Alan Merry

Error, Blame, and the Law in Health Care—An Antipodean Perspective

Until recently, while people understood that disastrous outcomes can result from avoidable failures in health care delivery, most people believed that these outcomes were fairly isolated events. In the last 10 years, however, reports from several countries have drawn attention to the fact that harm to patients receiving health care is not rare but common, systematic, ubiquitous, and, all too often, severe (1-6). Much of this harm is attributed to error, which in its broadest sense includes systemic problems in the organization of health care (7). However, the term often embraces widely diverse events, failing to distinguish true errors from behaviors that should properly be classified as violations. This failure may extend to behaviors at supervisory, administrative, and political levelsshould appropriate root-cause analysis be performed (8, 9). When things go wrong, the usual human response is to apportion blame, demand retribution and compensation, and seek assurance that the error will not occur again. Such redress is usually done through the legal system. However, less than 1% of people suffering preventable harm receive any compensation through the tort system, and there is little relationship between successful litigation and the degree to which negligent practice has contributed to harm (3, 10). Theoretically, tort should help promote high standards and provide compensation for injured patients. In practice, it often does neither; is grossly wasteful of resource; and is time-consuming, threatening, and unpleasant for both plaintiff and defendant. It may be thought that a no-fault system of compensation, such as in New Zealand and the Scandinavian countries, would solve these problems. However, whichever approach is used, it is common that neither patient nor physician feels well served and that few initiatives to improve safety eventuate (1, 8, 11). In the Scandinavian systems, approximately half the people who apply for compensation cannot demonstrate the causal connection or fulfill other eligibility criteria; in New Zealand, compensation has been limited in both scope and amount (11). These problems are largely attributable to a failure pwithin the legal and health care systems to understand the causes of system breakdowns, and attribute blame when none is due or to obtain compensation through tort (11). In this paper, we consider the concepts of error and violation, the proper attribution of blame, and the role of the legal system in health care, and we outline some of the approaches that are being considered in several countries. Errors Health care is an unusually complex system. Some features that predispose to errors and aggravate their consequences coexist and interact to a degree that is seldom found in other human endeavors. These features may not only set the stage for incidents and accidents but may be the prime causes (3, 8, 9). Health care is delivered in a dynamic environment with complex interactions between pathophysiologic and disease processes, medical staff, infrastructure, equipment, and policies and procedures (9, 11). Challenging procedures must be undertaken, sometimes under circumstances that are not chosen or controlled by those at the front line. While this is mostly successful, errors do occur. There are several definitions for the term error (1, 9, 11). We propose the unintentional use of a wrong plan to achieve an aim, or failure to carry out a planned action as intended. An important feature of this definition is its focus on the thoughts and actions involvednot the outcome. Errors are intrinsic to normal cognitive processes (8, 9). Moreover, an error thought to be preventable in an individual case may, in fact, be statistically inevitable in the career of a physician (12). Giving one drug to one patient correctly seems simple, but for an anesthesiologist, administering a half-million intravenous injections during an entire career, often under difficult circumstances, without ever making a mistake is almost impossible without technologic support systems (13). The consequences of errors may be severe, and such failures should not be accepted simply because they reflect an intrinsic human characteristic. However, it is irrational and unjust to hold an individual who makes a genuine error while trying to do his or her best morally culpable (11). Berwick stated that to expect to function without errors and to view error as a failure of character every system is perfectly designed to produce the results that it does (14). The solution to error lies not in accountability or punishment for individuals (although these have a place) but rather in redesigning the system to reduce the risk for error and limit and manage the consequences when errors occur (1, 3, 7-9). Designing health care systems specifically to compensate for the propensity for error on the part of physicians has, at least until recently, run counter to a powerful medical culture that led physicians to expect to function without errors and to view error as a failure of character (15). This reluctance to admit error, coupled with advice from insurers that nothing should be said to injured patients, for fear that it be taken as an admission of liability, led to a conspiracy of silence that inhibited the development of systemic safety approaches and has had a corrosive effect on trust in the medical profession. A prerequisite for making systems safer is understanding why and how errors occur. Because most of the things that go wrong occur infrequently (3, 16) and few data about system failures or human error are recorded in the medical record (17), information has to be gathered after the event. Eliciting detail about contributing factors in incident reports dealing with near misses or events with only minor outcomes, as well as by root-cause analysis after serious or sentinel events, is vital to find out not only what is going wrong but how and why these problems are occurring, so that appropriate preventive and corrective measures can be devised (3, 7, 18). If those who provide this information are personally and professionally threatened by the very process necessary for the solution of the problem, the information will not be forthcoming. Separating the processes for accountability and system improvement and the importance of a blame-free environment for the latter are discussed later in this paper (11). Making some errors is normal and inevitable. Most individuals who make errors are not morally culpable, and outcome is a poor index of moral culpability. Violations Violations are quite a different matterwhile errors are unintentional, violations are deliberate. A violation may be defined as a deliberatebut not necessarily reprehensibledeviation from those practices appreciated by the individual as being required by regulation, or necessary or advisable to achieve an appropriate objective while maintaining safety and the ongoing operation of a device or system (11). Again, the definition focuses on the mental state and actions involvednot on the outcome. Violations are common in all human activitiesexceeding the speed limit while driving, for example. Speeding is not usually an error (although it may be) because the driver usually chooses to break the rule. There is no intention to harm associated with most violationsthe operator believes that the violation will produce a successful result, and it often does (getting to ones destination more quickly, for example). Violations may even be necessary at timesfor example, speeding may be justified in an emergency (9). Sometimes they are forced on people by the systemworking beyond safe limits in the case of junior physicians, for example. However, violations increase the risk for error, or for the consequences of error (speeding and drunk driving are obvious examples). Furthermore, violations are always deliberate, and they can be avoided by choice. For example, hospital management could choose to organize their services to avoid excessive physician fatigue. In contrast to error, harmful violations might be prevented by deterrence, provided the deterrent measures are well targeted. Punishing a junior physician for making an error after the violation of working excessive hours was forced on her by the system would be wrong, even if the patient died, as in a notable Irish case (19). On the other hand, prosecuting the hospital authorities for corporate manslaughter might produce desirable change. Strategies to reduce violations (and some errors) include automating tedious checking, documentation, and monitoring tasks and improving compliance with protocols by ensuring that they are appropriate, flexible, and readily accessible. Continuous monitoring is necessary after such changes because each change will offer opportunities for new problems (3, 8). Strategies at the system or health care level include setting appropriate standards for safety and quality and making proper compliance a condition for accreditation; currently, some gross violations of industry standards are the norm in sections of health care (working hours, for example). Blame and the Law Before the Industrial Revolution, the law had generally considered that people were strictly liable for harm that was caused by their actions; they would be punished and would have to compensate the wrong. However, with the burgeoning of new industries and risks, concepts of strict liability were seen as an inappropriate brake on the innovation of the new entrepreneurs and the view became that there should be no liability without fault (20). This was the birth of the concept of fault as an essential component of the tort system (tort is derived from the word wrong in Latin). In practical terms, fault implies blameworthiness. Blame is of major emotional significance to those blamed, and people wrongly blamed have a powerful sense of being the victims of injustice. Some lawyers dismiss the distress of physicians publicly an