To that end, this second edition of Charles Vincent's classic text pursues a subtle but very important objective established in the original edition in 1995. Diverse contributors, both clinical and nonclinical, provide an international, multidisciplinary perspective on medical error. Clinicians, nurses, paramedics, technical staff, risk managers, clinical auditors, quality assurance reviewers, quality improvement managers and all others involved in the delivery of health care can find here the common threads of a long-overdue, integrated approach to patient safety, one that creates an atmosphere in which we can all start talking to each other. This new culture of clinical risk management is freed to a large extent from its historical preoccupation with containing litigation costs and reducing harm to institutions. It becomes, instead, a vehicle for directly improving the quality of care. A new movement in error reduction is now underway, and Vincent's book is one of its first texts.

The book is divided into four parts. The first, "Principles of Risk Management," opens with a chapter by psychologist James Reason, surely now the recognized father of human error theory. With clarity and insight, he sets the theoretical background for the practical issues that follow. A chapter by Eric Thomas and Troyen Brennan provides a good review of benchmark studies on error and adverse events in medicine in the US and Australia. Vincent has recently published a preliminary study from the UK using a similar methodology; it would be useful to have comparable information on adverse events in Canadian hospitals to see if we fare better or worse than other countries.

The second part, "Reducing Risk in Clinical Practice," provides an overview of error issues in obstetrics, pediatrics, anesthesia, surgery, emergency, general practice and psychiatry. Many clinical data are referenced here that might not otherwise readily come to light. It is difficult, for example, to find data on error in general practice, and the studies referenced in Stephen Rogers's chapter, especially those from the Medical Defence Union of the UK, will be useful to many. In contrast, some readers on this side of the Atlantic might find that the chapter on risk reduction in obstetrics, which deals almost exclusively with the UK system, might have benefitted from the inclusion of data from other systems. The book's lone Canadian contribution, a chapter by Jan Davies and Alan Aitkenhead on clinical risk management in anesthesia, is a well-balanced presentation of some of the special insights anesthetists have historically brought to error theory and safety issues in medicine.

There is also an excellent and surprisingly readable chapter on clinical oncology; it is with some relief that we learn that maverick clinical decision-making in clinical oncology is in decline. Generally, we do not hear much from this area until a catastrophic event occurs, such as the intrathecal injection of an intravenous chemotoxic agent, a tragic and peculiarly repeating error. The contribution on psychiatry provides a thorough treatment of suicide and violence to others, certainly major issues in risk management. Regrettably, however, there are so many other areas in which the psychiatric patient is failed, and these are not dealt with here. Overall, despite the wide variability in what we do and where we do it, this section very usefully illustrates the ubiquity and persistence of error phenomena. Each specialty faces its own, but similar, demons.

The third part, "The Conditions of Safe Practice," is an excellent collection of new chapters on a variety of topics, all of them germane. Two important themes emerge. The first, widely appreciated in theory but less so in practice, is that error is inherent in clinical work because it derives from uncertainty. This is the notion of "necessary fallibility." Where there is uncertainty, there is a need for clinical decision-making, and this opens the door to error. Given the inevitability of error, we might be excused for punishing ourselves a little less than society or the legal profession might otherwise allow. The second and equally important point is that we are often only as good as the tools we are given; some error may be unavoidable in the conditions under which we work. James Reason sees this as the passive facilitation of error by the design of the environment in which it occurs. Inevitably, when resources are limited there is a trade-off in quality. The not-so-subtle nature of this argument often seems to have been lost on those who allocate resources to health care.

The third section begins with the issue of communicating risk to patients and their families. This is a useful overview that should be required reading for those in training. The chapter on "guidelines and pathways" deals sensibly with some of the predictable, now-tired arguments against "cookbook" medicine. It is high time we accepted and incorporated guidelines into the armamentarium of care; we need fewer mavericks and more evidence-based consensus. Does anyone really believe some loss of autonomy to be too high a price for patient safety?

Another new and burgeoning area, human factors engineering, is introduced in this section, which addresses the interfaces between the medical environment, the equipment it contains and those who operate it. New kinds of errors will result from new technology, and we would do well to anticipate them. The chapter on stress and fatigue is especially important and, in the medical domain, is almost worth a book in its own right. Historically, the delivery of health care has exacted too high a toll on front-line workers. Many work practices, especially those involving shift work, are painfully outmoded and would not be tolerated in private industry. Shift work has become a public health issue that urgently demands reform.

The chapter by Jenny Firth-Cozens entitled "Teams, Culture and Managing Risk" is clear, concise and commonsensical. Medicine invariably benefits from inviting discerning onlookers, in this case a psychologist, to take a look at what we do.

The final section of the book, "Implementation of Risk Management," might appear on the face of it to be the least clinical part of the book, but it does raise a number of important issues for clinicians. A major problem facing the new science of error reduction in health care is the monitoring and reporting of error. It has been estimated that current incident monitoring techniques miss perhaps more than 90% of errors. Aside from "tombstone learning," which typically occurs after a catastrophic outcome, one might ask how much we learn from what we are doing unless we develop an accurate database. The chapter on clinical incident monitoring recognizes and details this important problem and advances a number of useful suggestions for making adverse incident reporting work. An important consideration for busy clinicians will be finding ways to simplify reporting and minimize paperwork. Given the miserable record of physicians in reporting even relatively straightforward adverse medication events, any optimism on this score must be guarded.

After incidents have been detected and recorded, the next step is to investigate them. One particular approach, used by the Clinical Risk Unit at University College, London, is described in some detail in the chapter by Charles Vincent and Sally Taylor-Adams. Problems identified in care management are analyzed in accordance with a set protocol. Those entrusted with incident investigation might find this a useful template for developing their own protocols. It might also serve as an effective tool for gathering data on potential adverse events and the working conditions under which they occur.

Vincent's chapter on caring for patients harmed by treatment itself provides a sensitive and empathetic treatment of a difficult area. Several useful guiding principles are reviewed; again, this should perhaps be required reading for those in medical training or, indeed, for any physician placed in this unfortunate position. The subsequent chapter by David Hewett on supporting staff involved in adverse events is also a thoughtful treatment of the impact of error on physicians. It includes a description of the principal tenets of the National Health Service Litigation Authority, the central pool or repository for liability of hospital doctors in the UK. These appear solid and well founded, and some thought might be given to following a similar paradigm here.

The final three chapters are concerned with the mechanisms of complaint and claims management and with dispute resolution. The very cultural nature of these processes dictates a largely domestic treatment, and North American readers will find little to help them here. However, there is some useful discussion of alternatives to litigation that might inspire those interested in the badly needed reform of our adversarial litigation system. As things stand, patients and their families often face heavy psychological and financial costs in resolving clinical negligence disputes. Many feel that the time has come for some form of nonadjudicatory dispute resolution. Would it be too simple to have all claims evaluated by a neutral panel of medical experts, including some who understand error theory, with financial settlement based on actuarial input?

Overall, there is much to commend this second edition. It comes at a time when issues of patient safety are clearly in the ascendant. In the 1995 edition, Clements defined risk management as "the reduction of harm to an organization by identifying and, as far as possible, eliminating risk." This expanded and rich collection will serve as both a text and reference source for anyone involved in the new integrated approach to clinical risk management, which now sensibly places the emphasis on reducing harm to individuals.