Trust or Contract: How Far Does the Contemporary Doctor-Patient Relationship Protect and Promote Autonomy?

Chapter 5
Trust or Contract: How Far Does the Contemporary Doctor-Patient Relationship Protect and Promote Autonomy?

Sheila McLean published her doctoral thesis in 19891 and it is now trite to say that its influence on medical practice in general was profound. The title – The Patient’s Right to Know – clearly signalled that her particular interests lay in an analysis of the doctor-patient relationship within the therapeutic ambience which, as we will see, has failed to mature into an alliance, despite the increasing political and judicial attention devoted to the concept of patient autonomy.² McLean’s ground-breaking contribution came during what can be seen as the developmental stages of a United Kingdom medical jurisprudence. It is a matter of history that, around the 1980s, the medical profession was under intense and critical review. This was partly self-induced by way of a number of scandals related to indefensible examples of malpractice.³ The major impetus was, however, derived from a growing, and increasingly accepted, belief in individualism in general and the parallel, specifically medical, demand for self-determination based loosely on the patient’s right to protection from untoward interference with his or her person – which is, effectively, a matter of assault.

However, the practice of medicine is by its very nature, based on personal interference. Thus, the distinction between a good and a bad doctor-patient relationship is defined – or refined – in the word ‘untoward’. No-one would suggest that Dr Cameron of Tannochbrae4 intended to harm his patients; his offence under the revisionist mantra was that of paternalism. Most doctors practising at a time when their pharmacopoeia was largely both homeopathic and herbal would have seen this as synonymous with the reputable principle of beneficence. Nevertheless, the medical profession fell foul of the public when it persisted in this stance in an era in which pharmacologically powerful drugs, and associated complex techniques, became freely available. Very rapidly then, ‘untoward’ treatment came to mean unconsidered treatment – that is to say, treatment which was given without, at the same time, giving the patient adequate opportunity to accept it or reject it. Thus was born the procedure-driven concept of informed consent which now constitutes the main pillar of medical jurisprudence in, at least, the greater part of the anglophone world – and this despite, in respect of the United Kingdom, the early, yet seemingly authoritative, words of Dunn LJ that: ‘The concept of informed consent forms no part of English law’.⁵ The unfortunate result of these terminological uncertainties has been that the right to informed choice of treatment and respect for autonomy have come to be regarded as synonymous. In our view, this is a misconception with worrying conceptual and practical consequences. McLean’s understanding of consent could not have anticipated this collapse in the distinction between two related but distinguishable concepts which she attributes, broadly, to a differing jurisprudential interpretation of the meaning and purpose of ‘informed consent’ in the United States and the United Kingdom.⁶ Another important distinction that McLean is keen for all to grasp is that between due and appropriate concern for a person and unwarranted paternalism: ‘… care must be taken to avoid opportunities for [the latter] to resurface where it interferes with autonomy’.⁷ The line for McLean can be drawn thus: ‘While there may be a role for beneficence … this is not the same as a justification for paternalism’.⁸

This chapter walks the fine lines between autonomy and consent, and beneficence and paternalism, as these have been drawn in the last 30 years, and against McLean’s own analysis. We suggest that a failure to maintain sufficiently clear and bright lines, principally on the part of the courts, has led to a distortion of the doctor-patient relationship. Crucially, we contend that, today, this is a relationship of three parties – clinician, patient and the courts – in that the last police the parameters of the relationship dynamic and have, increasingly, taken it upon themselves to interfere in the name of autonomy in ways that might at times serve to undermine the all-important value of trust. While the medical profession in the United Kingdom has made considerable efforts to redress any imbalances in power – as demonstrated by the General Medical Council’s guidance on Tomorrow’s Doctors with its emphasis on effective communication and joint decision-making⁹ – it is far less clear that the same objective is being delivered by the law, especially by the way in which central notions of consent and autonomy are constructed and protected.10

