Informing patients: The Bolam legacy

Abstract

In the context of medical advice to patients, the UK decision in Montgomery v. Lanarkshire Health Board rejected the application of Bolam v. Friern Hospital Management Committee. This article argues that the rejection is neither complete nor settled. We explore doctrinal, conceptual and practical limitations of Montgomery to demonstrate the vestiges of Bolam’s relevance to medical advice. Medical advice does not end with disclosure of material risk but incorporates information on prognosis, diagnosis and treatment alternatives. Montgomery does not always apply in these cases, nor outside the medical mainstream or where patients lack capacity to consent. We identify ways in which the extension of patient-centred care in the giving of medical advice can be achieved through incremental development of Montgomery and application of the Bolitho gloss to require that processes conform to Montgomery principles of partnership and autonomy.

Introduction

For some time, the non-disclosure of information to patients in England and Wales was governed by Bolam v. Friern Hospital Management Committee (Bolam), a highly influential first instance decision which set out the standard of care for skilled professionals applied in relation to both treatment/diagnosis and the giving of advice about risks inherent in medical treatment. ¹ In 2015, the UK Supreme Court in Montgomery v. Lanarkshire Health Board (Montgomery) held that Bolam should no longer apply in the context of negligent non-disclosure of information about risks and reasonable treatment alternatives. ² Lords Kerr and Reed, who jointly gave the leading judgment, stated: ‘There is no reason to perpetuate the application of the Bolam test in this context any longer’. ³ In this article, we expose Bolam’s ongoing legacy in relation to aspects of healthcare professionals’ (HCPs’) advisory role. We suggest ways in which legal developments might mitigate issues flowing from the dual application of Bolam and Montgomery in different aspects of HCPs’ advisory role and endorse a patient-centred approach to medical advice that applies beyond negligent non-disclosure of information.

In ‘The Bolam identity’ ⁴ and ‘The Bolam betrayal’, we briefly set out the Bolam decision, its expansion and subsequent attempts to apply judicial controls, particularly in relation to non-disclosure of information in the case of Montgomery. The next sections set out new arguments as to Bolam’s legacy from two perspectives. We begin with a doctrinal examination of Bolam’s current relevance. While Montgomery enhances commitment to patient values, it also (rightly we would argue) recognises the pertinence of clinical expertise, and with it, the relevance of Bolam. In ‘The Bolam retribution’, we argue that the dissection of the duty of care established in Montgomery is neither settled nor clear. We derive three explanations from Montgomery for the respective relevance of Bolam and Montgomery tests and show that they can conflict. The next section considers Bolam’s legacy beyond negligence. In ‘The Montgomery enigma’, we criticise the consumeristic conception of patient choice alluded to in Montgomery. We consider the inherent limitations of Montgomery given its focus on material risk which relates to but one of many aspects of the HCP advisory role, and set out practical challenges in its application. We set out a normative argument for the development of a better appreciation of the role of partnership in both treatment selection and choice between treatments. In the final section – the ‘Montgomery supremacy’ – we consider the potential for Montgomery’s incremental legal development and influence.

Montgomery is a landmark decision and adoption of its principles within and beyond clinical negligence has potential to transform clinical practice. However, while Bolam’s influence has declined, we point to its continued relevance in various aspects of clinical advice, and while Montgomery represents a watershed, there are barriers to the proliferation of its principles beyond clinical negligence. In working to overcome these challenges, we emphasise the importance of balancing patient values with professional expertise and set out ways in which that might be achieved.

The Bolam identity

Mr Bolam agreed to undergo electroconvulsive therapy to treat his mental illness but sued in negligence when he suffered injury. He complained that he should have been given relaxants and restrained to reduce the risk of physical injury, and also that he should have been warned of the risks. McNair J set out the test for assessing the standard of care in negligence cases involving skilled professionals:

[H]e is not guilty of negligence if he has acted in accordance with a practice accepted as proper by a responsible body of medical men skilled in this particular art. ⁵

The result is that, where skilled professionals reach the standard of a responsible body of opinion, they will not be found negligent merely because other skilled professionals would have acted differently. Applying the test in Bolam, expert evidence established that medical opinion was opposed to both relaxants and restraints. Furthermore, it was accepted practice not to warn of small risks unless the patient specifically asked. The jury (as the case was then decided) did not establish negligence in relation to either the treatment or the non-disclosure of information as to risk.

In relation to non-disclosure of risk, the relevance of the Bolam test was reconsidered in Sidaway v. Board of Governors of the Bethlem Royal Hospital (Sidaway). ⁶ Should errors regarding disclosure of information be viewed in the same way as errors in the operating theatre? The judgments in Sidaway varied in approach, ⁷ with Lord Diplock siding firmly with application of the Bolam test, Lord Scarman embracing a more patient-centred approach and Lord Bridge of Harwich (with whom Lord Keith agreed) and Lord Templeman somewhere between. Lord Diplock was concerned that the law relating to negligently caused injury should not be carved up across different aspects of patient care:

This general duty [of care owed by a doctor] is not subject to dissection into a number of component parts to which different criteria of what satisfy the duty of care apply, such as diagnosis, treatment and advice…[S]uch dissection…is neither legally meaningful nor medically practicable. ⁸

The majority judgments have been interpreted ⁹ to require that a patient was entitled to receive only the amount of information that a responsible body of medical opinion considered relevant. The implication was that disclosure of information (or, as Lord Diplock termed, ‘advice’) was considered an aspect of treatment and diagnosis, and thus a matter of clinical judgment. We will see in the next section that this did not prove workable because it did not allow the law to protect patients from medical paternalism. Nonetheless, we will later argue that Lord Diplock’s warning about the practical difficulties in dissecting the duty of care was prescient.

