Disappointingly clinical. McCulloch and others (Appellants) v Forth Valley Health Board (Respondent) (Scotland) [2023] UKSC 26

Negligent treatment

The claimants argued that several aspects of Dr Labinjoh’s care constituted negligence, ¹⁷ including failing to prescribe NSAIDs as treatment for Mr McCulloch’s pericarditis. ¹⁸ It is well established that questions of breach in the context of diagnosis and treatment are governed by Bolam v Friern Hospital Management Committee (‘Bolam’) ¹⁹ —that, provided the clinician has ‘acted in accordance with a practice accepted as proper’ by a responsible body of professional opinion, the legal standard of care is satisfied. ²⁰ This is qualified by the decision in Maynard v West Midlands Regional Health Authority ²¹ as well as the Bolitho v City and Hackney HA ‘gloss’, ²² which, taken together, mean that: where multiple schools of thought exist regarding appropriate treatment, the court is not at liberty to choose which they prefer ²³ ; however, this need not imply absolute deference to expert testimony. It is within the court’s jurisdiction to scrutinise a body of opinion through testing its capability of withstanding logical analysis. ²⁴ Importantly, on the evidence, the prescription of NSAIDs to a patient who was not in pain (and who presented with Mr McCulloch’s comorbidities, for instance, stomach problems) was a matter on which clinical judgement legitimately differed. ²⁵ Thus, Dr Labinjoh was not negligent. Regarding questions of treatment, the Inner House unsurprisingly reasserted its unease with ‘stray[ing] into the realm of medical expertise’. ²⁶

Information disclosure

More important for our purposes, though, are the arguments pertaining to consent and information disclosure, and how the decision in McCulloch tempers the reach of Montgomery. The claimants alleged that Dr Labinjoh failed to inform Mr McCulloch of the material risks associated with, and variant treatments for, his condition. ²⁷ Specifically, Dr Labinjoh did not explain the possibility of using NSAIDs to treat pericarditis, along with any associated risks. It was argued that Mr McCulloch was thus deprived of necessary information which, had he possessed, his family were certain he would have acted on. To appreciate the effect of the Court’s findings in this regard, it is necessary to look at the case on which the claimant’s arguments were built: Montgomery.

Prior to Montgomery, the standard of information disclosure was considered in Sidaway v Bethlem Royal Hospital (‘Sidaway’). ²⁸ In this case, the Law Lords were divided on the appropriate test, with Lord Diplock stating that Bolam provided a ‘principle of English law that is comprehensive and applicable to every aspect of the duty of care owed by a doctor to his patient.’. . . ²⁹ Essentially, a doctor would only be required to disclose a risk inherent in treatment if no professional body of opinion would support not disclosing it. Lords Bridge, Keith, and Templeman accepted that the issue should ‘be decided primarily on the basis of expert medical evidence’, ³⁰ subject to important caveats. ³¹ Lord Scarman, dissenting, stated as part of the ‘proper respect’ for patients’ rights and self-determination, information disclosure ought to be governed by a patient-centric approach. ³² The jurisprudence following Sidaway struggled – perhaps understandably – to make sense of the judgement (with some courts adopting Lord Bridge’s strict Bolam test, ³³ while others applied Bolam, subject to their own caveats and qualifications). ³⁴ Still, reliance on Bolam in this context increasingly became subject to academic ³⁵ and judicial criticism ³⁶ for several reasons, including (1) that it reinforced paternalistic attitudes unsuited to our modern socio-cultural climate, ³⁷ and (2) that it failed to reflect a doctor’s ethical duty, as delineated by professional guidance, which promoted a ‘partnership model’ of medical decision-making. ³⁸ Thus, when the Supreme Court in Montgomery was presented with the opportunity to depart from Bolam in the context of information disclosure, it did so quite emphatically.