Autonomy is a status whereby the human being is recognised as having the fundamental right, qua human, of control over his or her own body and the direction of his or her own life in both physical and intellectual senses. It is a quality, which – when present – indisputably merits respect. It is related to human dignity¹¹ which is an inherent quality, possessed of all persons and which exists irrespective of their ability to express it. Contrariwise, autonomy is contingent both on one’s ability to recognise and to exercise it – and yet it is of no less fundamental importance for this fact. Consent, on the other hand, is a means to an end, that end being respect for the person and respect for their autonomy. It is an ethical and legal device or conduit by which we can achieve the desired end-point, but it is by no means the only way to do so. And yet, such is the sway which consent now holds over the discipline of medical law that this important distinction is often ignored. This error is compounded by the fact that it is consent rather than autonomy which emerges as the dominant paradigm of the two within the doctor-patient relationship – that is, that the means often comes to define the totality of the ends. Not only is this paradoxical and conceptually worrying, but it has profound practical consequences; once again, ones that we fancy were never intended or looked for by McLean when she wrote her seminal work.

From this prologue about conceptual confusion, the purpose of this chapter emerges as two-fold. First, we will explore what the collapse in the distinction between these central concepts has meant for medical law, arguing that it has impoverished both our understanding, and the very nature, of the doctor-patient relationship. This is true also with respect to the legitimate roles of those who play a part in the formation of the relationship – being both medical and judicial personnel. Second, we will argue for the reinstatement of the distinction between the roles of autonomy and consent, and for further clarification of the legitimate basis for beneficent as opposed to paternalistic interference, in order to reorient the way in which the relationship should be seen. These contributions depend on two different sets of circumstances: (i) when autonomy is present and accepted in the therapeutic relationship, and (ii) when it is disputed or open to question.

As to the former, we will demonstrate that the focus on consent to date has reduced the doctor-patient relationship to one that is driven by claims, rights and entitlements, and by concerns about breaches of duty and the pursuit of remedies; it has overlooked the importance of, and necessity for, trust as the basis for that relationship. We posit that the drift towards a pseudo-contractual, legalistic model should be resisted for the benefit of a genuine therapeutic alliance and for the betterment of medical law itself. As to the latter context, we will examine recent developments concerning the emerging legal category of the ‘vulnerable adult’, as driven by the English High Court and Court of Appeal. While this has been done in the name of facilitating autonomy, we argue that far more procedural attention must be paid to the limits of court powers over this category of person, lest the developments bleed into the medical domain and return clinicians to the role of paternalist.

We have already pointed to the semantic confusion that currently exists as to the distinction between autonomy and consent and we re-emphasise that autonomy is a multifaceted quality of humanness, the protection of which is now accepted as the prime requirement of contemporary medical jurisprudence. Consent is not a complete reflection of that need. Fundamentally, the protection of a patient’s autonomy as defined above lies in the criminal law of assault;12 on this view, consent is no more than a device to protect the patient’s autonomous right to accept or refuse the ‘assaults’ which are an integral aspect of modern medical practice. Thus, while it is undeniable that an intimate connection exists between consent and autonomy, the former represents no more than a secondary, or consequential, manifestation of the fundamental right. The two concepts are inseparable but they are not synonymous. In this section we explore some of the contours and the limits of consent as it is currently used in medical law, in both the therapeutic and research contexts. In particular, we posit that, whereas autonomy serves to promote the interests of the patient in general terms, the function of consent is limited to protection of his or her status against practices that drive the development of an unequal relationship.

Consent as an Aspect of Self-defence

Whether or not it is a derivative attribute only, the patient’s practical defence of his or her autonomy lies in the primacy of choice as to medical interference – be it in the form of consent or refusal. Yet, it is one of the deep ironies about the groundswell of support for consent in the guise of patient empowerment that the reality is that the power that the patient derives is almost entirely negative: it is the right to say no. It is all too often forgotten that the doctor-patient relationship is one between supposedly autonomous individuals, and the courts have consistently ruled that the patient cannot demand treatment against the doctor’s judgment.¹³ From this, it follows that adoption of the mantra of consent is, in effect, an abrogation of autonomy. Moreover, refusal carries with it a rejection of opportunities proffered by the medical professionals, leading to a closing off of options and a restriction of possible life paths. This might indeed be what the patient chooses, but the scope of possible choice is limited to a binary yes/no and the so-called choice is reduced to the single option of rejection. We do not deny that such a choice can be liberating, but this is so only in the sense of a negative liberty: the right to respect for the integrity of the human persona. For us, then, one of the major and oft-heralded victories in the campaign for patient autonomy – reflected in the following case law dictum – has something of a hollow ring to it: ‘This right of choice is not limited to decisions which others might regard as sensible. It exists notwithstanding that the reasons for making the choice are rational, irrational, unknown or even non-existent’.¹⁴