In relation to both treatment/diagnosis and information disclosure, Bolam enhanced certainty by giving HCPs confidence that differing perspectives would not necessarily be viewed in law as evidence of substandard care. From a patient perspective, however, Bolam made it difficult to establish breach and set a tone of deference to the medical profession that applied in both court and practice. In cases where different and discordant expert medical opinions existed, the HCP merely had to conform to one of them. ¹⁰ Bolam also offered a judicial solution to wider matters of medical ethics and complex clinical questions. Swoboda has described ‘The deep ossification of the Bolam test in the common law’. ¹¹ Brazier and Miola refer to a process of ‘Bolamisation’ ¹² whereby the courts abrogated responsibility for ethical issues and lacunae in the law into the hands of doctors. What ought to be done became, by default, what reasonable doctors would ordinarily do. ¹³ In particular, Bolam dominated the test by which a patient’s best interests was determined. This was particularly apparent in cases on sterilisation of patients lacking capacity ¹⁴ and the withdrawal of life-sustaining treatment. ¹⁵ Brazier and Miola also present evidence of a subtler infiltration in Gillick v. West Norfolk and Wisbech Area Health Authority, ¹⁶ which set out the test for child competence, and subsequent cases concerning treatment refusal, where they argue that covert application of Bolam effectively determined the reasonableness of treatment in the child’s best interests. ¹⁷

The extension of Bolam was influenced by the sociocultural context. Decided in 1957, the National Health Service was in its infancy. Social factors such as class differences between most doctors and their patients and costly NHS reforms that limited the potential for patient choice provided the backdrop to the decision. ¹⁸ Evidence of clinical efficacy was sparse or anecdotal in contrast to the era of evidence-based medicine in which we now reside. Doctors set themselves apart in their expertise, and the courts relied on them in complex and uncertain clinical and ethical cases. ¹⁹ Medical regulation was scant and there was little by way of detailed guidance from regulatory bodies. These factors resulted in greater reliance on clinical judgment and information disclosure was dominated by the principle of doing no harm. ²⁰

The Bolam betrayal

The context has since changed dramatically. The National Institute for Clinical Excellence (now the National Institute for Health and Care Excellence (NICE)) was established in 1999 and evidence-based medicine has become the norm. ²¹ Guidance from Royal Colleges and regulatory bodies is increasingly detailed and prolific. Where interaction was once dominated by a doctor–patient relationship, decisions are increasingly made between the patient and a team of HCPs. ²² Change was also driven by developments in human rights, ²³ an information revolution and the rise of consumerism. ²⁴ The social and legal changes have been heralded as moving from patient as ‘passive recipient’ to patient as active participant ²⁵ : from doctor- to patient-centred practice.

Devaney and Holm argue that paternalism and deference once went hand in hand, but legal and policy initiatives ²⁶ emphasising patient choice, have been successful in limiting paternalism. ²⁷ These sociocultural transformations impacted on Bolam’s dominance across different applications. We will focus on the changes with regard to disclosure of information and advice, but it is worth briefly noting constraints on the Bolam test in relation to treatment and diagnosis cases and beyond.

Bolam continues to apply to medical treatment and diagnosis cases, ²⁸ but the House of Lords made efforts to correct what they saw as a misinterpretation of Bolam in Bolitho v. City and Hackney Health Authority. ²⁹ This 1997 decision enhanced the potential for judges to question the logic of medical judgments. Lord Browne-Wilkinson added a gloss to the Bolam test, recognising that ‘if, in a rare case, it can be demonstrated that the professional opinion is not capable of withstanding logical analysis, the judge is entitled to hold that the body of opinion is not reasonable or responsible’. ³⁰ The Bolitho judgment reasserted boundaries: judges must scrutinise the foundation of medical opinion and must be satisfied that it rests on logical foundations, but it only goes so far. The Bolitho gloss will only apply where ‘a judge can be satisfied that the body of expert opinion cannot be logically supported at all’. ³¹

While Bolam applies in cases of misdiagnosis that result in harmful treatment, ³² it has less relevance to pure misdiagnosis cases, where a diagnosis is either right or wrong and does not result in the balancing of risks and benefits in order to determine whether to treat. In Muller v. King’s College Hospital NHS FT, ³³ Mr Muller was misdiagnosed by a histopathologist. He was told a malignant melanoma was in fact an ulcer which delayed effective treatment. One expert argued that a reasonable competent histopathologist could have made such an error. On a straightforward application of Bolam, there would be no finding of negligence. However, Kerr J was unconvinced of Bolam’s fit:

In a case involving advice, treatment or both, opposed expert opinions may in a sense both be “right”, in that each represents a respectable body of professional opinion. The same is not true of a pure diagnosis case such as the present, where there is no weighing of risks and benefits, only misreporting which may or may not be negligent. ³⁴

Bound by the law, Kerr J found that negligence was established, not by dis-applying Bolam to purely diagnostic errors but by applying the Bolitho gloss. ³⁵ Poole has argued that this point is likely to be developed should a case on pure misdiagnosis or misreporting reach the higher courts. ³⁶

The Bolamisation process referred to by Brazier and Miola has also been restricted. In the context of mental capacity, best interests had been equated with acting in accordance with an accepted body of medical opinion in Re F, ³⁷ but this notion was criticised in Re S. ³⁸ There, Butler-Sloss P acknowledged that the starting point in a medical decision is Bolam, but there may be more than one Bolam-compliant option and the court’s role is to determine the best option. ³⁹ Section 4 of the Mental Capacity Act 2005 now takes a patient-centred approach, evaluating the ‘best’ option in light of the patient’s values. ⁴⁰

What of non-disclosure of information? In Bolitho, Lord Browne-Wilkinson’s observations focused on treatment and diagnosis cases. ⁴¹ Two separate qualifications in relation to information disclosure were set out in Sidaway. They were, however, of limited value to patients. First, the judgment distinguished cases where questions were raised directly by the patient, in which case a duty would arise. ⁴² This was problematic, not least because the patient may lack the clinical expertise to know to raise the question. ⁴³ A second qualification related to disclosures that are ‘so obviously necessary to an informed choice on the part of the patient that no reasonably prudent medical man would fail to make it’, for example ‘if there was a substantial risk of grave adverse consequences, as, for example, the ten per cent risk of a stroke’. ⁴⁴ This qualification was described by Lords Kerr and Reed in Montgomery as ‘inherently instable’: the application of the Bolam test to information disclosure emphasised the relevance of clinical judgment which sat uneasily with the recognition that patients are entitled to make an informed choice. ⁴⁵ The Supreme Court recognised that, as a result, the qualification has been restrictively applied. ⁴⁶

In Sidaway, information disclosure was judged by the majority to constitute an integral part of medical treatment. Lord Scarman, dissenting, stated: ‘The profession, it is said, should not be judge in its own cause’. ⁴⁷ He pointed out the existence of ‘the patient’s right to make his own decision’ or ‘self-determination’, ⁴⁸ which is protected by the common law and ought not depend merely on medical considerations. Lord Scarman supported a test for materiality that put the patient centre-stage. HCPs would be required to disclose risks to which a reasonable patient would attach significance. ⁴⁹ As we have seen, Lord Diplock’s position was that doctors are better equipped to determine reasonableness than the courts, but the single test proved problematic in relation to information disclosure, because judgment as to its adequacy is less dependent on medical expertise. ⁵⁰