The Supreme Court stated that patients are entitled to decide for themselves which risks they are willing to take in the course of treatment as ‘persons holding rights’ to self-determination instead of ‘passive recipients of the care of the medical profession’. ³⁹ This was enshrined in the new test of materiality, which asks whether

in the circumstances of the particular case, 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. ⁴⁰

Importantly, the Court in Montgomery placed emphasis on the duty to warn – not only of the risks inherent in the treatment the patient actually received – but also of ‘any reasonable alternative or variant treatments’. ⁴¹ How far, exactly, this obligation should extend was precisely the issue in McCulloch: what was to be the interpretation and scope of ‘reasonable alternative or variant treatments’ for the purposes of information disclosure? The claimants argued that, in the spirit of Montgomery, ‘what constitutes a reasonable alternative treatment is to be determined by the Court, unshackled from the professional practice test’. ⁴² Conversely, counsel for the defendants submitted that ‘reasonable alternative or variant treatments’ were phrases synonymous with ‘clinically appropriate’ or ‘clinically suitable’. ⁴³ Thus, following this interpretation, Dr Labinjoh’s duty to discuss treatment using NSAIDs with Mr McCulloch did not arise because, on her assessment (which was supported in evidence as reasonable), NSAIDs were not clinically appropriate in Mr McCulloch’s circumstances.

Reframing Montgomery

While noting that the crux of the issue in Montgomery was not a failure to discuss alternative treatments, the Supreme Court in McCulloch thought it could nevertheless be reframed and viewed through this lens. ⁵³ This entailed the following thought experiment: rather than alleging that she was not informed of the risks involved in vaginal delivery, Mrs Montgomery might have instead argued that the alternatives to vaginal delivery had not been disclosed to her. With this in mind, the Supreme Court in McCulloch tested its proposed use of Bolam against Mrs Montgomery’s retrospectively reframed case. ⁵⁴ It concluded that the alternative treatment in question – that is, a caesarean section – legally would have needed to be discussed with Mrs Montgomery, because it was clear that this alternative was reasonable on a Bolam analysis ⁵⁵ (according to the court in McCulloch, ‘there was no responsible body of medical opinion denying that a caesarean section was a reasonable alternative procedure to the vaginal delivery’.) ⁵⁶

But in the Supreme Court’s view, Mr McCulloch’s evidentiary circumstances starkly contrasted with the facts in Montgomery. Thus, while no reasonable doctor would disregard a caesarean section as a reasonable treatment option, question marks, or clinical ambiguity regarding appropriate treatment did exist for Mr McCulloch. ⁵⁷ More specifically, one expert witness – Dr Bloomfield – testified that, given the patient’s lack of pain and his other ailments, reasonable doctors might disregard NSAIDs as a reasonable treatment option. ⁵⁸ As such, Dr Labinjoh was entitled to conclude that the prescription of NSAIDs for Mr McCulloch was (1) not necessary, (2) not a reasonable treatment option, and therefore (3) need not be discussed with him. This is because, in the Court’s finding, ‘the narrowing down from possible alternative treatments to reasonable alternative treatments is an exercise of clinical judgement’, ⁵⁹ and it is only those alternatives that are considered reasonable that ought to be brought to the attention of the patient.

