not reached full agreement.

(b) Resolving conflict in an ongoing medical and residence dispute for an adult

Mediation is rarely reported in cases that arise under the MCA 2005 and, where they are, very little detail is present. Therefore, we have not been able to identify any CoP case with sufficient detail in its judgment about mediation's use. Instead, we consider a case with a typology of sustained conflict and litigation between HCPs and family members and apply our TJ analysis to consider how mediation might meet TJ aims. In A NHS Trust v G & Others,¹¹⁵ a disagreement had arisen between the family and the clinicians concerning the treatment and residence of G, a 27-year-old woman suffering from a progressive and untreatable neurological disorder, as well as other medical problems. G had microcephaly and her development was significantly delayed since early childhood. She had quite exceptionally been hospitalised in children's hospital since the age of 13. She had had a tracheostomy and received ventilation support but could leave the hospital for hours to spend time with her family. Her parents were devoted to her, and her father, FH, was very closely involved in her everyday care – his understanding of medical issues was qualified as ‘impressive’ by one of the expert witnesses.¹¹⁶

The proceedings were lengthy, meaning that for several years there was no clear treatment plan for G. At the same time, it was recognised that whereas G's emotional needs would be better served if she were reunited with her family, such an arrangement would not be optimal for her physical welfare. In the end, the court's decision was in line with the NHS Trust's application: the central venous line was to be removed and G was to move to a care home, and her return to the family home should be considered in the future. One of the preconditions set by the court was that the strained relationship between FH and the treating clinicians should become ‘fully functional’.¹¹⁷ Unfortunately, this relationship deteriorated in the following weeks and months, becoming ‘entirely dysfunctional’, characterised by mutual distrust and poor communication.¹¹⁸ FH was also said to have become hostile and intimidating with nurses caring for G. The Trust considered it necessary to seek injunctive relief against FH, his mother N and G's mother M, which was granted. Whereas both the parents and the treating team agreed the paediatric ward was not the right place for her, G was still there six months after the first judgment had been given.

The adversarial nature of these proceedings arguably did not help to improve the already difficult relationship between the family and HCPs. Moreover, the judge chose and imposed the solution advanced by the Trust. While mediation was not attempted, as far as we know, we consider whether it might be an example of the typology of dispute where mediation could have TJ benefits. One of the drawbacks of mediating a case such as this is that it would have been difficult to secure the patient's direct participation in the process given her ill health in hospital and the clear ongoing conflict between the parties. Yet the flexible nature of mediation could have provided an opportunity for a therapeutic approach by encouraging creative ways to engage a party with impaired mental capacity situated in a hospital ward. For example, she could have participated indirectly through the mediator speaking to her as part of pre-mediation meetings, which are common in facilitative mediation.¹¹⁹ This would have enabled the mediator to gain an understanding of G's views and relay these to other participants. If G had an independent advocate she could also have been invited to participate in part of the mediation to speak on G's behalf. Moreover, mediators can encourage participants to discuss the patient at regular intervals during the mediation, reminding them that it is the patient's best interests that they are focusing on jointly in mediation. Similarly, participants might bring along photos, letters or videos that describe the patient's views or that show her interactions with those she is close to.

Another feature of TJ is participant wellbeing. Here, wellbeing improvements may have been strongly related to the ability to participate voluntarily and have one's emotions considered sensitively. The voluntary nature of mediation means that the parties would not have had a solution imposed upon them by the court. In a mediation process, FH's emotions and their impact could have been acknowledged with respect and dignity and not ‘used against him’ as they appeared to have been in the litigation where Hayden J extensively discussed the father's emotions, mostly to dismiss his concerns. Moreover, the confidentiality of the mediation discussion could have contributed to safeguarding the parents’ wellbeing, de-escalating conflict and rebuilding the relationship with the treating team, as their concerns could be aired in a confidential process without risk of disclosure to court. The ongoing conflict clearly had a negative impact on G's wellbeing, prolonging her stay in the inappropriate children's hospital. In addition, very likely she was aware of the strong negative emotions felt by her father and others responsible for her care. The everyday presence of G's parents, crucial to her emotional wellbeing, was potentially limited by the injunctive relief; that measure was for the benefit of the HCPs but arguably was of no benefit to G, and even possibly had an anti-therapeutic impact on her. All these elements were acknowledged by the court.

Finally, G herself may have gained a therapeutic benefit from her case being mediated. For example, if mediation enabled her family members and the HCPs to develop better communication or an improved relationship, this would likely have had wellbeing benefits for G. We cannot say this would have occurred in this specific case, but these benefits have been established in literature elsewhere, which emphasises the creativity and flexibility of mediation agreements, as well as the potential for improved communication and relationships.¹²⁰ However, there remains the risk that mediation in this typology of dispute prolongs decision making, with delay undermining the patient's best interests. Furthermore, HCPs may bend to the strong emotions of family members and agree outcomes which are not optimally in G's best interests. As we have noted in our earlier discussion of mediation, there are some safeguards against these concerns, specifically that either party would retain the ability to apply to court if they believed agreement was not in G's best interests. Furthermore, a mediator experienced in participatory approaches should ensure the range of perspectives is heard as part of the mediation process, albeit this relies on a mediator having the necessary skills and expertise to secure this.

Conclusion

The conflict that can arise in medical treatment disputes can be challenging for all participants and can be made worse through litigation. This paper has highlighted examples of this conflict and discussed possible causes. Yet there has been relatively little focus on the use of mediation as a tool for resolving medical disputes. Moreover, there has been no consideration of mediation's potential to secure certain features of TJ in medical treatment disputes. There are, of course, significant challenges in using mediation here, including securing the best interests of the patient, avoiding ‘mission drift’ and ensuring that a participatory approach, including for the patient, is facilitated. However, by considering the synergies between TJ and mediation, we have provided a framework against which mediation of medical treatment disputes can be analysed for its therapeutic potential. Further empirical analysis of mediation's ability to achieve TJ is necessary, given the limitations of relying on reported judgments and the lack of evidence on mediation in practice. If mediation is to provide a therapeutic way forward for dealing with medical treatment disputes, the challenges we have identified will need to be addressed and mediation will need to incorporate practices that support therapeutic benefits for all participants.