So began a disaggregation of treatment/diagnosis and information non-disclosure. ⁵¹ The courts effected an evolution which others have charted in detail ⁵² and we summarise only briefly. In Pearce, Lord Woolf proposed a variation on the standard of disclosure set out in Sidaway: ‘if there is significant risk which would affect the judgment of a reasonable patient, then in the normal course it is the responsibility of a doctor to inform the patient of that significant risk’. ⁵³ Subsequently, in Chester v. Afshar ⁵⁴ – a case on causation rather than breach – Lord Steyn asserted that: ‘[The patient’s] right of autonomy and dignity can and ought to be vindicated by a narrow and modest departure from traditional causation principles’. ⁵⁵

The steps towards a patient-centred approach to information disclosure culminated in the Supreme Court decision of Montgomery v. Lanarkshire HB, which rejected the test set out in Bolam and perpetuated in Sidaway (though we will argue that the extent of that rejection is unclear) in favour of a patient-centred test. ⁵⁶ The decision recognised the right of persons to determine what shall be done with their bodies. Montgomery effected the dissection of negligence that Lord Diplock (above) warned against: the tests in relation to treatment/diagnosis and information disclosure were segregated. Bolam remains pertinent where different ‘schools of thought in medical science’ may apply, but Lords Kerr and Reed found that its application in relation to discussions of risk could lead to acceptance of ‘divergent attitudes among doctors as to the degree of respect owed to their patients’. ⁵⁷ The new test of materiality in information disclosure set out in Montgomery is whether:

a reasonable person in the patient’s position would be likely to attach significance to the risk, or the doctor is or should reasonably be aware that the particular patient would be likely to attach significance to it. ⁵⁸

Communication of risk, benefit and reasonable alternatives is now judged according to what the reasonable and sometimes the actual patient wants to know. ⁵⁹ Montgomery adopts a mixture of subjective and objective assessments. In determining materiality, the Supreme Court emphasised that it cannot be defined as a matter of statistical analysis but is specific to the facts and the characteristics of the patient. ⁶⁰ The Supreme Court Justices did not make clear precisely how to determine the subjective element, but Lords Kerr and Reed did provide guidance as to the dialogue necessary to establish relevant patient characteristics. ⁶¹

Practical guidance to doctors enhances the patient-centred approach taken in Montgomery. Pre-Montgomery guidance from the General Medical Council (GMC) on consent requires that patients ‘must’ be given the information ‘they want or need’ in relation to a range of issues ⁶² and that doctors ‘should’ ‘explore these matters with patients, listen to their concerns, ask for and respect their views, and encourage them to ask questions’. Post-Montgomery, new draft GMC guidance replaces ‘should’ with ‘must’. ⁶³ In addition, it requires doctors to tailor information to patient needs, wishes, values and priorities as well as their level of understanding and the nature of the proposed options. ⁶⁴ While Montgomery endorsed a view already adopted by the GMC, the Council’s draft revised guidance is testament to the practical importance of the legal development and the need to operationalise the Montgomery principles beyond the law of negligence.

The Bolam retribution

One of the themes that has been taken forward in case law post-Montgomery is that ‘the Bolam approach is no longer appropriate in cases of informed consent’. ⁶⁵ However, this section specifies ways in which the Bolam legacy endures. We acknowledge the need to dissect the law of negligence in Montgomery, so as to recognise different duties with respect to treatment/diagnosis and information disclosure. A failure to do so would have perpetuated medical dominance and paternalism. However, we also recognise the force in Lord Diplock’s warning in Sidaway that the dissection would lead to problems of impracticability. We argue that there is a lack of clarity as to the dissection in Montgomery and its subsequent common law development that muddies the boundaries of Bolam’s continued relevance.

The effect of Montgomery was to overrule Sidaway and disapply Bolam with respect to information non-disclosure. Lady Hale concluded that ‘once the argument departs from purely medical considerations and involves value judgments of this sort’, Bolam ‘becomes quite inapposite’. ⁶⁶ This section identifies areas in which there is debate as to whether value judgments based on the needs and preferences of the patient are relevant to aspects of information non-disclosure. Where medical expertise is relevant and value judgments of the sort envisaged in Montgomery are not pertinent to the issue, there is an argument that Bolam remains relevant. Montgomery leaves open the possibility that the new test for materiality might apply to the reasonableness of knowledge of risk, selection of treatment alternatives, the operation of the therapeutic exception and potentially disclosure of differential diagnosis. But an alternative and dominant view reasserts the relevance of the Bolam test alongside the Montgomery test, drawing on Lords Kerr and Reed’s

fundamental distinction between, on the one hand, the doctor’s role when considering possible investigatory or treatment options and, on the other, her role in discussing with the patient any recommended treatment and possible alternatives, and the risks of injury which may be involved. ⁶⁷

First, there was initial uncertainty as to Montgomery’s effect on the relevance of the Bolam test in cases concerning non-disclosure of information. In Spencer v. Hillingdon, HHJ Collender QC applied the Bolam test in the context of advice on post-operative risk, subject to a Montgomery ‘gloss’. The test applied was: ‘would the ordinary sensible patient be justifiably aggrieved not to have been given the information at the heart of this case when fully appraised of the significance of it?’. ⁶⁸ This interpretation was later rejected in Thefaut v. Johnson, where Green J did not accept that the Montgomery test is a variant of Bolam. ⁶⁹

Second, Bolam has potential application in relation to the therapeutic exception, which in exceptional circumstances HCPs might employ if they consider disclosure of material information to be detrimental to the health of the patient. ⁷⁰ Because of the relevance of medical expertise to its application, Bolam is likely to apply. ⁷¹ Montgomery was influential in the recent Singapore Court of Appeal in Hii Chii Kok v. Ooi Peng Jin London Lucien (Hii). ⁷² In that case, Bolam was held to be relevant to the consideration of reasonable justifications for withholding information, such as waiver, emergency treatment and the therapeutic exception. We comment no further in light of arguments previously articulated by one of us that the therapeutic exception is of dubious relevance and seems unlikely to be developed or relied upon in legal proceedings. ⁷³

Third, the question arises as to the test to be applied in relation to the risks associated with treatment. HCPs cannot inform patients of risks they know nothing about, but in some cases, they might be found to have constructive knowledge of risk: if information on risk is material, but the HCP did not know of it, then the question is raised as to the reasonableness of the HCP’s position. Ought this matter to be governed by Bolam? Austin notes that the Singapore Court of Appeal in Hii applied Bolam on the basis that the lack of information would lead to a wrongful diagnosis or negligent treatment. ⁷⁴ Austin is critical of this conclusion on the basis that the crux of the action is advice on risk. The matter came before the Court of Appeal in Duce v. Worcester Acute Hospitals NHS Trust where a patient underwent surgery to relieve pre-existing pain and was not warned that the operation could result in serious and permanent pain. Montgomery was framed as a two-part test:

(1) What risks associated with an operation were or should have been known to the medical professional in question. That is a matter falling within the expertise of medical professionals [83].