The Court’s illustration of how Bolam would operate in these two cases boils down to a seemingly simple formulation: if expert testimony reveals clinical ambiguity regarding the ‘reasonableness’ of the claimed alternative, then that alternative need not be discussed with the patient. ⁶⁰ Conversely, if the evidence suggests that the list of reasonable alternatives is clear cut, or reveals medical consensus that a particular treatment is reasonable, then the clinician must discuss it with the patient and will have breached their duty to the extent they do this inadequately. In practice, though, this reasoning seems difficult to apply and raises the possibility of criticism and contention. Clinical ambiguity more realistically comes in shades of grey as opposed to being something that is either present or absent (indeed, research aimed at trying to understand clinical ambiguity has found that there is no ‘gold standard’ for measuring its extent, or its influence over clinical decision-making). ⁶¹ Sometimes, clinical ambiguity might only be established after the fact. ⁶² Montgomery itself is a case in point. The Supreme Court’s simple assertion in McCulloch that a caesarean section was definitively a reasonable alternative for Mrs Montgomery (and, so, legally ought to have been discussed with her) relies on aspects of the Montgomery judgement – specifically, the interpretation of expert and professional evidence – which have since been called into question. For instance, commenting on the case, Jonathan and Elsa Montgomery argue that ‘the Supreme Court displayed a poor grasp of the risks and benefits of the procedures in question, and engaged inadequately with the guidelines drawn up to assist clinicians to make evidence-based decisions’. ⁶³ They note, after examination of the evidence, that Dr McLellan was actually ‘supported [by] clinical guidelines and experts on both sides’ – a fact that was not explicitly appreciated by the Court. ⁶⁴ Yet, this suggests that Montgomery, similarly to McCulloch, is more accurately characterised by clinical ambiguity – even though this was downplayed or oversimplified during proceedings.

With this in mind, and applying Bolam, the failure to discuss a caesarean section might not have been as patently negligent as the Court in McCulloch claims. Contrary to the contention that the facts (as they related to the reasonableness of alternative treatments) in the cases of Mrs Montgomery and Mr McCulloch were diametrically opposed, they instead seem quite analogous. The Supreme Court’s assertion then that the claimant in Montgomery would have been entitled to information about alternative treatments, but on the same reasoning, Mr McCulloch would not, rests on either unsafe presumptions or on an entirely different basis than that suggested by the Court. Regardless, this aspect of the Court’s reasoning is problematic.

Unnecessary fragmentation?

McCulloch ultimately results from the fact that the Supreme Court in Montgomery did not spell out when an alternative treatment is to be considered ‘reasonable’. Yet, looking at the wording in Montgomery, it is difficult to support McCulloch’s interpretation that ‘reasonable alternatives’ should fall within the remit of Bolam, rather than the patient-centric test of materiality. Importantly, in Montgomery, the court stated that there is a

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. ⁶⁵

The inclusion of the need to discuss with the patient any ‘possible alternatives’, formulated as if this was in addition to ‘any recommended treatment’ options, supports the contention that they are not one and the same – that is, the possible alternatives to be discussed with the patient are not merely those that the doctor recommends or finds reasonable. This supports our contention that the court in McCulloch effectively further fragments the doctor’s duty of care in the context of information disclosure, but it is not clear that this was necessary or aligns with a faithful reading of Montgomery.

Further evidence of this exists. In delineating the new test of materiality, for example, the Supreme Court in Montgomery devised it as encompassing ‘. . . the alternatives available, and the risks involved in those alternatives. The assessment is therefore fact-sensitive, and sensitive also to the characteristics of the patient’. ⁶⁶ Thus, in constructing the subjective, patient-centric approach to information disclosure, the court embraced the discussion of alternative treatments as falling within the bracket of matters that, contrary to being solely a matter of professional judgement (as found by the Court in McCulloch), are instead ‘sensitive to the characteristics of the patient’. Finally, this point – that the discussion of reasonable alternatives forms a distinct yet integral part of the patient-oriented approach in Montgomery – is implied by the following extract:

Countless other examples could be given of the ways in which the views or circumstances of an individual patient may affect their attitude towards a proposed form of treatment and the reasonable alternatives. The doctor cannot form an objective, ‘medical’ view of these matters, and is therefore not in a position to take the ‘right’ decision as a matter of clinical judgment. ⁶⁷

Thus, the Supreme Court in Montgomery makes no such distinction as the one endorsed in McCulloch, to the effect that reasonable alternatives are to be defined by Bolam, rather than the new approach to information disclosure more holistically construed (which entails consideration of the patient’s circumstances and characteristics). Practically, if the right to be informed of alternative treatments is properly subject to a prerequisite Bolam analysis, then we might ask what the point of its explicit inclusion in Montgomery was in the first place?

