The so-called evidence gap between proving breach of duty of care and proof of injury, and then demonstrating a causal link between them. The meaning of “causation” in law was also highlighted, sometimes it means simply the historical facts of events, and i.e. did the defendants’ actions have anything at all to do with the subsequent events.
At other times lawyers mean the consequential connection between two events i.e. the administered medication resulted in his death and not the ischemic heart disease from which he also suffered[3]. In Gregg the equally difficult question of can loss of chance be considered an injury for which a head of damages can be recovered in medical negligence cases?
First I shall look at how causation is determined in law and then how this was used in Gregg and Chester. In order to prove causation as I have alluded to above requires proof of a breach of duty of care and that injury occurred as a consequence of that breach. In cases of non-disclosure of risk the claimant is also required to prove that if they had been made aware of the risk they would have taken steps to avoid it. The “but for test” is the main way that causation is determined in law. In medical negligence cases, the claimant must prove that but for the doctor’s negligence no injury would have occurred. An example of this is the case Barnett v Chelsea[4], here three men ingested tea later found to be contaminated with arsenic, and they subsequently began vomiting. When they later presented to the casualty department of the defendants’ hospital the casualty officer Dr Banerjee refused to see them. Neild J held that Dr Banerjee had breached his duty of care by not seeing the men; however the question causation was not made out. If he had seen the men he would not have made the correct diagnosis due to the obscurity of symptoms and the rarity of arsenic poisoning therefore the only antidote could not have been administered in time to save their lives. Here then the injury (death) was caused by the arsenic not as a consequence of the breach. The case of Wilsher v Essex[5] the court of appeal by a majority verdict upheld a claim by the plaintiff that the doctors’ negligence led to retrolental fibroplasia (RLF), a condition resulting in Martin Wilsher’s near blindness. The House of Lords would later reverse this decision on the grounds of causation. It was claimed that the doctors’ negligent insertion of a venous catheter rather than an arterial one had led to dangerously high oxygen levels resulting in RLF. The Lords felt that Martin Wilsher had experienced multiple possible causes of RLF in his early life, and the necessary consequential connection between high oxygen levels and RLF was not successful therefore causation could not be proved[6]. Conventional causation principles when rigorously applied will result in many failed cases in negligence, due to failure to bridge the evidence gap.
The Law allows for the relaxation of conventional causation principles, an early example of this was in McGhee v National Coal Board[7], and here McGhee’s claim was that the lack of provision of washing facilities had prolonged his exposure to brick dust which caused dermatitis. Due to lack of medical knowledge regarding how dermatitis developed a causal link could not be proved. The House of Lords concluded that Coal Boards breach of duty materially contributed to McGhee contracting dermatitis his claim was therefore made out in negligence. The evidence gap could therefore be bridged in individual cases. Another case cited in judgement in McGhee was Bonnington Castings Ltd v Wardlaw[8], here the claimant had worked in an environment containing silica dust and contracted pneumoconiosis. There were two sources of silica dust; his employers accepted they had breached their duty of care by not protecting him from the dust from the swing grinders. The second source the pneumatic hammers one of which Wardlaw operated could not be made safe. His employers could not be held liable for dust from this source. The House of Lords again concluded that the dust from the swing grinders materially contributed to him developing pneumoconiosis and his claim succeeded. Another excursion from conventional causation principles occurred in Fairchild v Glenhaven Funeral Services and others[9], here the claimant had worked for two employers who had both negligently exposed him to asbestos particles from which he had contracted mesothelioma a cancerous condition of the pleura. There was no method of proving from which asbestos he had contracted his disease. The court of appeal rejected their claim as the but for test failed to prove on the balance probabilities their mesothelioma would not have occurred but for the breach by either employer. The Lord Bingham of Cornhill stated that “if there had been only one tortfeasor the claim would have been successful and he would be entitled to recover. However as there were two who had breached he was entitled to recover from neither, because he was required to prove what is scientifically un-provable. If the mechanical application of generally accepted rules leads to such a result, there must be room to question the appropriateness of such an approach in such a case”[10]. Also Lord Nicholls of Birkenhead agreed and added “any other outcome would be deeply offensive to instinctive notions of what justice requires and fairness demands”. The House of Lords granted the appeal, while setting clear limits on how these rules may be applied in future cases. Clearly then the but for test remains the primary method of proving causation, but the use of both binding legal authority and persuasive legal authority such as notions of justice and fairness may be used to justify relaxation of causation principles in specific cases.