(2) Whether the patient should have been told about such risks by reference to whether they were material. That is a matter for the Court to determine [83]. This issue is not therefore the subject of the Bolam test and not something that can be determined by reference to expert evidence alone [84]–[85]. ⁷⁵

The two parts are distinguished on the basis of the relevance of expertise to the judgment of reasonableness: Bolam governs the matter of whether the risks were or should have been known; if so, Montgomery governs the question of whether the patient should have been informed. In Duce, the second part of the test (materiality) was irrelevant in light of the fact that the first part was not satisfied: the risks were not considered by medical experts to be known risks at the relevant time. From a patient’s perspective, the dual operation of Bolam and Montgomery and reliance on expert opinion in determining constructive knowledge of risk might seem to perpetuate the HCP-centred standard set out in Bolam. They might argue that their values are potentially relevant to the matter of what risks should have been known.

Fourth, a duty to warn of reasonable alternatives that carry fewer or no risks was upheld in Birch v. University College London Hospital NHS Foundation Trust, ⁷⁶ and a requirement to discuss ‘any reasonable alternative or variant treatments’ was recognised in Montgomery. ⁷⁷ In Montgomery, the non-disclosure of risk relating to shoulder dystocia was directly related to the requirement to inform the patient of alternative treatment by caesarean. ⁷⁸ This may be the extent of the duty to discuss alternatives, in which case Bolam remains pertinent to the reasonableness of clinicians’ decisions regarding treatment selection, except where selection is necessitated by the communication of a material risk. If a failure to consider an alternative does not breach Bolam, then it is less likely to be considered a ‘reasonable’ alternative, even if it would not breach Bolam for a clinician to offer that alternative. Poole warns that this has potential to undermine Montgomery:

Suppose there is an innovative treatment being used by 10 per cent of clinicians in a particular field, and that the patient’s clinician knows about it but considers it not to be a reasonable alternative because it has not been sufficiently tried and tested. If the Bolam test applies then the Court might well find that it was not negligent to have failed to have mentioned the alternative to the patient. Is that not the very paternalism that the Supreme Court in Montgomery was striving to end? ⁷⁹

An argument might be advanced that Montgomery has wider relevance to treatment alternatives: if a reasonable patient would find the alternative significant or the clinician is aware that the particular patient would find it so, then it ought to be put to the patient. ⁸⁰ On this basis, alternatives should sometimes be communicated to the patient, even if they would pose greater clinical risk than the clinically optimal treatment, because it would be for the patient to choose between them according to their values and preferences. Bolam would still be relevant to the selection of treatment alternatives insofar as the clinician should offer alternatives unless no reasonable doctor would do so. This would maximise the patient’s opportunity to decide whether a clinically riskier treatment is in fact better suited to their values and preferences.

However, there has been a marked judicial reluctance to adopt such an approach. We focus in this section on the doctrinal development to demonstrate the continued relevance of Bolam in aspects of the HCP’s advisory role. In the next section, we turn to the rationale behind it. In Bayley v. George Eliot Hospital, the claimant argued that she ought to have been advised of an alternative treatment that she would have been willing to fund privately. ⁸¹ HHJ Worster rejected this argument. The alternative was a possible option, but in 2008 it was not an appropriate option. ⁸² Elements of Bolam flow from the judge’s focus on the reasonable competent surgeon in 2008. ⁸³

In Scotland in 2018, Lord Boyd issued an Opinion arising from the use of vaginal mesh products in the treatment of the pursuers in which the definition of reasonable treatment options was contested. Lord Boyd found that Montgomery’s application was restricted to communication of the risks associated with treatment options. ⁸⁴ The Hunter v. Hanley ⁸⁵ test (broadly the Scottish equivalent of Bolam) was relevant to decisions about diagnosis and treatment. A different approach was taken in Hii (Singapore) where the court articulated certain alternatives HCPs are not bound to disclose including ‘fringe’ treatments, alternative medicine, or ‘mainstream treatment options which are obviously inappropriate on the facts’, ⁸⁶ the assessment of which comes under Bolam. ⁸⁷

In England and Wales, it seems that a narrow approach to Montgomery’s application to reasonable alternatives is emerging, particularly where treatment is urgently required. A failure to identify an alternative is most likely to amount to breach of the duty if it is relevant to the risk of proposed treatment. ⁸⁸ However, a wider approach may yet be consolidated in relation to elective procedures where, for example, dialogue around more conservative alternatives may be key, ⁸⁹ or in novel procedures, where dialogue around standard options is important. ⁹⁰

A related issue arises around the duty to inform patients of alternative treatments where diagnosis is uncertain (‘differential diagnosis’). 2008 GMC guidance requires that patients are given information about diagnosis and prognosis and uncertainties thereof. ⁹¹ Pre-Montgomery, in Meiklejohn v. St George’s Healthcare NHS Trust, it was said that the duty to warn of alternative diagnoses comes under Bolam and Sidaway. ⁹² Post-Montgomery it might be argued that dialogue around differential diagnosis is required where it affects the range of alternatives and their relative risks and benefits. However, in light of Duce and the restrictive approach taken to date around treatment alternatives, if the differential diagnosis does not breach the duty of care according to Bolam, Montgomery’s relevance as to its communication to the patient seems likely to be limited.