Artificial distinctions

Moreover, the approach taken by the court in McCulloch seems unduly artificial and somewhat counterintuitive. Whether an alternative is reasonable is necessarily a function of both facts about the intervention and the patient’s clinical characteristics, ⁶⁸ and the patient’s circumstances and non-clinical characteristics and what matters to the patient (for instance, lucidity versus absence of pain, etc). ⁶⁹ The risks and benefits of various treatment options will be weighed differently by different patients. Whether a low risk of some adverse outcome is significant for a particular patient may depend on the patient’s lifestyle and circumstances, their work, their hobbies, what they value, and so on. Importantly, this is not just the case when we assess reasonableness from the patient’s own point of view, but also when we consider reasonableness of any recommendations for the patient (from an ‘objective’, or the clinician’s, point of view). It is certainly the case if we are obliged, in line with Montgomery, to consider whether,

in the circumstances of the particular case, 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. ⁷⁰

This is to say that the very idea of a purely clinical assessment of reasonable treatments is a misnomer. Disappointingly, though, it is unclear following McCulloch whether or how the patient’s characteristics will be included when litigating the matter. A blueprint of how it might be achieved is provided by the case of Bilal v St George’s University Hospital NHS Foundation Trust (‘Bilal’), ⁷¹ although Bilal was decided prior to McCulloch, and, despite raising similar issues, was not mentioned in the latter case. This is all the more surprising, given that the Court of Appeal in Bilal, ostensibly, applied the same Bolam logic to the question of what constitutes a ‘reasonable alternative’ for the purposes of information disclosure. ⁷² The word ‘ostensibly’ is crucial here, because despite formally adopting Bolam, it is clear that more than just clinical judgement or expert testimony factored into deciding whether alternatives to surgery (i.e. analgesics or medication) were reasonable (and ought to have been discussed with the patient). For instance, in Bilal, reference is made to the patient’s history and experience: the judge notes that the patient had previously tried analgesia to manage his debilitating pain, but did not like the side effects (which were, objectively, relatively minor, but were clearly significant to that particular patient). ⁷³ In addition, the patient’s known views and his philosophical outlook formed part of the assessment. It was said that the patient had previously indicated that he was ‘not keen on trying to mask his pain with medication’, ⁷⁴ and, despite it being a more conservative treatment option, did not want to ‘become dependent on some of the stronger drugs’. ⁷⁵ Ultimately, it was determined that the alternative to surgery, that is, medication, would ‘fail to secure the benefits which [the patient] was desperate to seek to achieve’. ⁷⁶

Clearly (even if not explicitly), elements of the patient’s characteristics – while not being determinative – played an important role; and this serves as a template for how balance could be better achieved between clinical judgement and a patient’s known characteristics. It is unfortunate, then, that no reference was made to the individual circumstances of Mr McCulloch, or what he might have considered to be reasonable, based on the knowledge available to his clinicians. For instance, the gastrological effect that NSAIDs were likely to have on Mr McCulloch was discussed by the Court, but not whether he, subjectively, would have been willing to endure those side effects if there were even a small chance of alleviating his life-threatening condition. But, it is important that patients’ characteristics and values should not become lost, lest we risk a retreat from the progress made in Montgomery. We might have less cause for concern if the operation of Bolam allows for inclusion of these types of facts, as the judgement in Bilal might suggest; but it is unclear what direction the courts will take post-McCulloch (even if one were to read Bilal and McCulloch as compatible).

Pragmatic concerns

The Court’s reasoning is perhaps borne from a concern that, without the convenience of Bolam delineating the remit of reasonable treatments, the duty to discuss will become too onerous. Indeed, an extensive duty to disclose might be worrisome: it raises the spectre, as signalled by the Supreme Court in McCulloch, of unnecessarily ‘bombarding the patient with information’ ⁷⁷ —a practice which both the courts ⁷⁸ and professional guidance ⁷⁹ take exception to. Moreover, an extensive duty to inform would be burdensome, impractical, and unattainable in the context of an under-resourced, time-poor NHS.

