This articles discusses a Supreme Court of Appeal judgment in relation to various tests courts use to determine whether a particular risk should have been disclosed to the patient.
When is a risk of medical treatment material?
By Maria Wilson
Source: De Rebus
The case of Oldwage v Louwrens  1 All SA 532 (C) was discussed in 2004 (Sept) DR 55 in the context of informed consent to medical treatment. Subsequently, on 21 September 2005, the Supreme Court of Appeal (SCA) delivered judgment in the matter of Louwrens v Oldwage (case 181/2004 unreported). This article discusses the SCA judgment in relation to various tests which the courts use to determine whether a particular risk of medical treatment should have been disclosed to the plaintiff. Firstly, I will argue that it is not appropriate to lean heavily on precedent when determining whether a risk in a particular case is material. Secondly, I will argue that when applying the medical opinion/reasonable doctor test to decide whether a risk should have been disclosed, the courts do not adequately take into account all the circumstances and human elements of the particular case.
I will conclude that the application of the subjective patient-centred test promotes a flexible and inclusive approach, in which the likelihood of the risk occurring and medical opinion as to its relevance, are but a few of the factors to be considered when determining the boundaries of a material risk.
The defendant, a surgeon, performed vascular surgery on the plaintiff, Mr Oldwage, following complaints by the plaintiff of excruciating pain in his right leg.
Following the vascular surgery the plaintiff suffered from claudication in his left leg, which prevented him from enjoying the lifestyle he was accustomed to prior to the surgery. The plaintiff argued, inter alia, that the defendant had not warned him of the risk of claudication occurring. As a result there was inadequate consent to surgery and the operation therefore constituted an assault.
The trial court applied the formulation as set out in Castell v De Greef 1994 (4) SA 408 (C) and found in favour of the plaintiff. The court in Castell held that a doctor is obliged to warn a patient of the material risks inherent in the proposed treatment, a risk being material if in the particular circumstances:
• a reasonable person in the patient’s position, if warned of the risk, would be likely to attach significance to it; or
• the doctor is or should reasonably be aware that the particular patient, if warned of the risk, would be likely to attach significance to it (426F-H).
On appeal, however, Mthiyane JA (Mpati DP, Streicher JA, Lewis JA and Ponnan JA concurring) did not apply the formulation as set out in Castell and instead quoted with approval from Richter and Another v Estate Hammann 1976 (3) SA 226 (C) at 232G-H:
‘In reaching a conclusion a Court should be guided by medical opinion as to what a reasonable doctor, having regard to all the circumstances of the particular case, should or should not do’ (25).
The SCA set aside the decision of the court a quo and found in favour of the appellant doctor.
How do the tests referred to above work in practice and what are the practical implications of applying one test as opposed to the other?
The medical opinion/reasonable doctor approach
In the English case of Sidaway v Bethlehem Royal Hospital Governers (1984) 1 All ER 1018 (CA); AC 871 (1985) 1 All ER (HL), a patient was left paraplegic after a cervical cord decompression. The question was whether the doctor should have disclosed to his patient the risk of spinal cord damage with consequent paraplegia as well as damage to the nerve roots.
The court a quo and on appeal adopted the English ‘Bolam’ test with minor variations (from Bolam v Friern Hospital Management Committee  2 All ER 118). In effect, the court held that a doctor should act ‘in accordance with a practice accepted at the time as proper by a responsible body of medical opinion, even though other doctors adopt a different practice’ (1030h).
Medical experts in the Sidaway case testified that the risk of damage to nerve roots from a cervical cord decompression was 1 to 2%. They advised that it was in accordance with accepted medical practice that the likelihood of harm was material and should be disclosed (1035 h-j).
However, their evidence indicated that the risk of paraplegia was less than 1% and therefore not material as it was too remote. There was therefore no need to disclose it, even though it was a more serious risk than the damage to the nerve roots (1035h–j).
The courts a quo and on appeal were guided by the evidence of the medical experts as to the likelihood of the risks occurring and their consequent materiality. It was, therefore, held that the risk of paraplegia, although very serious, need not be disclosed (1028h–j; 1035h–j).
The approach in Sidaway is not followed in many other jurisdictions which adhere to a patient-centred approach to informed consent. One such is Australia.
