Sidaway v Board of Governors of the Bethlem Royal Hospital [1985] UKHL 1

Mrs Sidaway underwent cervical spine surgery and suffered partial paralysis, a recognised but small risk of the procedure. She claimed her surgeon negligently failed to warn of this risk. The House of Lords held that, applying the Bolam test, there was no negligence and rejected a general doctrine of informed consent.

Facts

The appellant, Mrs Amy Doris Sidaway, suffered long-standing neck and arm pain linked to deformities in her cervical spine. In 1960 the neuro-surgeon, Mr Murray A. Falconer, successfully operated to remove a disc between the fifth and sixth cervical vertebrae and fuse them, ultimately relieving her pain.

Mr Falconer reviewed her annually until 1970. In 1973 he wrote to enquire after her condition; she reported renewed and worsening pain in the right arm and shoulder and in the left forearm. Following further consultation, she was admitted to the Maudsley Hospital on 11 October 1974. A myelogram on 17 October 1974 revealed a partial block and narrowing at the C4/5 level, consistent with pressure on a nerve root.

Mr Falconer decided to operate. On 29 October 1974 he performed a laminectomy of the fourth cervical vertebra and related procedures to free the fourth cervical nerve root. The trial judge described the operation as follows:

“The operation consisted of a laminectomy of the fourth cervical vertebra and a facetectomy or foraminectomy of the disc space between the fourth and fifth cervical vertebrae. A laminectomy is an excision of the posterior arch of the vertebra. It gives the surgeon access to the foramen or channel through which nerves travel from the spine laterally. Randomly placed in the foramina, running alongside the nerves, are small blood vessels known as the radicular arteries. These supply blood to the cord and are extremely vulnerable because of (a) their size and (b) the unpredictable nature of their siting. In one foramen, there may be one, two or more radicular arteries. Their rupture or blockage may cause damage to the cord by depriving it temporarily or permanently of its blood supply at the relevant level. At the operation, Mr. Falconer freed the fourth cervical nerve root by removing the facets, or small bony protuberances, from the fourth verbebra and used a dental drill to free the nerve within the foramen.”

The specific recognised risks included damage to a nerve root and damage to the spinal cord, the latter carrying a small (<1%) but serious risk of partial paralysis. One surgeon estimated the overall risk of either type of damage at between one and two per cent.

Mr Falconer’s diagnosis, decision to recommend surgery, and performance of the operation were all found to have been reasonable and skilful. However, the operation resulted, without negligence in performance, in damage to the spinal cord and severe, permanent partial paralysis. Damages were agreed at £67,500 subject to liability.

Mr Falconer died before trial, so he could not give evidence. The medical records did not clearly record what he had told the appellant. The trial judge rejected Mrs Sidaway’s assertion that she had received no warning and, relying on evidence of Mr Falconer’s usual practice, found that he had probably explained the nature of the operation and mentioned the possibility and consequences of damage to a nerve root, but had not referred to the risk of spinal cord damage or to the fact that the operation was one of choice rather than strict necessity:

“As to the risks, I think it is probable that he mentioned the possibility of disturbing a nerve root and the consequences of doing so, but I am satisfied that he did not refer to the danger of cord damage or to the fact that this was an operation of choice rather than necessity.”

All the neuro-surgeons who gave evidence agreed that there were specific risks beyond those inherent in all surgery under general anaesthetic, and that some warning of specific risks should be given. They also agreed that the extent of any warning depended on medical judgment, and that responsible practice could include not mentioning the remote risk of paralysis where it was thought contrary to the patient’s interests to do so. The appellant’s own expert, Mr Uttley, accepted that a surgeon might properly decide not to frighten a patient by referring to paralysis and that such a decision could accord with practice accepted by a responsible body of competent neuro-surgeons.

Mrs Sidaway had signed a standard consent form stating that the nature and purpose of the operation had been explained by a junior doctor, who in turn made clear that he left specific warnings about risk to Mr Falconer. Her claim was based solely on an alleged failure to warn of the risk of spinal cord damage.

Issues

Primary legal questions

• Whether, as a matter of English law, a doctor’s duty of care includes a duty to warn a patient of risks inherent in proposed treatment, particularly surgical treatment.
• Whether the standard of that duty is governed by the Bolam test (practice accepted as proper by a responsible body of medical opinion) or by an “objective” patient-centred standard akin to the North American doctrine of informed consent.
• Whether failure to warn of a small but serious risk (here, less than 1% risk of spinal cord damage causing partial paralysis) constituted negligence in the circumstances.
• Whether Mrs Sidaway had an independent cause of action based on a specific right to be informed, distinct from negligence in the Bolam sense or trespass to the person.