However, it is far from clear that a patient-sensitive understanding of ‘reasonable alternatives’ would bring about the negative consequences alluded to. In fact, similar claims relating to ‘defensive practices’ (an implication of which is increased litigation) featured as arguments against a patient-centric test for information disclosure prior to Montgomery. ⁸⁰ The Supreme Court in Montgomery considered these arguments to be unpersuasive, however, noting that ‘a degree of unpredictability’ in terms of clinician’s practices ‘can be tolerated as the consequence of protecting patients’. ⁸¹ Moreover, in much the same way that there is no evidence, following Montgomery, that doctors have found the new (legally speaking) patient-centric test for materiality unmanageable, it is perhaps unlikely that they would struggle with an interpretation of reasonable alternatives that aligns more closely with the spirit of Montgomery. Indeed, one might expect that proper communication with a patient involves a more holistic and transparent conversation, whereby standard treatments – even if not necessary in the treating clinician’s view – are put forward to the patient.

Perhaps then the difficulty lies in conceptualising what ‘reasonableness’ is and how it should be assessed – particularly regarding the kinds of considerations that can render a treatment option not recommended, or the grounds on which clinicians may permissibly decide not to inform patients about an option. For instance, might a doctor legitimately fail to discuss an option that they do not think will be good for the patient for broader welfare reasons? This raises similar problems to the issues discussed above: if reasonableness is a matter of clinical consideration alone, we get a narrow conception of what is good for patients which fails to capture a lot of what patients themselves may take into account (and indeed, give significant weight to).

Moreover, clinicians will not, unless they have a long-standing relationship with the patient, ordinarily be aware of the factors that will bear on their weighing of different options. And indeed, it would seem very onerous to expect them to have such intimate knowledge of their patients’ situations – more onerous than discussing alternative treatments other than the one(s) they recommend. However, if clinicians are not under a duty to discuss advantages and disadvantages of alternatives other than the one(s) they consider most appropriate, then, it seems unclear how clinicians would be able to gather the information from their patients necessary to tailor information disclosure in the way required by Montgomery. This is especially so if we take seriously the concerns about overstretched clinicians in an under-resourced health service.

Still, the Supreme Court in McCulloch held that incorporating a patient’s subjective circumstances into decisions on reasonable alternatives would create too much uncertainty, or prove to be unworkable in practice. ⁸² Yet its interpretation of what a duty to disclose alternatives would look like seems to lack nuance. We can accept that clinicians should be under a more extensive duty to disclose alternatives, while resisting the implication that such a duty should extend to disclosing alternatives that would be inappropriate and harmful to the patient. We can assess particular cases and agree on a range of alternatives that are reasonable and thus need to be disclosed, and on alternatives that fall outside this range and about which the clinician thus does not need to inform patients, to (eventually) arrive at some criteria for what makes something reasonable. Courts can and often do assess the reasonableness of conduct, and it does not seem inherently harder in this context than elsewhere. The decision then seems to be a missed opportunity to more explicitly preserve some scope for courts in determining the reasonableness of alternatives.

Finally on this note, it does seem significant that Mr McCulloch’s working diagnosis, as agreed by his medical team and stated on his medical notes, was pericarditis. On the evidence, it was agreed by all experts that prescribing NSAIDs is standard treatment for pericarditis. ⁸³ Even in the absence of pain, both Dr Weir and Dr Flapan would have prescribed NSAIDs to Mr McCulloch, and even Dr Bloomfield (whose testimony saved Dr Labinjoh from a finding of negligence on the Bolam test) did not go as far as to say that he, personally, would not have prescribed NSAIDs to Mr McCulloch. ⁸⁴ While there is ambiguity as to whether NSAIDs can treat the underlying inflammation (and thus whether they reduce the risk of tamponade), it could not be said, even in the absence of pain, that NSAIDs would have been a decidedly unreasonable treatment option for Mr McCulloch. Moreover, the argument that doctors need not discuss treatments that they would not recommend is perhaps undermined by the fact that Dr Labinjoh did, in fact, discuss with Mr McCulloch other treatments that she would not recommend, and even considered to be potentially harmful – such as pericardiocentesis (the draining of the fluid around the heart). ⁸⁵ If it were reasonable not to discuss this treatment in McCulloch, then, it seems to open up for a rather restricted duty to inform about alternatives.