In the medical ambience, where a patient is faced with a choice between two therapeutic management proposals, the ability – or entitlement – to choose between the two is scarcely empowering unless it is assisted by the doctor as therapeutic partner. Yet, too often the patient is left alone in the name of autonomy to choose between options, the doctor having fulfilled his or her duty of establishing that the patient is in possession of the relevant facts and has capacity. Consent, if it is given, thus becomes a protector of the doctor rather than of the patient.15 It is acquiescence with the proposed medical intervention, but it is scarcely patient-empowering beyond such limits. The enduring power remains one to say no – at any stage and for any reason¹⁶ – but this may be of little assistance to the patient, or research participant, whose interests might be better served by a therapeutic or research alliance where the boundaries and contours of a relationship are negotiated on mutual terms. We suggest that patients and doctors alike are currently prevented from entering such an alliance by the constraints of consent as it has come to be constructed by these two key qualifying elements: capacity and informedness.

Capacity

Capacity is a factor that is internal to the patient, albeit that it is mostly judged by medical professionals. In law, it is a threshold device that is used to determine whether further (caring) intervention can legitimately be taken towards individuals who purport to say ‘no’ or who cannot speak for themselves. The default, as we know well, is that capacity is present in all mature adult persons. As such, there is no incentive to promote autonomy. Indeed, quite the opposite is true. Only if there is incapacity can an individual be treated non-consensually in their own best interests.¹⁷

McLean has revisited the subject as a whole in her recent monograph.¹⁸ Very early in her presentation, she describes mental competence as the sine qua non of autonomy.¹⁹ Certainly, many of the legal battles fought and won on the consent field over the last quarter century have been in and around questions of capacity. Case law²⁰ and legislation²¹ now abound with attempts to define legal capacity and its role as a threshold mechanism for determining whether someone is sufficiently autonomous with respect to the particular decision they are facing to merit respect for any choices they purport to make.

We confess to being unimpressed with the majority of these templates which, of necessity, deal with the generality rather than the particular.²² They do little more than recapitulate what most doctors would see as standard medical practice; as a result, the legal definition of capacity rests on the highly subjective opinions of individual practitioners. The answer to the broad question – has this adult person still got capacity? – is largely medically determined, but the legal consequence of a finding of capacity necessitates the unquestioning acceptance of any and all choices which are then made. As things stand, setting capacity as a legal threshold focuses all of our attention on the establishment of capacity and none on what happens thereafter. Indeed, there is a positive disincentive in law to explore with a patient how he or she might maximise their autonomy, through choices or otherwise. Once the Rubicon has been crossed, the doctor is not permitted to accompany the patient on his or her journey. Put bluntly, the law supports the potential abandonment of autonomous patients to the vagaries of their own choices. This is at considerable odds with what many patients would reasonably expect of their healthcare professionals. Moreover, it fails to reflect the possibility that we are all, in some sense, vulnerable and ‘inevitably dependent’ throughout our lives, and most especially when we are unwell.23

Informedness

Achieving autonomy does not only depend on the intrinsic quality of capacity. It can depend as much on the externally-driven provision of information about the nature and quality of choices that are to be made. Once again, this apparently leaves considerable control in the hands of the medical profession, and yet – despite the triumph of the clarion call for the patient’s right to know – we would suggest that the legal result does little to promote either autonomy or a healthy therapeutic relationship.