If a breach of duty leads to loss of a chance of a specific outcome can this be recoverable in damages. In non medical cases it is long established that this can be so, for example Little Miss Chaplin was entitled to compensation for loss of chance to be crowned the most beautiful girl in Battersea[11]. Quantification of damages will depend on the value of the “prize” and percentage chance lost. In medical cases the situation is more problematic. In Hotson v East Berkshire Health Authority[12], The Claimant fell and injured his hip, on his first presentation to casualty the correct diagnosis of a fracture to his epiphysis was missed. Five days later the correct diagnosis was made and appropriate treatment instigated. The plaintiff would later develop avascular necrosis of his femoral head leading to disability and inevitably to osteoarthritis. The claim was made that the breach of duty of care in the delay in making the correct diagnosis and starting the correct treatment lead to the loss of chance of recovery without developing avascular necrosis. The chance was estimated at 25% by the plaintiff’s expert witness. The House of Lords was not persuaded that he had proved that the consequential connection between the delay in diagnosis and him developing avascular necrosis of his femoral head had been made. His Lordships however did not answer the questions (1) could loss of chance be a compensatable head of damages in its own right? And (2) how would the damages be quantified?
The proof that consequential injury due to breach of duty of care is of the utmost importance in medical negligence, how then did the House of Lords deal with Chester and Gregg. First I shall briefly review the acts in each case.
Ms Chester had suffered years of back pain and had consulted an eminent neurosurgeon Dr Afshar who recommended surgery. He failed to inform Ms Chester of a 1-2% chance of paralysis inherent in the procedure. He subsequently performed the surgery expertly but the complication of paralysis materialised. Ms Chester made claim on the basis that had she been told of the risk she would not have had the operation on that day. The original trial judge was persuaded that Ms Chester would indeed have avoided the risk on that day and if she had then had the operation on another day performed by Dr Afshar or by another surgeon the risk was so small that it would not have materialised. Therefore using a narrow interpretation of the but for test causation was proved and her claim was successful. The House of Lords unanimously rejected this interpretation of causation; they felt she had not made out her case on conventional causation principles. Their Lordships were then split 3:2 deciding to dismiss the appeal and allow Ms Chester to recover substantive damages. Their Lordships were not invited to decide if Ms Chester would have avoided the risk for ever, they were asked would she have avoided the risk for that day. The dissenting Law Lords (Lord Bingham and Lord Hoffmann) felt that she would have been exposed to the same risk on another day; therefore the injury was not caused by a failure to warn her of the risks. The majority (Lord Steyn, Lord Hope and Lord Walker) made reference to Fairchild and the relaxation of conventional causation principles. They therefore relied heavily on policy considerations in their decision to allow substantive damages to be recovered. The Policy considerations were the right to autonomy and dignity, desirability of the doctrine of informed consent and the notion of justice. Lord Hope felt “the function of the law is to enable rights to be vindicated and to provide remedies when duties have been breached. Unless this is done the duty is a hollow one, stripped of all practical force and devoid of all content”[13].
Looking now at Gregg, Malcolm Gregg presented to his GP Dr Scott on another matter but was noted to have a lump under his axilla; Dr Scott considered the lump to be a lypoma a benign fatty swelling. Some time later Mr Gregg presented to another GP who also thought the lump to be benign but referred him for a biopsy. The lump was found to be a Non Hodgkin’s Lymphoma (NHL) requiring chemotherapy. Treatment was started some nine months after first presentation to Dr Scott, by this time the tumour had spread to the pectoral region. His chance of cure equated to 10 year survival was 25%. His original survival chances when seen by Dr Scott were estimated at 42% the background risk of all patients irrespective of history. Mr Gregg sued for damages on the basis of loss of chance of survival; he put forward two arguments (1) Dr Scott’s breach of duty of care led to a delay caused injury namely the spread of his cancer, resulting in a reduction in life expectancy. (2) Reduction in chances of survival should be a compensatable head of damages in its own right.