The relationship between the right to be informed and the ability to request treatment

As noted earlier, the Lord Ordinary (with whom the Inner House and Supreme Court agreed) held that interpreting clinicians’ duty to disclose information in line with the appellant’s suggestion – that is, in a patient-centric way – would constitute an unwarranted extension of Montgomery. The Supreme Court characterised the issue as one of conflict in the doctor’s role, whereby the law would require a doctor to inform a patient about a medical treatment which the doctor, supported by a responsible body of medical opinion, would not consider reasonable. ⁸⁶ It did not directly consider, though, that imposing such duties to disclose might create further conflict with the position established in English law that a patient’s right to self-determination does not extend to an ability to demand a particular treatment. ⁸⁷ If a particular alternative treatment is discussed with a patient, it might create the impression in that patient’s mind that the treatment is, explicitly or implicitly, on offer.

We might wonder what the purpose is of being informed of a treatment that the doctor will, ultimately, decline to provide, and to which the patient is not, legally, entitled to request. The answer to this lies in part in the importance attached to the right to self-determination, along with other, practical, outcomes (besides demanding treatment) that might accompany its exercise. For instance, the appellants in McCulloch argued that their interpretation of ‘reasonable alternatives’ would not only accord appropriate respect to a patient’s autonomy, but that it would also afford the patient the opportunity to seek a second opinion. The Supreme Court in McCulloch, beyond citing extracts of the judgement in AH v Greater Glasgow Health Board ⁸⁸ – a factually similar case – did not, however, address this matter of second opinions. Rather, we can infer from its lengthy discussion of the conflict doctors might experience in having to disclose treatments that they consider inadvisable, that the Court did not really even consider whether denying the patient the opportunity to ask for a second opinion, or the opportunity to reflect on the alternatives, could constitute an injury in itself. ⁸⁹

While it is understandable that the court would refrain from engaging deeper theoretical issues in the context of a particular case, it would have been helpful to more clearly delineate the different issues at play. While duties to assist – provide treatment, in this case – are limited by assistors’ views about what might be good for a patient, respect for autonomy entitles us to do to ourselves what others do not think is best for us. As the Supreme Court noted in Montgomery, patients are ‘persons holding rights’ to self-determination, rather than ‘passive recipients of the care of the medical profession’ ⁹⁰ and therefore entitled to decide for themselves on the risks they are willing to expose themselves to. ⁹¹ Information provision needs to be such that it facilitates this: it needs to enable patients to also make idiosyncratic choices or choices that we think might be bad for them. There are risks attending to letting clinicians’ professional judgement circumscribe information provision such that only information about options they consider good for the patient must be provided. In particular, the two claims in McCulloch that (1) clinical judgement determines which alternatives should be disclosed and (2) a clinician’s non-disclosure of an option can be reasonable even though other clinicians might disagree with it, when taken together, seems to give too much scope for discretion regarding whether interventions should be disclosed, and limited grounds on which to challenge clinicians’ decision-making.

Thus, and even though in the majority of instances it is likely that patients will simply agree with their clinician’s reasons for rejecting a particular treatment option, encouraging wide disclosure is still necessary and valuable, as it (1) affords patients the respect they deserve by not treating them as ‘medically uninformed’ ⁹² ; (2) might, in some instances, provide patients with the opportunity to reflect more deeply about their treatment options and seek out further advice or a second opinion, and (3) provides the clinician with opportunities to learn things about the patient that will help them make better judgements about which treatment options best promote their interests.