Book Review Editor

Errors, Medicine and the Law is a most timely contribution to the enduring debate about how society should respond to accidents in general, and medical accidents in particular. It is a work that challenges us to rethink our approach to the evaluation of error and the role that blame plays in that evaluation. Its central thesis is that many events that are categorised as errors do not involve moral culpability and that litigation, far from promoting greater safety in medicine, induces secrecy and silence that conspire to inhibit the reduction of medical accident. This is a provocative and challenging book that is in tune with an emerging consciousness that we must fundamentally rethink how we deal with medical error in order to make the NHS safer for patients. The Bristol Inquiry Report⁽¹⁾ has recommended that the clinical negligence action, as a barrier to openness, should be abolished and replaced with an alternative compensation system. The government has announced that a White Paper on reforming the way clinical negligence claims are handled will be published early in 2002.^[2]

Alan Merry is a practising cardiac anaesthetist whose research focuses on the reduction of error, particularly in anaesthesia, and Alexander McCall Smith is Professor of Medical Law at the University of Edinburgh. The fact that the separate disciplines of law and medicine are so ably represented gives this text a rich flavour and illuminates the discussion in a practical way that a lawyer alone could not hope to do.

The authors state that the book is a “study of how mishaps occur and how people are blamed for them. In many areas of human activity there is a strong tendency to attribute blame for accidents which, on further investigation, may be shown not to involve any culpable conduct” (Introduction p.1). The authors argue that many cases that have been/might be characterised as examples of clinical negligence do not involve moral culpability and that a safer system of care would be promoted if we understand better the factors that underlie the errors that cause medical harm.

Chapter 1 is described as a “chapter of accidents”(p.6). The authors describe the various events that led up to the sinking of the Titanic and the explosion of the space shuttle Challenger. Both are good examples of how difficult it frequently is in the aftermath of a disaster to isolate cause. Each incident occurred as the cumulative effect of a series of conditions. In the case of Challenger, for example, ring seals failed that caused fuel to escape and explode. That failure was caused by the exposure of the seals to low temperatures on the ground. The launch was approved by a team of people who were grossly sleep-deprived and under pressure to meet deadlines. That pressure reflected the operational culture of NASA, with the result that the advice from engineers was overruled by managers who were under budgetary pressure from politicians. This catalogue of conditions led to tragedy. The authors use this and other examples to highlight the multi-factorial nature of accidents and the fact that the instinct of human beings is to seek someone to isolate and blame. Unfortunately, the fact that people may appreciate that an accident is multi-factorial is not accompanied by a change in blaming behaviour. This is the case with medical accidents, where, as Merry and McCall Smith point out, in many adverse events it is the system itself that is at fault: tired medical staff and inadequate resources may contribute to medical harm, therefore should blame necessarily be laid upon the individual doctor where a systemic failure occurs?

The authors use the example of drug error to illustrate their argument that some types of error should be regarded as slips or lapses that are an inevitable part of human behaviour and not culpable. They give the example of a Dr. Yogosakaran who faced with an emergency after an operation, the surgeon and scrub nurse having left the operating theatre, mistakenly injected the patient with the wrong drug. It would seem that the doctor had gone to the right drawer, but another person, who was never identified, had put the wrong drug in the drawer. The patient suffered brain damage and died. The doctor on finding the empty ampoule realised what he had done and reported his error to the hospital authorities. The doctor’s honesty led to his conviction for manslaughter.

The authors argue that while this looks like a case of negligence other factors were at work that would argue against a finding of culpability. They argue that the patient was high risk and should not have been treated at the relevant hospital and that the defendant was not adequately supervised so that there was a systems failure. The authors also argue (this is developed in Chapters 2 and 3) that the legal process should take account of human cognition function and recognise ‘confirmation bias’ i.e. that there is a tendency for people to read what they expect to see. Thus, a similarly named drug in the wrong drawer might reasonably lead to a mistake being made. The authors highlight the fact that courts and lawyers look for the ‘smoking gun’ when determining blame and focus on isolated acts, rather than context, a theme to which they return in subsequent chapters.

In Chapter 2, The human factor, a number of alarming statistics are put forward that demonstrate just how ‘highly prone’ is medical practice to going wrong (p.43). The central argument is that error is inevitable in any doctor’s career and that “disciplinary or legal processes which may flow from such events need to distinguish genuinely culpable practices from the human errors which are inevitable in any doctor’s career” (p.45). It is at this point that this reviewer began to feel less sympathy with the authors’ central arguments regarding error and blame. The requirements to prove breach of duty and causation as well as the cost of litigation make it notoriously difficult to mount a successful negligence action against members of the medical profession. The authors state in their Introduction that they are writing about accidents and blame generally, but the focus of the work is on medical accidents and this chapter seems to suggest that we should make a special case for doctors. It is true that the personal and professional consequences for a doctor who is subject to civil proceedings may be severe (p.246), but why should the negligent be exculpated? This reviewer was not convinced when reading Chapter 2 that the authors would necessarily make the same argument for other professionals, although this must be the logical consequence of their discussion. However, in their Conclusion they state that “the occupation of the individual is not relevant, it is the general issue of culpability in particular circumstances” (p.247).