Competing standards for disclosure

The House examined two competing approaches:

• The Bolam standard: In Bolam v. Friern Hospital Management Committee the duty to warn of risks was treated as one aspect of the doctor’s overall duty of care, with breach determined by whether the doctor acted in accordance with a practice accepted as proper by a responsible body of medical opinion.
• The “informed consent” model: As developed in Canterbury v. Spence and supported in Canada by dicta in Reibl v. Hughes, under which a doctor must disclose “material” risks defined from the perspective of a reasonable patient, subject to limited therapeutic exceptions, with the court (not the medical profession) setting the standard.

Judgment

Outcome

The House of Lords unanimously dismissed the appeal and affirmed the Court of Appeal’s order. No negligence was established. Costs orders were made in accordance with the Legal Aid Act 1974; the appellant’s costs were to be taxed under Schedule 2 and the respondents’ costs were to be paid out of the Legal Aid Fund.

“That the said Order of Her Majesty’s Court of Appeal of the 23rd day of February 1984 complained of in the said Appeal be, and the same is hereby, Affirmed and that the said Petition and Appeal be, and the same is hereby, dismissed this House.”

Lord Scarman (dissenting in principle, concurring in result)

Lord Scarman accepted that, on the evidence, the appellant could not prove negligence even on his preferred legal test, mainly because Mr Falconer had died and there was no direct evidence of his assessment of the patient or of the detailed content of his warning. He expressly concluded:

“After an anxious consideration of the evidence I do not find it possible to say that it has been proved that Mr. Falconer failed in his duty when he omitted – as we must assume that he did – to warn his patient of the risk of injury to the spinal cord.”

However, Lord Scarman considered that the trial judge and Court of Appeal had erred in law in treating the duty to warn purely as a Bolam issue. He argued that a patient has a right of “self-determination” and that the law should recognise a distinct duty to warn of “material” risks as part of the duty of care, not solely controlled by professional practice.

Drawing heavily on Canterbury v. Spence and related commentary, he endorsed the core propositions that:

• Every adult of sound mind has the right to decide what shall be done with his or her own body.
• That right requires that consent be based on information about material risks of proposed treatment.
• A risk is material if a reasonable person in the patient’s position would be likely to attach significance to it in deciding whether to undergo the treatment.
• There should be a recognised therapeutic exception where disclosure would, on a reasonable medical assessment, be seriously detrimental to the patient’s health.

He summarised his conclusion as to English law:

“To the extent that I have indicated I think that English law must recognise a duty of the doctor to warn his patient of risk inherent in the treatment which he is proposing: and especially so, if the treatment be surgery. The critical limitation is that the duty is confined to material risk. The test of materiality is whether in the circumstances of the particular case the court is satisfied that a reasonable person in the patient’s position would be likely to attach significance to the risk. Even if the risk be material, the doctor will not be liable if upon a reasonable assessment of his patient’s condition he takes the view that a warning would be detrimental to his patient’s health.”

Applying this test, he held that the appellant had failed to show that the risk of cord damage in this case was proved to be “material” in the relevant sense, nor could the court assess Mr Falconer’s clinical judgment about the impact of a warning on this particular patient. For want of proof, the claim failed even on his more patient-centred approach.

Lord Diplock (majority)

Lord Diplock emphasised the paucity of factual evidence about the appellant’s personality and about the precise discussion that occurred. He reiterated that all medical experts, including the appellant’s own, agreed that a responsible body of neuro-surgical opinion would have done as Mr Falconer did, i.e. warn about nerve root damage but not explicitly about spinal cord paralysis.

He treated the doctor’s duty of care as a single comprehensive duty covering diagnosis, advice (including warnings) and treatment, all governed by the Bolam test. He rejected the doctrinal foundation of “informed consent” in English law, noting that English medical liability has historically sounded in negligence rather than trespass. He criticised the attempt, seen particularly in Canadian case law, to graft a consent-based, battery-rooted concept onto negligence, arguing that it is inconsistent with the traditional principles of tort and impractical in practice.

In his view, deciding what risks to disclose, and how to describe them, is as much an exercise of professional skill and judgment as diagnosis or choice of treatment. Expert evidence as to responsible practice must therefore be the benchmark, and the Bolam test should be applied to warnings in the same way as to all other aspects of a doctor’s duty.

On that basis, since there was clear evidence of a responsible body of opinion supporting the level of warning given, there was no negligence.

Lord Keith of Kinkel (majority)

Lord Keith agreed with the reasoning and conclusion of Lord Bridge of Harwich and would dismiss the appeal for those reasons.

Lord Bridge of Harwich (majority)

Lord Bridge accepted that the patient is entitled to decide for herself whether to undergo treatment, especially major surgery, and that this underlies the attraction of the “informed consent” doctrine. He carefully reviewed Canterbury v. Spence and its “prudent patient” test, recognising its logical force but finding it impracticable in English conditions.