Capacity and informedness must co-exist to create the autonomous patient. But this raises the question: how and how far does a patient have to be informed about a proposed medical intervention before his or her consent is valid? McLean published The Right to Know in the immediate aftermath of the House of Lords’ decision in Sidaway.²⁴ This decision is remarkable in two important respects: first, it confirmed the application of the Bolam²⁵ principle to information disclosure cases in the UK and stated explicitly that the American doctrine of ‘informed consent’ had no place in our law. Second, there was no mention in the entirety of the judgement of the term autonomy. Such features were the strongest possible evidence to activists like McLean that medical paternalism was alive and well and that the battle for the patient’s right to know remained to be fought. And fight they did. And win they did. Indeed, so far has the pendulum swung from Sidaway that the decision must now be considered as little other than anathema. In its place sit a wide range of purportedly patient-centred judgements,²⁶ best typified by the Supreme Court ruling in Chester v. Afshar²⁷ which, again with deep irony, had astonishingly similar facts to those of Mrs Sidaway and yet which produced an equally remarkable – and diametrically opposite – result. What is particularly remarkable about Chester is not only its blatant distortion of the rules of causation so as to enable the Court to find in favour of the claimant’s argument that she was insufficiently informed, but also that the court appears to have brought us to the brink of a stand alone tort of affront to autonomy.²⁸

We are not concerned here to revisit the legal particularities of informedness in order to demonstrate the ways in which this feature of consent has developed. Rather, we are driven to point out the consequences for autonomy of this fixation on informed consent, albeit not in the US doctrinal sense. First, and most obviously, this conceptualisation implies that a consent that is insufficiently informed is not valid and, a fortiori, is not a genuine expression of autonomy. Yet, as the following examples show, this conclusion cannot be universalised across the whole spectrum of healthcare relationships. Secondly, a requirement of informedness can serve to undermine or even entirely thwart autonomy interests; this is, again, ironic since it is generally considered that consent processes are made procedurally robust precisely because they involve information giving. In particular, what this drive towards informedness to promote consent achieves, indirectly, is a potential undervaluing of circumstances where an otherwise autonomous individual might trust another party to ‘do the right thing’, be it with their being, their tissue or their data. We can choose to trust in circumstances where information is less than complete or, even, where the very provision of information is considered to be an affront in itself.

One of us was inspired by McLean’s work on the Right to Know to ask whether the converse claim was also sustainable, that is: is there a Right Not to Know?29 The notion is perfectly understandable because we can all imagine circumstances in which we might prefer not to be told of a particular diagnosis or family trait if, say, nothing can be done about it. Is it not an equal entitlement under our claim to autonomy to reject information as it is to require it? We wager most of us would intuitively agree that this is, at least, not an unreasonable exercise of our autonomy. Moreover, in doing so we would have to recognise that this is about a lack of information and relies on us being able to trust those in possession of information not to disclose it. We have argued for the subtleties of this elsewhere.³⁰ Our contention that consent (or refusal) cannot do all of the work to protect our autonomy and other personal interests³¹ is illustrated by the limited protection afforded by instruments such as the UNESCO Universal Declaration on the Human Genome and Human Rights (1997), Article 5c of which provides: ‘The right of every individual to decide whether or not to be informed of the results of genetic examination and the resulting consequences should be respected’. Similarly, the Council of Europe Oviedo Convention on Human Rights and Biomedicine (1997), Article 10(2) states: ‘Everyone is entitled to know any information collected about his health. However, the wishes of an individual not be so informed shall be observed’. These provisions are based on two challengeable assumptions: first, that a timely and effectively-informed opportunity to say ‘No’ is available; and, second, that a person can be informed that there is something to know about themselves without revealing the essence of the information itself! If we take seriously the idea that a genuine state of ignorance is valuable to our autonomous selves, then these formulations as matters of choice and control of information are deficient.

How Information Can Undermine Autonomy: (B) Broad Consent and Biobanks

It is clear, however, that the requirement of respect for autonomy goes deeper than that from many aspects. The extreme antithesis to individualistic autonomy can, perhaps, be epitomised in some such phrase as ‘forgone autonomy’ whereby the originator empowers the receptor to proceed on the assumption that specific consent has not been given but is encompassed within a general acquiescence with the circumstances. Such a situation arises when the originator does not – or cannot – know the full use to which the receptor is likely to put the material over which he or she has assumed control; the consent model then becomes difficult to apply as a means of justifying surrender of one’s autonomy insofar as it seems to ignore the significance attached to ‘informedness’ in the concept of ‘informed consent’. The exemplary situation is to be found in contribution to data and tissue research – or biobanks – in which, given that not even the controller of the bank can be certain of the future use of the material in his or her charge, application of the informed consent model is a virtual impossibility.32 This has resulted in the literature and in practice in a flurry of attempts to reconceive and distinguish forms of consent variously as implied,³³ explicit,³⁴ informed,³⁵ specific,³⁶ broad³⁷ and generic.³⁸ But, if as we suggest, there should be a decoupling of consent from autonomy, where lies the alternative? In our view, there is no single answer but, in our search for that answer, we must first consider the precise circumstances in which one is pleading that autonomy rules.