In the Scottish case of Taylor v. Dailly Health Centre, ⁹³ 32-year-old Mrs Taylor suffered chest pain and subsequently died after a General Practitioner (GP) misdiagnosed her with gastro-intestinal upset and musculoskeletal pain. One of the claims was that the GP had not obtained her informed consent because she was not advised of the risk of alternative diagnosis – acute coronary syndrome – and accompanying treatment options. Finding that Montgomery was irrelevant, Lord Tyre said:

A distinction fell to be drawn between (i) the doctor’s role when considering possible investigatory or treatment options, and (ii) the doctor’s role in discussing with the patient any recommended treatment and possible alternatives. The first remained an exercise of professional judgment, and [in this case] no duty existed to discuss that judgment with the patient or seek consent to treat the patient in accordance with it. ⁹⁴

In 2020, the Paterson Inquiry recommended strengthened procedures to ensure that patients are properly informed of their diagnosis as well as their treatment options. ⁹⁵ Should a case in England and Wales arise, this might influence judges to extend the remit of Montgomery to require communication of differential diagnosis, at least where it raises the possibility of less risky treatment.

It is possible to discern three explanations in Montgomery for the respective relevance of Bolam and Montgomery tests: The distinction between selection of treatment options versus the discussion of those options with the patient focuses on the difference between treatment/diagnosis and informed consent; the relevance of different schools of medical thought versus different approaches to respect for patient choice focuses on the role of medical expertise; and the relevance of the patient’s values to the decision focuses on the protection of patient rights. The explanations are generally harmonious but may conflict. For example, the division in Montgomery between the role of HCPs in selecting treatment options, making diagnoses and understanding the risks, and their role in disclosing material information to patients so as to allow them to make an autonomous choice is problematic insofar as the GMC acknowledges that patient values can be relevant to selection of treatment options. ⁹⁶ There may be cases where patient values render Bolam ‘inapposite’, notwithstanding the relevance of medical expert evidence to determine reasonableness. Subtle differences in rationale may also have resulted in latitude in the subsequent application of Montgomery, resulting in a degree of ‘Bolam creep’.

In conclusion of this section, we have argued that Lord Diplock’s warning about the practicability of dissecting the law of negligence has some bearing. Whilst there is force in the argument that his view flowed principally from conservatism and deference to the medical profession, ⁹⁷ we would argue that it is also based on the practical difficulties in some cases of distinguishing advice and treatment/diagnosis. Bolam’s ‘legacy’ flows from the unclear division, which raises questions as to the extent to which Bolam remains relevant to the assessment of reasonableness of aspects of medical advice. We have argued that the resolution of this debate is far from settled.

The Montgomery enigma

We move now from the continuing doctrinal relevance of Bolam in cases of non-disclosure of risk, reasonable alternatives and differential diagnosis, to its wider influence in relation to medical advice. Notwithstanding the metaphorical legal and cultural nails in Bolam’s coffin, the paternalistic ethos perpetuated in Bolam has proved difficult to curtail. In this section, we explore three reasons for this: that Montgomery seemingly endorses a consumeristic conception of patient choice, that the paternalistic ethos is deeply ingrained in practice, and that outside the confines of negligent non-disclosure of material risk, the principles endorsed in Montgomery are not equally relevant to all aspects of medicine.

The Montgomery decision was based on principles of self-determination, ⁹⁸ partnership, ⁹⁹ support, ¹⁰⁰ choice, ¹⁰¹ and a move away from paternalism ¹⁰² but also on consumerism. ¹⁰³ Seeming acceptance of consumerism has potential to create unrealistic expectations of patient choice and blur the divide between selection of clinically appropriate treatments and patient choice between them. In this section, we set out a normative argument for a clearer balance of patient choice and HCP expertise. Recognition that Bolam is relevant to the selection of reasonable treatment options is problematic if it leads to patient values being ignored. Similarly, recognition of the importance of patient choice is problematic if clinical expertise is ignored. Montgomery requires a revised professional autonomy rather than its abandonment. ¹⁰⁴ There is a role for clinical expertise in the selection of relevant treatment options and in support of patients when they choose between them that is potentially disrupted by emphasis in Montgomery on consumerism. Conceptual clarification is required to recognise and better define the extent of professional autonomy.

The rise of consumerism is not new. The internal market was introduced in the NHS and Community Care Act 1990, assigning the role of purchaser to District Health Authorities and encouraging providers to reduce costs and enhance efficiency. According to the King’s Fund, these early reforms had limited impact as incentives were weak, constraints strong and information insufficient. ¹⁰⁵ It resulted in culture change, however, encouraging greater focus on patient needs. ¹⁰⁶ New Labour retained the purchaser provider split and brought in commissioning. Market reforms were gradually introduced, starting with a new focus on patient choice to enhance the patient experience in the NHS Plan ¹⁰⁷ and introducing greater flexibility for providers through Foundation Trust status and Payment by Results. Market-based approaches were extended in the Health and Social Care Act 2012, ¹⁰⁸ underpinned by an ‘information revolution’. ¹⁰⁹ Competition resulted in a mix of private and state-run NHS organisations which, Sturgeon argues, made the NHS more business-like and consumeristic. ¹¹⁰ Sturgeon points out that the ‘necessity’ of consumption in the NHS makes for valuable commercial opportunities and thus drives consumerism. ¹¹¹

The Supreme Court Justices reference choice 17 times in Montgomery. Lords Kerr and Reed recognise that patients are ‘widely treated as consumers exercising choices’. ¹¹² Consumerism can have value in the pressure it brings to bear to reform practice, ¹¹³ but critics point to the unsuitability of a consumer model in healthcare on the basis that patients need support rather than an individualised choice architecture. ¹¹⁴ Consumerism can also lead to unmet patient expectations: it denotes a wide range of choice that is in practice limited. Choice is limited by resources, evidence of clinical need and beneficent obligations including Section 1 of the Health and Social Care (Safety and Quality) Act 2015 ¹¹⁵ which requires the Secretary of State to ensure that services cause ‘no avoidable harm’ to service users. ¹¹⁶ The Care Quality Commission states ‘providers must not provide unsafe or inappropriate care just because someone has consented to care or treatment that would be unsafe’. ¹¹⁷ Competition has been introduced, with varying degrees of success, to allow some choice of provider, though there is still considerable control by HCPs as to when to refer and to what specialism. In relation to choice of treatment, in 1985 Schwartz and Grubb claimed that Britain could not afford informed consent because the system could not respect patients’ reasonably possible choices:

Ultimately, the existence of a national health system, prospectively funded by a central planning authority and with limited and defined resources, makes the acceptance of such a doctrine an impossibility. ¹¹⁸