At the time of the House of Lords judgement and his prognosis of cure had varied considerably during that time. Following relapse estimates of chances of survival had been almost nil. However by 2005 he had been disease free since 1996 and estimates of survival had greatly improved. The House of Lords rejected his appeal by a majority of 3:2. It is important to look at some of the comments of the majority, Lord Hoffmann questioned if the injury could be attributed to the defendant and also what would be quantification of damage attributed to Dr Scott? He rejected the notion that the injury could be used as the “hook” upon which to hang a claim for damages. Also as causation could not be proved by conventional means, he felt that injury could not be used to allow damages to be awarded.
Lady Hale considered that progress of disease was affected by Mr Gregg’s genetic makeup and other factors which could not be attributed to the defendant. She also raised the question of quantification, commenting that the defendant would be entitled to credit for the chance that a particular outcome would not have occurred. Thus in almost all clinical negligence cases the claimant would recover something, but never everything and the defendants liable for large sums of damages. She felt that denial of a remedy to Mr Gregg was a price worth paying for the certainty of balance of probability approach to causation. Lord Philips commented that a personalised estimate of his chances of cure differed from the estimates given in court. Also it was now nine years since Dr Scott’s negligence, and Mr Gregg remained alive and well. He therefore felt this was an unsuitable case for introducing into the law the right to recover damages for the loss of a chance of cure. It has been argued that there can be no recovery of damages if the outcome is not known[14]. This non fixed nature of the case and the fact that no final conclusion had been reached did not help Gregg’s claim. In both Fairchild and McGhee the court inferred rightly that the outcome was caused by the defendants’ negligence but damages are awarded for the outcome itself. This is at odds with loss of chance cases from non medical areas, damages were awarded on calculations based purely the loss of chance of a particular outcome[15]
Clearly then legal authority and policy considerations were used to inform the House of Lords’ judgements in both cases.
In Chester the extent to which the House of Lords appeared to favour Ms Chester is a little alarming, on the face of it. It would appear they have gone to extreme lengths to support her claim. They appeared to feel that the risk that a professional who had caused no injury may have to pay damages to a claimant was a price worth paying to uphold the principles of autonomy and the doctrine of informed consent. This view is supported by Prof Tony Honoré who commenting on the Australian case of Chappel v Hart concluded that the doctor who has failed to warn should in effect indemnify their patient against injury[16], in fact this commentary was quoted with approval by Lord Steyn in his judgement in Chester. Can these doctrines be of such importance to the public? Certainly surgeons may feel that they will be required to invest a large amount of time and effort both to perform an operation and to adequately inform about the procedure prior to it[17]. This would undoubtedly benefit the public and would be in accordance with guidance from professional bodies such as the Royal College of Surgeons’ guidance “Good Surgical Practice” (2002). Should I be persuaded that despite my initial concerns, the promotion of informed consent is good? Similar developments in informed consent in the United States, has led to increased liability of the doctor and some have argued that this had been the purpose of its development[18]. Such policy would be helpful to claimants especially if injuries were due to medical accidents rather than doctors’ negligence, because under this new era they would be more likely to succeed. In England such an expansion of medical liability would deplete the public purse, if every patient who received an unfavourable outcome were entitled to recover a head of damages. This could result in deserving cases receiving reduced amount of damages due to lack of resources. What of non medical areas here the doctrine of informed consent also applies, in a recent solicitors’ negligence case. The Court of Appeal in Paul Davidson Taylor v White[19], the claimant brought a claim against a solicitors for failure to advise him of the effect of a notice to quit. Both Lord Justice Ward and Lady Justice Arden considered Chester v Afshar but felt it did not establish a new general rule in causation, rather it held a special place in medical negligence cases. The Court of Appeal therefore dismissed the appeal on the grounds that causation was not proved. This I am sure was to the relief of many non-medical professionals.
