TO sustain a claim of medical malpractice, a plaintiff must show a pre-existing duty on the part of the physician and then a breach of that duty that causes the plaintiff damages. 1The duty arises most commonly when a physician–patient relationship is formed and requires the physician to provide patient care that a similarly situated practitioner in good standing would provide in the same clinical circumstances. 2Care falling below this standard represents a breach of duty; if the breach causes the patient injury, the physician is liable for damages. Plaintiff claims of a breach of the standard of care and causation generally must be supported by expert testimony. 3However, in this issue of ANESTHESIOLOGY, Edbril and Lagasse 4report data that indicate adjudication of medical malpractice claims may not follow these legal precepts. They find that upon review of anesthesia care by anesthesiologists provided in an academic setting, substandard care was unrelated to litigation risk and adverse malpractice adjudication. Thus, the standard of care defined by anesthesiologists, as mandated by the legal system, does not seem to comport with litigation risk and malpractice adjudications against anesthesiologists.
This study adds to the growing literature indicating that the legal system may not be adjudicating malpractice claims according to the legal rule. Using data from the Harvard Medical Practice Study, Brennan et al. 5found that malpractice liability was only correlated with severity of patient injury, not negligence. Furthermore, in their seminal work from the American Society of Anesthesiologists closed claims study, Cheney et al. 6found that more than 40% of patients who were provided appropriate, nonnegligent care as defined by neutral anesthesiologists still collected payments from these anesthesiologists. But physicians also disagree with jury verdicts that hold for the defendant physician. Liang 3reported that a homogeneous sample of neutral anesthesiologists in an academic center disagree with jury verdicts, even in some cases in which juries found no defendant anesthesiologist negligence. Radiologists have shown similar behavior. 7These findings are compelling because it has also been reported that lay persons, without medical or legal knowledge, are statistically better able to predict jury verdicts than anesthesiologists, who are legally informed as to the standard of care through their clinical training. 8Even studies that report that the tort system assesses negligence appropriately find that non-negligent physicians are still required to pay malpractice judgments against them. 9,10Thus, one major goal of the medical malpractice tort system—to provide efficient and appropriate physician incentives to render nonnegligent care to minimize patient injury—seems to be unfulfilled by the traditional tort system.
Furthermore, Edbril and Lagasse report that anesthesia care deemed negligent by the study's reviewers did not result in patient suit or malpractice system compensation. This result, too, is consistent with results from the Harvard Medical Practice Study, which reported that very few negligently injured patients are compensated by the tort system. 11–13Hence, the other major goal of the malpractice system—compensation of patients who are negligently (as defined by the legal standard) injured—also seems to be unfulfilled. These conclusions raise the possibility that the billions of dollars spent on the malpractice system annually 14may not be an effective allocation of social resources to minimize patient injury, maximize patient safety, and compensate injured patients. 3
Edbril and Lagasse suggest that to combat these legal system weaknesses, a peer-review process should be used to assess provider negligence. However, this approach has significant difficulties. Hindsight bias plays a tremendous role in ex post review of clinical circumstances 15–17; thus, negligence may be overestimated. Indeed, Edbril and Lagasse's characterization that all adverse events are somehow a result of “error” seems to reflect this bias. Most researchers in this field would not characterize adverse events in this manner because even appropriate provider actions may result in untoward clinical results. This is best exemplified by Edbril and Lagasse's own definition of system error, including accidental occurrences that result from performing a technique properly, equipment failure despite proper use, miscommunication while following established protocol, inability to diagnose a disease process because of limitations of presently available screening and monitoring standards, inability to treat a disease process because of limitations in present standards of care, and inability to meet the demands for resources of equipment or personnel. These “system errors” accounted for 88% of the total errors reported, none of which are preventable. We do not believe these are errors at all and instead represent untoward events that are not preventable. We must emphasize that the human errors that resulted in a disabling injury as reviewed by Edbril and Lagasse form the basis of their report, not the system errors. Finally, it is difficult to determine whether such a peer-review system would result in cost savings relative to the current litigation system. Attorneys presently reject most requests for malpractice claim representation; reducing such barriers to facilitate plaintiff claim filing through peer review or other low-cost administrative mechanisms may consume and even surpass the cost savings from avoiding court. 14,18,19
Other investigators have suggested the use of clinical practice guidelines. 20The use of clinical practice guidelines is fraught with difficulty at the outset because these guidelines are not recognized legally as the standard of care. 20Furthermore, guidelines, similar to a peer review process, assume there is a single standard of care for all clinical circumstances; this may be too narrow a viewpoint. 21Significant discordance between evaluating physicians has been reported when assessing clinical scenarios, 7,8,22,23reflecting the well-known variability phenomena in clinical care. 24,25The standard of care consensus in the study by Edbril and Lagasse most likely reflects homogeneity of a single anesthesiology department, which was the sole source of reviewers. The vast majority of practice guidelines are also formulated by consensus rather than double-blind study and may conflict in their recommendations. 26An excellent example of a clinical practice guideline with such problems is the pulmonary artery catheterization guideline published in ANESTHESIOLOGY. 27This guideline does not provide objective evidence to support the benefits versus the risks of pulmonary artery catheterization, but merely lists the pros and cons regarding catheter use and clinical scenarios in which some practitioners have found them useful. Guidelines are also limited by the source from which they emanate, 28which may reflect specialty turf battles rather than clinical indication or medical appropriateness. Finally, these guidelines are often outdated as soon as they are published and, in any event, are manipulated by local physicians and managed care organizations before being put into use. 29
The reported results have significant implications for patient safety efforts. Instead of providing anesthesiologists with a clear incentive to provide nonnegligent care, the uncertainty of legal adjudication in practice may result in a heightened level of defensive medicine and paradoxically an increased risk of patient injury. 3,8,30Although there is significant debate as to whether the legal system actually induces defensive medicine, 30other mechanisms that instead induce physicians to proactively adopt behaviors that minimize patient injury and maximize patient safety would be a far better way to spend current malpractice system dollars.
To fulfill the goal of maximizing patient safety, a commitment to using these dollars to focus on evidence-based medicine and patient safety outcomes is essential. 31A concurrent and necessary step would be to provide for immunity against legal discovery of data from internal and external safety reporting systems to encourage error reporting. Currently, legal incentives and potential discovery of this information in tort suits substantially chill reporting medical error. 30,31In combination, open reporting of medical error and resources devoted to studying the issues so identified would provide significant progress toward systemically maximizing patient safety. 30Similarly, accreditation organization support via data standardization, nonpunitive reporting approaches, and education would begin the process of broad, interprovider analyses that may yield insights into methods to maximize patient safety and minimize error. 30
To accomplish the goal of compensation, one possible mechanism could be the use of a workers’ compensation-type system for patient injury caused by medical error, with direct patient suits available only for situations of reckless provider actions. For the small minority of direct patient suits involving such actions, court-appointed experts representing the court, not the litigants, would be far more appropriate, eliminating financial bias from an expert's opinion. A similar system has recently been proposed for the aviation industry. 32This approach would focus on corrective action to improve patient safety, reduce the threat to clinicians of reporting errors, and still deter inappropriate, high-risk behavior.
Overall, Edbril and Lagasse's work highlights for the anesthesia community the significant weaknesses of the traditional tort system and the incentives it creates. 30,31Their work also puts into stark relief the fundamental need for new methods to appropriately affect physician behavior so that the health delivery system can continuously improve in its efforts to provide safe, effective medical care while minimizing medical error and patient injury. 30