Primacy cannot be given to patient rights to decide which treatment suits them best in the way that it can in a system where consumers more directly fund their treatment. ¹¹⁹ Notwithstanding creeping privatisation of the NHS, treatment in the NHS remains largely free at the point of delivery and Schwartz and Grubb’s argument remains relevant. Patient choice cannot extend to the right to demand treatment, though there is evidence of increasing pressure on clinicians to comply with patient demands even when the options sought by patients are not clinically indicated. ¹²⁰ Rather, patient preferences must often give way to medical criteria, including the allocation of resources. Though Lords Kerr and Reed are clear that professional expertise remains central to the selection of suitable treatment options, the emphasis on consumerism in Montgomery is unfortunate. In dismantling HCP-centred medicine, care must be taken to distance anti-paternalism from unmitigated patient choice. The HCP role in selecting treatment options can be carried out in a patient-centred manner that takes into consideration patient values, without compromising the professional duty to base medical treatment decisions on clinical need, evidence of efficacy and available resources. ¹²¹

We have described a conflict of principles set out in Montgomery. Emphasis is placed on consumerism that has potential to cause opacity in relation to the selection of treatment options and the choice between them and raise patient and HCP expectations that patients can choose treatment options that HCPs do not consider clinically justifiable. But the Supreme Court also acknowledged the importance of dialogue and partnership. It is apparent that the rejection of Bolam in Montgomery does not denote a rejection of the role of HCP support in the choice between options. Nor does acceptance of the role of medical expertise in the selection of treatment options mean that consideration of patient values is excluded from the exercise. Shared decision-making is relevant to both selection and choice between options. ¹²²

As Montgomery is incrementally developed in the courts, it is important that the relevance of shared decision-making and patient autonomy are recognised and promoted. This would help facilitate the articulation of Bolam and Montgomery’s respective application, whilst promoting workable patient-centred standards  in a manner that is alive to the needs of some patients for support and to the relevance of HCP expertise in determining the suitability of treatment and in providing advice.

Another part of the Montgomery ‘enigma’ is the potential gap between legal principle and application. The GMC conducted research in 2017 reporting that awareness of the Montgomery decision is low and doctors fear that a lack of time in consultations can prevent a full understanding of the particular patient’s needs. Doctors can find assessments of capacity difficult and are not all equipped to translate complex information in a digestible form. ¹²³ If cultural change can be brought about, resources are needed for implementation. But limitations on resources impact upon the time HCPs have to engage in dialogue and the alternatives that can be offered. ¹²⁴ Laing has highlighted the stresses on the partnership model that the Montgomery judgment advocates, ¹²⁵ that result from the financial pressures on the NHS and the considerable strains under which the work force operate. ¹²⁶

There is also evidence of variation in Montgomery’s perceived relevance between specialties. The production of specific consent guidance post-Montgomery from the Royal College of Obstetricians and Gynaecologists, ¹²⁷ Association of Anaesthesiologists, ¹²⁸ Royal College of Surgeons, ¹²⁹ Royal College of Nursing ¹³⁰ and British Orthopaedics Association ¹³¹ reflect the fact that these are the specialisms most likely to face clinical negligence litigation. Outside these areas, there is less focus on the legal changes from the Royal Colleges, at least in terms of published guidance responding to Montgomery. We can surmise that there may therefore be less understanding of the new responsibilities among HCPs in those specialties that are less likely to face clinical negligence suits.

Some patient choices bypass HCPs altogether, which self-evidently limits opportunities for partnership and support relied upon in Montgomery. In essence, patients may rely on advice in a situation where there is no individual HCP to challenge in a lawsuit, in which case there may be less pressure on those responsible for the advice to make it compatible with Montgomery principles. For example, the information revolution can leave vulnerable people exposed to harmful information as seen in social media suicide posts ¹³² and depictions of Anorexia Nervosa as a positive lifestyle choice. ¹³³ Access to complex healthcare information that bypasses the HCP gatekeeper can pose risks of misunderstanding, as in the case of some direct-to-consumer predictive and pre-dispositional genetic testing ¹³⁴ and online pharmacies. ¹³⁵ In 2010, the Nuffield Council on Bioethics concluded that the government needs to do more to ensure information is accurate and consent is informed. ¹³⁶ Patients can access more information than ever before and this can steer people who do not need medical intervention away from healthcare services, help patients to frame their questions for HCPs and reassure them post-consultation. ¹³⁷ On the other hand, many evidence-based articles are hidden behind a paywall and misinformation can put strain on services and exacerbate public health concerns. ¹³⁸ In such cases, Montgomery is not directly applicable, but the principle of patient-centred care remains relevant. Adherence to that principle may necessitate pressure on the government to ensure that public information is fit for purpose.

Variations in treatment type might also affect Montgomery’s application. For example, medical advancements allow increasingly individualised care which can decrease choice and impact on the partnership model. The drive towards personalised (or precision) medicine tailors advice, prevention, diagnosis, treatment and aftercare to the individual, usually based on genetic data, rather than the disease. For example, cancer treatment can be matched to the patient’s genetic make-up and tumour growth. Montgomery emphasises the importance of patient values, ¹³⁹ but it is not clear how values will fit into personalised medicine. Clinically indicated options might be reduced to one and the advisory function of the HCP diminished, with the result that less attention is paid to the patient’s particular values and preferences. Adherence to a patient-centred approach might lead to greater focus on tailoring the choices to the person as well as their genetic profile if individual values and personalised medicine are brought into tension.

Some aspects of law are regulated by statutes enacted when Bolam was more dominant. Take, for example, the Abortion Act 1967. ¹⁴⁰ Evidence of an HCP-centred approach ¹⁴¹ flows from the medicalisation of abortion. Though during the coronavirus crisis temporary approval was given to early medical abortion at home, ¹⁴² the 1967 Act takes a medicalised approach whereby two registered medical practitioners must form an opinion in good faith that the abortion fits within one of a series of defences to what would otherwise constitute a crime under sections 58–59 of the Offences Against the Person Act 1861. The Act pays little heed to the process of consent and the woman’s understanding of relevant information. Fifty years after the Act came into force, Lee et al. describe a shift in professional attitude that exposes tensions in the Act flowing from better support for women’s choices and the rejection by HCPs of medical hierarchies. ¹⁴³ The articulation of principles of information disclosure in Montgomery may lead to new challenges of Bolam-era legislation if evidence suggests that the legislation fails to support patient-centred practice.