In Gregg v Scott policy considerations regarding justice and loss of chance were considered by Lady Hale, but here the possibility of increasing the number of claims dissuaded her from supporting Gregg’s claim. Also his survival for nine years could mean he was soon to be considered cured and this made Lord Phillips feel that it was difficult to justify moving away from normal causation principles. It should also be remembered that the majority in Gregg were critical of how the case was framed. They felt the use of 10 year survival was helpful for the medical profession but made quantification difficult in law. There were also a number of other heads of damage which he could have used, namely loss of earnings, increased pain and suffering due to spread of disease, earlier need for radical treatment. There is no simple actuarial route for assessing pre- tort life expectancy and estimates of post –tort life expectancy will be affected by personal attributes and also the actual experience of the claimant will affect estimates of post-tort life expectancy[20]. Ultimately his good personalised response to treatment and his attempt to only try for the more lucrative “survival” injury may have reduced the court’s sympathy for him.
Gregg may still ask why the courts did not look more favourably on him, as the breach in duty of care was precisely that which the law had directed the GP to protect him from. The GP had a duty to preserve any chance of survival or prevent any loss of chance irrespective of cause[21]. Was the duty of the doctor to inform in Chester, more important than a duty to protect Gregg’s future chances?
Could Chester have argued that the breach by Dr Afshar lead to a loss of chance of a second opinion. Alternatively some have argued that the judgement of the court of Appeal had been the correct decision in Chester that her injury had resulted as a consequence of Dr Afshar’s breach and negligence was made out by conventional causation principles[22].
Both cases however leave some residual issues (1) the cases of McGhee and Fairchild treat the meaning of causation as the consequential relationship between the breach and their injury. The difficulty with proving “what was un-provable”, in the interests of justice led to the courts to relax conventional causation principles. These cases were then used in Chester here the meaning of causation was simply one of historical fact. The result appears that some practitioners will in the future pay claimants for injuries they have not caused. (2) No clear guidance has been given from the House of Lords regarding when relaxation of normal causation principles is appropriate in the future. (3) Lord Hope’s comments in Chester that duty would be a worthless notion if the law did not provide a remedy for its breach. I do not feel that doctors would consider they had no valid duty to their patients in such a case.
[1] Chester v Afshar [2005] 1 AC 134.
[2] Gregg v Scott [2005] UKHL 2.
[3] Jane Stapleton, “Occam’s razor reveals an orthodox basis for Chester v Afshar, LQR 2006, 122(Jul), 426-448.
[4] Barnett v Chelsea & Kensington Hospital Management Committee – [1968] 1 All ER 1068
[5] Wilsher v Essex Area Health Authority – [1986] 3 All ER 801
[6] Wilsher v Essex Area Health Authority – [1988] 1 AC 1074
[7] McGhee v National Coal Board – [1973] 1 WLR 1
[8] Bonnington Castings Ltd v Wardlaw – [1956] 1 All ER 615
[9] Fairchild v Glenhaven Funeral Services Ltd and others; Fox v Spousal (Midlands) Ltd; Matthews v Associated Portland Cement Manufacturers (1978) Ltd and others – [2002] 3 All ER 305
[10] [2002] 3 All ER 305 at 312
[11] Chaplin v Hicks [1911] 2 KB 786.
[12] Hotson v East Berkshire Area Health Authority – [1987] 2 All ER
[13] Chester v Afshar [2005] 1 AC 134. Page 163
[14] Edwin Peel, “Loss of a chance in medical negligence”, [2005] LQR 369
[15] Kitchen v Royal Air Force Association [1958] 1 WLR 563.
[16] [1999] Lloyd’s Rep Med 223 in “Medical non-disclosure, causation and risk: Chappel v Hart” (1999) 7 Torts Law Journal, p.1.
[17] Sarah Devaney, Commentary – Autonomy Rules OK, Med Law Rev 2005. 13(102)
[18] Gerald Robertson, “Informed consent to medical treatment,” LQR [1981], 97(102-126)
[19] Paul Davidson Taylor (a firm) v White – [2004] All ER (D) 304 (Nov)
[20] Jane Stapleton, Loss of Chance of Cure from Cancer, (2005) 68(6) MLR 996-1015
[21] Simeon Maskrey, William Edis, “Chester v Afsharnad Gregg v Scott: mixed messages for lawyers.” JPI Law 205, p1-14
[22] Jane Stapleton, “Occam’s razor reveals an orthodox basis for Chester v Afshar, LQR 2006, 122(Jul), 426-448.
- Tags:
- medical law