Reconceptualising Autonomy in Practice – Alternative Models

From our above analysis, we suggest that there are two ways in which a patient’s autonomy arises as a contentious issue. The first, which broadly reflects the status quo and which we tentatively describe as defensive autonomy, relates to the autonomous right to choose what is to be done to or with one’s body. It is exemplified, as we have seen, by the respect that is due to our choices as to our medical management and is intimately bound to the consent model on which hangs much of the contemporary definition of autonomy. It is heavily dependent on two key thresholds being overcome: having the capacity to make choices and having adequate levels of information by which to make choices. Attempts to exercise one’s autonomy beyond these limits are treated as suspect, flawed and non-deserving of respect. This having been said, it is precisely in the realm beyond autonomy that there have been considerable statutory legal reforms in the guise of the Mental Capacity Act 2005 and the Adults with Incapacity (Scotland) Act 2000 to determine how incapacitated persons should be treated. There now exist extremely robust procedural mechanisms by which we can place trust that patients will be treated in their own ‘best interests’ (or ‘to their benefit’ in Scotland) when autonomy and consent do not come into play. When they do, however, we suggest that a defensive need to obtain consent drives a culture of potential patient abandonment whereby individuals are left to their own devices to explore and exercise their autonomy. This casts consent as an aspect of control, albeit illusory in many cases. It follows that there is an expectation that control be facilitated.

Arguably, this has been compounded in recent years by the entry of the property model into the doctor-patient relationship.39 Property is the paradigmatic legal device through which to exercise control. We pre-empted this more than a decade ago: ‘The concepts of property and consent need not be mutually exclusive; indeed they should operate in tandem to ensure full and proper respect for individual rights’.⁴⁰ We did not anticipate, however, that an unlooked-for consequence of these parallel developments would be the increasing approximation of the doctor-patient relationship to something akin to pseudo-contract. We suggest that the metaphor of contract is pertinent in that it evokes a relationship characterised by an essentially oppositional dynamic focused on expectations of rights and entitlements, and with a growing cohort of remedies for non-performance and failure to satisfy, viz the apparent emergence of the stand-alone tort of affront to autonomy (above). The patient today, then, is entitled to information, is entitled to unquestioning respect for their refusals, is entitled to recognition of their property interests, and, ultimately, is entitled to be left alone entirely to determine his or her own self rule. This is, however, a view of autonomy in the most meagre of terms. It is a view that is likely to take the care out of the caring professions.

The alternative is to cast autonomy as permissive whereby the individual can legitimately and voluntarily surrender some aspect of his or her persona, as is typified by the originator entering the data- or bio-bank system or the organ transplant programme. On this interpretation, autonomy is no longer being used as a defensive shield but is, rather, being surrendered as an expression of community trust; and, again, consent fits uneasily into the resultant ethical pattern. The power to say no and withdraw usually remains (subject to the impracticalities of destroying data or samples) but, in this concept, there is no pretence of continuing control over said data or samples nor, indeed, of any necessary personal benefit. Indeed, the trusted surrender of autonomy might even bring risk, as in the example of recognising an interest in not knowing information about one’s self: this can be a threat to privacy interests. Still, why should a pastiche of control – via consent – be the gold standard of our behaviour when our autonomy might be facilitated and exercised in ways which have no need for such control and are, instead, a reflection of our capacity and desire to trust? Developments in the law to date seem largely blind to these possibilities.

From Pseudo-contract to Trust

The relationship between trust and law is under-examined. Unlike the connections or disconnections between autonomy and trust explored by O’Neill and others,41