The subjective patient-centred approach
In the Australian case of Rogers v Whitaker (1993) 67 ALJR 47, a patient was left totally blind in her left eye due to a remote risk eventuating following eye surgery on her right eye. When considering whether to have the surgery, the patient questioned her doctor closely as to the possible complications in respect of both eyes. The likelihood of the risk eventuating was 0,007% (1 in 14 000).
The doctor argued that once it was found that there was a responsible body of professional opinion which supported the view that a warning was not required, that was the end of the matter. However, the court found that the matter was not to be determined exclusively by reference to the current state of responsible and competent professional opinion and practice at the time, although both are relevant considerations. This view was also held in the Castell case:
‘Expert medical evidence would be relevant to determine what risks inhere in or are the result of particular treatment (surgical or otherwise) and might also have a bearing on their materiality but, in the words of the Supreme Court of Canada in Reibl v Hughes [114 DLR (3d) 1], “this is not a question that is to be concluded on the basis of expert medical evidence alone” ’ (426H).
The court in Rogers adopted substantially the same test as in Castell.
‘A risk is material if, in the circumstances of the particular case, a reasonable person in the patient’s position, if warned of the risk, would be likely to attach significance to it or if the medical practitioner is or should reasonably be aware that the particular patient, if warned of the risk, would be likely to attach significance to it’ (52).
The court held that, although the likelihood of the risk occurring was very remote, the doctor was nevertheless obliged to tell the patient of this risk, because he should have been reasonably aware that the particular patient, if warned of the risk, was likely to attach significance to it. In so doing the court criticised the medical opinion/reasonable doctor approach:
‘One consequence of the application of the Bolam principle to cases involving the provision of advice or information is that, even if a patient asks a direct question about the possible risks or complications, the making of that inquiry would logically be of little or no significance; medical opinion determines whether the risk should or should not be disclosed and the express desire of a particular patient for information or advice does not alter that opinion or the legal significance of that opinion’ (50).
The court also emphasised that it is the court and not medical opinion which decides the question as to the materiality of the risk in question. The above approach was approved in the later Australian case of Rosenberg v Percival  HCA 18.
In short, the subjective patient centred approach, as applied in Rogers v Whitaker and Castell v De Greef, provides that
• professional practice, whilst relevant, is not conclusive;
• the responsibility for deciding the content of the doctor’s duty of care rests with the court and not with professional medical colleagues; and
• in deciding what risks need to be disclosed to the patient, the court must have regard to the reasonable patient and also to the particular patient in the specific circumstances of the case at hand.
The approach in Louwrens v Oldwage
Although the court referred to the Castell case and approved the medical opinion/reasonable doctor test as set out in the Richter case, it appears from the judgment that the court did not apply either of the tests.
The court found, based on the evidence of the defendant’s medical expert, that the likelihood of the risk of claudication occurring was 2% (the court did not accept the evidence of the plaintiff’s expert that the risk was 4%). However, the court did not then go on to consider the views of the medical experts regarding the materiality of the risk based on the likelihood of it occurring:
‘If there was only a two percent chance of [the risk] occurring then the risk to the plaintiff was so negligible that it was not unreasonable for the defendant not to mention it’ (25),
‘[i]n my view of the evidence, the likelihood of [the risk occurring] … was so negligible that no duty arose on the defendant to mention it’ (25).
The court referred to the Richter judgment in support of its finding that a remote risk need not be disclosed (25). However, in its judgment the court did not consider exactly what was meant by ‘remote’ in the circumstances of the Richter case.
In the Richter case the doctor performed a unilateral nerve block, which has a slight chance of affecting the side on which the block is performed. However, the problem in this particular case was that both the patient’s sides of were affected, which the medical experts found was ‘a very unusual result’, ‘very uncommon in any man’s experience’ and, importantly, ‘could not have been expected by any stretch of the imagination’ (230F). The court therefore held that the risk was remote and need not be disclosed.
I submit that the difference between a risk of 2% and a risk that cannot even be imagined is not insignificant. Further, even if the risk had been 2%, the Richter case reflects the opinion of medical experts regarding remoteness and materiality in 1976 and in a particular set of circumstances.
South Africa has come a long way since 1976. The Constitution requires respect for autonomy and self-determination, and that people have the opportunity to make informed decisions about their bodies.