Brazier and Montgomery persuasively argue that Bolam is gradually being ‘left behind’. They point to the Supreme Court decision of Darnley as an example. ¹⁴⁴ There, it was held that inaccurate information that leads to harm is actionable in an accident and emergency department, just as it would be in other settings. They see this as evidence of ‘a more orthodox approach…to determining the standard of care in hospitals’. ¹⁴⁵ We agree that the adoption of patient-centred principles in Montgomery and related cases is having a positive effect, but caution that Montgomery’s potential should not be exaggerated. Montgomery concerned a physical intervention on a living patient. Differentials among health providers (from hospitals to care homes, surgeries to hospices, private and NHS) result in the prioritisation of different missions and values. The principles of partnership and patient autonomy endorsed in Montgomery have relevance in research, observational studies, use of personal data, and in issues after death such as organ or tissue transplantation. In these settings, however, competing values may apply. The plurality of values outside mainstream medicine makes for uneven application of patient-centred practice and partnership. Data-sharing encompasses public interest and commercial value of data; public health emphasises the public interest, as has become so apparent in the coronavirus pandemic ongoing at the time of writing. ¹⁴⁶

It is also worth noting that Montgomery focuses on a restricted patient group – namely ‘adult patients of sound mind’. ¹⁴⁷ Patients who lack capacity, who are subject to compulsory treatment under the Mental Health Act 1983, or who are under the age of 16 and lack the competence to consent have dubious claims for negligent non-disclosure of risk, but the principles of support, participation, partnership and patient-centred medicine are relevant here too and have potential to influence legal development outside of negligence. This matter is explored in a separate paper. ¹⁴⁸

This section has not sought to offer a comprehensive assessment of Montgomery’s relevance across healthcare, but rather to point out the potential for variability in commitment to patient values when giving medical advice, given constraints on HCP time, resources and awareness and the differences in HCP-type, patient group and healthcare setting. Montgomery is not a panacea in the move away from HCP-centred practice in information provision that followed the Bolam decision. There are cases where patient autonomy and partnership must give way to competing values.

The Montgomery supremacy?

We have focussed on the Bolam legacy in two contexts: the first is within the law of clinical negligence and the second looks beyond negligence to the potential to mitigate an entrenched Bolamised approach and deliver a patient-centred approach to medical advice. We are some way off a ‘Montgomery supremacy’ in both regards.

Taking the latter first, we have identified aspects of advice, areas of medicine and patient demographics where Montgomery’s bearing is limited. Two 2020 independent inquiries have outlined failures with regard to informed consent and the need to strengthen commitment to a patient-centred approach embracing partnership and patient autonomy. The Paterson Inquiry chaired by Rev’d Graham James made recommendations to improve the consent process, including writing to patients to outline their condition and treatment and to make sure the patient’s GP is also informed, and giving patients a short period of time to process information about diagnosis and treatment options before surgical procedures. Additional points were made in Baroness Julia Cumberlege’s medicines and medical devices safety review. ¹⁴⁹ Describing Montgomery as ‘landmark’ and ‘a watershed’ ¹⁵⁰ the 2020 report stated:

No longer can informed patient consent be anything other than a true equality of partnership in the decision making process between patients and their treating physicians. Their care and treatment should not be a series of events that happened to them. Rather, every patient should be able to stand back, look at their patient journey and say ‘I recognise my handwriting all over those choices. ¹⁵¹

The report suggests better use of patient decision aids and access to patient testimonials as part of the informed consent process. It further recommends recording conversations where appropriate, so that patients have opportunity to reflect. Both reports advise the GMC to incorporate recommendations reflecting the spirit of Montgomery into guidance and to monitor compliance as part of ‘Good Medical Practice’. ¹⁵² If this is achieved then Montgomery’s relevance will be expanded further beyond the law of negligence to aspects of practice where patients have a legitimate expectation of partnership and choice. In doing so it will be important to acknowledge that choices are often limited and that partnership does not equate to a patient right to demand interventions that are not clinically justified.

With regard to claims for negligent non-disclosure of information, though Montgomery signalled the dissection of the duty of care and the end of Bolam’s reign with regard to the test for materiality of risk, we have identified ways in which Bolam remains pertinent. The disapplication of Bolam in Montgomery is not complete, but this does not mean that the Bolam legacy cannot be confined where it threatens unwarranted paternalism in place of partnership and sensitivity to patient values. The significance of advice does not end with risks inherent in treatment or even with treatment. Aspects of diagnosis/treatment and information disclosure will converge when, for example, advice is needed for diagnostic testing; or where treatment necessitates advice regarding prognosis. ¹⁵³ There is potential for the incremental development of Montgomery to enhance respect for patient autonomy and emphasise shared decision-making (or ‘partnership’ as it is referred to in Montgomery) beyond the confines of non-disclosure of risk, benefits and reasonable alternatives, to encompass wider aspects of the HCP’s advisory role.

One method at the courts’ disposal is to apply Montgomery when patient values are pertinent on the facts. As we have seen, where disclosure of risk relating to one treatment option can be linked to the reasonableness of disclosure as to an alternative option, the Montgomery test will be relevant: the failure to offer an alternative becomes a matter of disclosure rather than treatment selection, because of the relationship to the risk inherent in the accepted treatment. The same argument might apply to other aspects of the decision-making process. For example, a differently framed Montgomery case could potentially have focussed on the issue of differential diagnosis. Montgomery involved a misdiagnosis of fetal weight. ¹⁵⁴ While the question of whether there had been a misdiagnosis would fall under Bolam, the same is not necessarily true of failure to communicate doubt as to diagnosis (of fetal weight) if that can be linked to the risk inherent in the proposed course of action.