He identified three principal objections to adopting Canterbury as English law:

• It gives insufficient weight to the practical realities of the doctor–patient relationship, particularly the need to tailor information to individual patients and to avoid over-emphasising remote risks.
• It would unrealistically limit expert medical evidence to primary medical facts and exclude evidence of professional practice and judgment about disclosure, depriving courts of vital guidance.
• The “reasonable patient” test is so imprecise that it would render litigation outcomes unpredictable.

He therefore rejected a wholesale “objective” standard but accepted, with some nuance, that a doctor’s duty to warn is still ultimately a duty of care to be assessed primarily, though not exclusively, by reference to professional practice under Bolam. He endorsed the reasoning of Laskin C.J.C. in Reibl v. Hughes to the extent that expert evidence on risks and their materiality is not the sole determinant, while insisting that decisions about adequacy of disclosure are primarily clinical judgments.

Lord Bridge suggested that, exceptionally, even where a responsible body of medical opinion would not warn, a court might still find non-disclosure negligent if the risk was very substantial and of grave consequences, giving as an example the ten per cent risk of stroke in Reibl v. Hughes. In such a case, a doctor could “hardly fail” to warn, absent cogent clinical reasons not to. That left open a narrow, judge-controlled limit on professional practice in extreme cases.

However, in Mrs Sidaway’s case, the risk of spinal cord damage comparable in seriousness to what occurred was less than one per cent, and no finding identified a specific fraction of that already small risk. The appellant’s expert accepted that non-disclosure of such a risk accorded with responsible practice. On these facts, applying Bolam, the respondents had a “complete defence” and there was no breach of duty.

Lord Templeman (majority)

Lord Templeman focused on the practicalities of explanation and patient decision-making. He accepted the trial judge’s finding that Mr Falconer had likely given a simple explanation of the operation and mentioned the possibility of nerve root damage but not spinal cord damage. In his view, a proper explanation of the nature of a major operation involving bone removal and work close to the spinal cord was sufficient to alert a reasonable patient to the possibility that serious harm might occur, even without a specific reference to paralysis.

He considered that, had Mrs Sidaway asked further questions, she should have been told that there was an aggregate risk of between one and two per cent of some damage to the spinal cord or nerve root, with consequences ranging from irritation to paralysis. But he saw this as reinforcing what was already obvious from a correct general explanation, especially when coupled with the fact that the surgeon was recommending the operation, which implied that he considered the risk suitably remote.

Lord Templeman accepted that a doctor must draw a patient’s attention to any danger which is special in kind or magnitude, or special to the particular patient, illustrating this with the Canadian case of Reibl v. Hughes where there was a four per cent risk of death and ten per cent risk of stroke from the operation designed to prevent stroke.

He emphasised that the doctor’s obligation is to explain the arguments for and against a major operation and the possibilities of benefit and danger, but that the doctor is not required to provide every possible detail nor to overwhelm the patient with information he or she may be unable to evaluate. He saw the doctor–patient relationship as rooted in an implied obligation to act in the patient’s best interests, which includes providing “information which is adequate to enable the patient to reach a balanced judgment” while also avoiding harm, including psychological harm, that might result from excessive or poorly framed disclosure.

He concluded that, on the judge’s findings, adequate information had been provided, and the omission of a specific warning of cord damage did not constitute negligence.

Implications

This decision established that, under English law at the time, the scope of a doctor’s duty to warn patients of risks of treatment was primarily governed by the Bolam standard: a doctor is not negligent if acting in accordance with a practice accepted as proper by a responsible body of medical opinion, and this applies to advice and warnings as well as diagnosis and treatment.

The majority rejected the wholesale importation of a North American-style doctrine of informed consent based on a judge-set, “objective” prudent patient test. They affirmed the central role of professional judgment and expert medical evidence in determining which risks must be disclosed, while acknowledging that in extreme cases a court might override professional opinion where a very substantial risk of grave harm is plainly one that a patient must be told about.

Lord Scarman’s partial dissent left a significant doctrinal marker by articulating a material risk test and emphasising patient self-determination, although his approach did not command the support of the majority. His speech highlighted the tension between respect for professional judgment and the patient’s right to decide, a tension that would influence later debate on medical negligence and consent.

On the facts, the case illustrates the evidential difficulties when a key witness (the treating surgeon) is deceased and contemporaneous records are limited, leading the courts to infer practice from general evidence and to apply the governing legal test to that reconstructed practice. The result was that a patient seriously injured by an accepted but small risk had no remedy, because the failure to warn of that risk accorded with responsible medical practice as understood at the time.

The case remains an important authority in the law of medical negligence on the standard for disclosure of risks and on the limits of judicial interference with clinical judgment as to what to tell patients about the dangers of proposed treatment.

Verdict: The House of Lords affirmed the Court of Appeal’s order and dismissed Mrs Sidaway’s appeal, holding that there was no breach of duty by the surgeon in failing to warn specifically of the small risk of spinal cord damage; no liability in negligence was established.

Source: Sidaway v Board of Governors of the Bethlem Royal Hospital [1985] UKHL 1