In Chapter 2, the authors continue their theme that it is inappropriate to judge isolated events when we evaluate the behaviour of the professional. They make the point that human beings are not machines and should not be expected to behave as such: indeed, their very strength is that, unlike a machine, they are able to react to events for which they have not been specifically programmed. For example, unlike a machine, a human being can stop a task and evacuate a building when it is on fire. Therefore, say the authors, “it is irrational to judge human actions in isolation or by the standards which we would apply to the functioning of a machine. An individual’s actions should be judged in relation to the other people and technologies involved. Such judgements also need to take account of the way in which the human mind actually works” (p.56). The authors are arguing here that when we determine what are appropriate standards of behaviour account should be taken of human cognition processes. This a highly innovative argument: in clinical negligence litigation, the focus is on what the actor actually did and evidence is adduced of what s/he should have done, not the likely thought processes that might reasonably lead to error. To take just one example, the issue of ‘confirmation bias’, referred to above, provides a good illustration put forward by the authors: “The effect of this is that, having once formulated an idea of events or of their surroundings, people have a strong tendency to interpret other information in such a way as to confirm or strengthen their initial interpretation of circumstances” (p.61). The simple example of proofreading a manuscript demonstrates that the human mind will see what it expects to see so that a missing or repeated word may easily go unnoticed. The authors argue that viewed in such a light, a drug administration error (such as that made by Dr. Yogosakaran) may appear far less reprehensible.

While this argument may be persuasive, it must not be forgotten that a drug administration error may (fortunately, infrequently) lead to very serious harm and even death, as in the example used by Merry and McCall Smith. To take into account the vagaries of human cognitive processes looks unduly favourable to the medical profession when compared with patient suffering. The authors argue that the regulation of medicine is about achieving a balance between patient and professional interests. To forgive certain errors because they are understandable fails sufficiently to take account of the injured patient who may require to be compensated. One way of doing this would be to remove the possibility of suit in negligence and to replace the current tort system with a no-fault compensation scheme. However, the authors do not explicitly argue for this and they suggest that the standard of care should reflect cognition and other issues, which would make a successful medical negligence action yet more unlikely. This would surely tip the balance too far in favour of the medical profession.

Chapters 3 and 4 contain a discussion of ‘Errors’ and ‘Violations’, respectively, with a view to demonstrating that a proper understanding “of the processes which underlie human errors and contribute to modern-day disasters is essential to any meaningful analysis of blame” (p.97). The authors argue that a distinction should be drawn between errors and violations. Errors are an inevitable by-product of complex systems of human activity. An error is unintentional, a mistake, and inherently less reprehensible than a violation which is a deliberately unsafe act. This does not mean that the error is associated with an automatic action, it is not. What happens is that there is “some intended aspect of the act or plan which ‘goes wrong’, although the error itself is unintentional. In such situations the doctor is doing their best to achieve a good result, but something goes wrong. A violation on the other hand involves the “knowing breach of an identifiable principle” (p.100).

The authors do not give a clear answer to the question of whether an error should ever be considered morally culpable. They state that since an error does not involve any deliberate element, perhaps the answer should be no. However, there may be cases that manifest as if an error, but in fact occur through an antecedent violation. The example is given of a doctor who makes an honest mistake, perhaps by failing to carry out a diagnostic test, but where his/her skill has been limited by poor application at medical school, which would be a violation. Merry and McCall Smith acknowledge that it may be very difficult in some circumstances to decide whether an event should be classified as an error or a violation and that the answer may well hinge upon the honesty of the individual.