Another method is to extend Montgomery’s application beyond treatment to encompass other aspects of medical advice. The courts have shown some willingness in this regard. Shortly after the Montgomery decision, in Spencer v. Hillingdon Hospital NHS Trust, the court found that a patient was not given adequate information as to the symptoms of post-operative deep vein thrombosis. ¹⁵⁵ Subsequently, in Gallardo v. Imperial College Healthcare NHS Trust, ¹⁵⁶ Mr Gallardo was treated in 2001 for a suspected stomach ulcer that proved to be a malignant stromal tumour, but the malignancy and risk of recurrence was not disclosed to him until 2010. Judge Peter Hughes QC held that: ‘By analogy [with Montgomery], it is the patient’s right to be informed of the outcome of the treatment, the prognosis, and what the follow-up care and treatment options are’. ¹⁵⁷ The principles set out in Montgomery that supported the right to be informed, also supported the patient’s ‘right to be told’. ¹⁵⁸ In Webster v. Burton Hospitals NHS FT, ¹⁵⁹ it was agreed that a failure to undertake additional ultrasound scans during pregnancy was negligent. The High Court (sitting pre-Montgomery) applied Bolam and found that a reasonable and responsible body of obstetric opinion supported the Consultant’s actions. The Court of Appeal (post-Montgomery) held that this was no longer the correct approach. ¹⁶⁰ Had there been a proper dialogue, Mrs Webster would have opted to have labour induced. Fox argues that application of Montgomery in this case demonstrates the potential for application beyond non-disclosure of information to advice more generally, ¹⁶¹ at least where the patient has not been properly informed and harm has resulted. ¹⁶²

A further possibility is to accept the application of Bolam to certain wider issues of informed consent where the matter in question turns on medical expertise and expert evidence is required to establish its reasonableness but adapt its application to limit the HCP-centred, paternalistic focus. This could potentially be achieved through the application of the Bolitho gloss. It will be recalled that the House of Lords in Bolitho corrected a misinterpretation of Bolam, by requiring that judges are satisfied that medical opinion rests on logical grounds. Two recent cases provide evidence that the courts are willing to question established practice on the basis that it is illogical because it is not compliant with Montgomery. Both are claims for damages for the upbringing of a child. In ARB v. IVF Hammersmith, ¹⁶³ ARB claimed in breach of contract against an IVF clinic that thawed and implanted a frozen embryo into ARB’s former partner without his consent, resulting in the birth of his genetic child. Jay J found at first instance that the consent form had been forged by ARB’s ex-partner. The claim failed on policy grounds, but the Court of Appeal held that the clinic was in breach of its strict obligations given the lack of genuine consent. Montgomery was relevant to the finding that the IVF clinic’s processes for obtaining consent were illogical, unreasonable and irresponsible. ¹⁶⁴ This was so notwithstanding that the process was established and common to other clinics. Subsequently, in Mordel v. Royal Berkshire NHS Foundation Trust, ¹⁶⁵ Jay J said of the ARB decision:

This authority is valuable to the extent that it vouches that a system which does not entail the taking of reasonable steps to ensure that relevant consent is informed may be regarded – subject always to a host of other considerations – as irresponsible, unreasonable and unrespectable even if there may exist expert evidence to support it. ¹⁶⁶

In Mordel, the logicality of processes designed to establish the choices of pregnant women with regard to screening for Down’s syndrome was called into question. Though Montgomery was not ‘central to the issue’ it was nonetheless found to be relevant:

to underscore that the issue of consent and the need for it to be truly informed is a question of right rather than discretion; and that the patient must be provided with sufficient information to make an informed decision. ¹⁶⁷

If Bolam is to sit more comfortably alongside Montgomery, then the Bolitho gloss might be utilised to ensure that reasonableness is viewed through the prism of Montgomery principles. In Birch, Cranston J opined that even if the failure to warn of the existence of a less risky alternative did not deny the patient the opportunity to give informed consent, ‘the failure to discuss with Mrs Birch these matters could not be described in law as reasonable, responsible or logical’. ¹⁶⁸

Post-Bolam, judges were initially reluctant to evaluate an HCP’s reasonableness and responsibility, deferring instead to medical expert opinion. Post-Bolitho, Teff questions the circumstances in which expert evidence might be disregarded. ¹⁶⁹ Lord Browne-Wilkinson’s reference to ‘logic’ might focus narrowly on internal consistency and credibility of the expert. Conversely, ‘use of the term logic may be seen as a residue of deference, not to be understood in any unduly constraining sense in future cases’, ¹⁷⁰ in which case there is scope for judges to question the substantive content of the decision.

In 2018, the Court of Appeal took a restrictive stance, setting a high standard for a claimant to satisfy: except in stark cases, expert evidence will usually be required to demonstrate illogicality. ¹⁷¹ Nonetheless, wider application of Montgomery principles of partnership and patient-centred practice might require HCPs to negotiate the ‘schools of thought in medical science’ ¹⁷² with the patient in mind. If so, then there are grounds upon which to argue that HCP-centred practice is illogical and untenable in those rare cases that alternative ‘schools of thought in medical science’ apply, but the identification of treatment options or constructive knowledge of risk flows from ‘divergent attitudes among doctors as to the degree of respect owed to their patients’. ¹⁷³

In this section, it has been argued that commitment to the Montgomery principles of autonomy and partnership would lead to gradual extension of the patient-centred Montgomery approach by analogy. Even where Bolam applies, the relevance of Montgomery to the Bolitho gloss has been established in ARB and is likely to lead to other instances where processes that do not comply with Montgomery principles are challenged.

Conclusion

The emphasis on HCP-centred medicine in Bolam has gradually been undermined by social and legal developments, ¹⁷⁴ but we have set out two ways in which the Bolam legacy remains relevant when judging the adequacy of medical advice. First, Montgomery only goes so far in challenging the pervasive nature of Bolam. Inherent limitations flow from reference in Montgomery to consumerism and there are practical impediments to Montgomery’s application. As the Cumberlege report ¹⁷⁵ recently exemplified, paternalistic behaviour is still commonplace. In extending the relevance of a patient-centred approach beyond the law of non-disclosure of material risk, we have argued that it is important to manage consumeristic patient expectations. HCPs have a legitimate role in limiting recourse to treatment options that are not clinically indicated.

Second, Bolam retains doctrinal relevance. Lord Diplock warned in Sidaway that the dissection of negligence would be neither meaningful nor practical. We have argued that the dissection was a meaningful and relevant response to the ethical requirement to challenge medical dominance over the law, but we have acknowledged and described difficulties as to practicability and coherence. The division of negligence is in part a distinction between treatment and diagnosis on the one hand and information and advice on the other. It is in part a distinction according to the relevance of medical expertise and patient values. As we have seen, the different explanations do not always align. The focus on medical expertise is not a principle of law, but a practical distinction based on the fact that diagnosis and treatment are more likely to raise technical questions. ¹⁷⁶ Where medicine is less technical in nature, or where patient values are pertinent to the decision, judges may show increasing willingness to question the clinical view whether through direct application of Montgomery or the development of the Bolitho gloss.