I submit therefore that medical opinion as to the importance of a risk based largely on the likelihood of it occurring, should not be the major determinant of whether a risk is material. Of equal, if not more importance, should be what risks the reasonable person expects to be informed of and whether the particular patient had expressed concern to the doctor about the risks of the proposed treatment.
The court did not consider the above factors when it discussed at length the likelihood of the risk occurring and whether this was a material likelihood. Finally, the court concluded that the risk was not material because its percentage risk was too small (25).
The court went on to hold that the harm the plaintiff had suffered had not in any event been caused by the risk and the harm had not been caused by the defendant (25).
Only then did the court remark, almost as an aside, that the defendant’s expert had said that if the risk occurred ‘it can be rectified by a minor operation’ (25). However, the court did not refer to the view of the plaintiff’s experts in this regard, nor did it discuss, analyse or contextualise the relevance of the above statement in relation to its argument that the risk in question was not material. Therefore, I submit that the weight of the statement by the defendant’s expert must be regarded as very small in the decision of the court whether the relevant risk was material and should have been disclosed.
The relevant findings of the cases discussed above are represented in the diagram alongside.
The subjective patient-centred approach
The decision in Rogers v Whitaker, discussed above, recognises that whilst statistics and medical expert opinion are relevant, the patient is also a factor in deciding whether a risk is material.
The subjective patient-centred approach does not rely exclusively on medical opinion as to the materiality of the risk, nor does it give undue weight to the percentage likelihood of the risk occurring.
This approach also takes cognizance of our changing society and the prominence given to human rights, as well as to human responsibility, the reverse side of individual rights. It highlights the respect we afford people when we recognise their ability to choose and give them the information that is necessary for them to accept the responsibility of making an informed choice.
The medical opinion/reasonable doctor approach
The judgments in the leading cases discussed above focused on medical expert opinion regarding statistics when deciding whether a risk should be disclosed to a patient.
From a reading of the above judgments it appears that medical practice as to the disclosure of risks is often swayed by the statistical likelihood of the risk occurring. I submit that in practice this approach tends to sideline the relevance of the circumstances of the particular case and the human elements and relationships involved.
I submit that it would be more objective and inclusive for the court to determine analytically, in the light of the medical evidence as to current practice, what a reasonable person would expect to be informed of and what knowledge the particular patient actually requires.
Louwrens v Oldwage
The decision in Louwrens v Oldwage stands on its own in the diagram below because it appears to some extent to have relied on precedent for its decision.
The judgment was disappointing. The Castell, Rogers and Sidaway judgments reasoned and argued the facts and law in detail, making them a pleasure to read and easy for future courts to understand, apply and, importantly, to avoid misinterpretation.
While the court did not apply the subjective patient-centred approach, it did not overrule it. There is, therefore, as yet no binding judgment by the SCA as to what the correct approach to determining the boundaries of a material risk to medical treatment may be. In the absence of such a judgment, courts are still free to follow the patient-centred approach, which was extensively and cogently argued in Castell v De Greef.
Maria Wilson BA (Unisa) LLB (UCT) LLDip (Wolverhampton, UK), a registered general nurse, midwife, psychiatric nurse and a first-aid instructor, is an attorney practising for her own account in Johannesburg.
|The medical opinion/reasonable doctor test|
|The likelihood of|
1-2% of the risk occurring is a
|Sidaway v Bethlehem (1985) 1 All ER (HL)|
|The court should be guided by medical opinion as to what is a material risk||PRECEDENT|
|The likelihood of less than 1% of a risk occurring is too remote to be material||A remote risk is not material – Richter v Estate Hammann|
|Sidaway v Bethlehem|
|A very unusual risk that could not have been expected by any stretch of the||The likelihood of 2% of the risk occurring is too remote to be material.|
|imagination is remote and not material||Louwrens v Oldwage(SCA; 21-9-2005)|
|Richter v Estate Hammann 1976 (3) SA 226 (C)|
The subjective patient-centred test
|Medical opinion is relevant in deciding whether a risk is material, but decisive is the reasonable and the particular patient in the specific case.|
|The risk of 0,007% is material in the circumstances of this case.|
|Rogers v Whitaker (1993) 67 ALJR 47|