Chapter 5 provides an analysis of the concepts of negligence, recklessness and blame with a view to eliciting indicators as to when behaviour that causes ‘accident’ becomes culpable. The authors classify blame into five levels, from level one which is pure causal blame where an actor in fact caused an ‘accident’ but through no fault, through to level five, which would reveal an intention to cause harm. This chapter is of particular interest to tort lawyers in particular the discussion of level two blame where conduct is measured “against a standard which sets out what ought to be done in the circumstances in question” (p.133). This is the language of negligence, the setting of an objective standard of conduct that must be reached. The authors point out that whether conduct is determined to be negligent will depend on the question asked. In relation to drug error, if

“the question is: is it reasonable to give the wrong drug? The answer to that would seem to be no. No expert could ever say that giving the wrong drug is the appropriate thing to do, that is, the expected or standard thing to do. The conduct is therefore, in this sense at least wrong. If, instead, the question is: could this have been done by a reasonably skilled and competent practitioner? Then the answer, as we have seen in chapters 2 and 3 [where such an error might be classified as a slip or lapse, rather than a violation, and therefore as not culpable], must be yes. In fact, the empirical and theoretical data both lead us to believe that all practitioners make this type of mistake at some time even when trying their best to avoid errors. .... Thus one might go a step further and phrase the question as follows: is it reasonable for a practitioner to make one mistake in (for example) one thousand drug administrations? On this basis it is hard to conclude that such an act should be construed as negligent (p.134).”

Thus, the authors are critical of the fact that in the negligence action a person’s professionalism will be judged by reference to a single act in what might be a lifetime of high professional standards. Even the most skilled and conscientious practitioners make errors and it might be seen as unreasonable to impose this standard of care given that error is an inevitable occurrence in professional life. It is suggested by the authors that the ‘snapshot’ approach might be acceptable in a legal action that is utilised for loss adjustment purposes only and that carries no overtone of opprobrium. The authors therefore examine the negligence action to establish whether the basis of liability carries a moral component. They conclude that, formally, there are no moral overtones, but “such an action is almost always viewed, both by public and the defendant, as an allegation of culpable conduct. ... There is evidence of considerable impact on doctors and other health professionals of malpractice actions” (p.140). However, the danger is that in the concern to ensure “fairness” of treatment of the health care professional the interests of the patient are susceptible to being diminished. In this reviewer’s view it would be completely inappropriate to adjust the normative standard to take account of either a professional’s generally reasonable behaviour or the fact that the error in question is one that a reasonable practitioner might make. Until large-scale reform occurs a patient must use the tort system to secure compensation. While it is unfortunate that a ‘good’ doctor’s career may be blemished by a finding of negligence, that should not be used as an argument to deny a patient a justified claim to compensation. A career is one thing, the physical health of individuals quite another.

Chapter 6 examines the standard of care applied in negligence and asks whether the Bolam standard (Bolam v. Freiern Hospital Management Committee [1957] 2 All ER 118), as modified by Bolitho v. City & Hackney Health Authority [1997] 4 All ER 771 is excessively high. The authors are critical of the fact that according to Bolitho the court reserves to itself the power to evaluate expert opinion in order to ensure that it is logical and defensible. In such circumstances, argue the authors, how can a medical practitioner imagine how his conduct will be viewed by a court in the future. A doctor can know if he is acting in accordance with received medical opinion: it is much more difficult to second-guess a court. This is all true, but, equally, Bolitho represents a welcome break from the paternalism and protectionism of Bolam. It is interesting to note that the authors do not apparently voice any disapproval of Edward Wong Finance Co. Ltd v. Johnson, Stokes and Master [1984] AC 296, where the court rejected the customary conveyancing practice of Hong Kong solicitors.

Chapter 7 provides an interesting examination of the role of the expert witness in assessing the standard to be achieved when a defendant’s action is being evaluated. The point is made that an expert is likely to be unusually knowledgeable in the relevant area that after all is why s/he is appointed by the court. The expert will focus on a narrow area, it will be research interest, s/he will be completely up to date and furthermore, having prepared for a trial will be completely on top of the detail. By comparison the doctor may well be a general practitioner with a wide, but less detailed knowledge base, and s/he may have been acting in a pressurised situation. Thus, the scales are not evenly balanced.

Finally, in Chapter 8, the authors examine how the needs of the injured might be addressed apart from the tort system. They acknowledge that if there were no alternatives then it is arguable that the negative consequences of tort litigation should be accepted as inevitable. The authors detail a number of alternatives including the accident compensation scheme in New Zealand and the no-fault compensation scheme that operates in Finland. The concept of fault was reintroduced into the New Zealand scheme in 1992. In their Conclusion the authors state that “If a way could be found for compensation to be paid without any finding of negligence with all that entails in terms of blame and moral censure, this would be highly desirable” (p.247). Undoubtedly, publication of the government’s forthcoming White Paper is awaited with interest.

To conclude this a highly readable, accessible text that should be read by all those who are interested in the compensation debate, and the field of tort law in particular.

⁽¹⁾ Learning from Bristol. The Report of the Public Inquiry into children’s heart surgery at the Bristol Royal Infirmary 1984-1995, Cm 5207, 2001.

⁽²⁾ Hansard: 4 July 2001: cols 221-222W.