• medical injury
• patient risk
• “no fault” system
• medical error

Retrospective medical record reviews suggest that 4–16% of hospitalised patients suffer harm, which is judged preventable in about half of cases. In the study by Davis et al¹ published in this issue of QSHC, 5.2% of admissions in New Zealand led to a preventable in-hospital event, a similar rate to that in the UK and within the broad range of other studies. Clearly these results are important for New Zealand health care, but they also have a special significance because of New Zealand’s “no fault” compensation system. To understand this it is necessary briefly to consider the basis and justification for different approaches to compensation for medical injury.

In most countries compensation for medical injury is based on the tort system or other “fault based” models. The claimant—the injured patient—must take legal action to prove duty of care, injury, causation, and negligence. In a “no fault” compensation system an expert panel will assess whether the injury has indeed been caused by health care, but the patient does not have to go to court and does not have to prove negligence in order to be eligible for compensation. Most “no fault” systems do not compensate all injuries from health care. For instance, the New Zealand system compensates for injuries caused by medical error and rare mishaps, but generally not for injuries deemed not to result from error.

The tort system, however, is not simply a system of compensation but is also intended to emphasise accountability of individuals and institutions and to be a deterrent to substandard care.² Simply put, the likelihood of being sued is supposed to inject a certain caution into clinical practice and decision making which is supposed to improve patient care. If this were so, one might think that countries operating tort systems would have a lower level of adverse events. However, the findings of Davis and colleagues, while not a direct test of the impact of tort, suggest at the very least that such an effect is fairly marginal—if it exists at all.

Tort systems may have had some positive effects. The rise in litigation, and reflection on its causes, was certainly one powerful driver towards assessing the full extent of harm to patients. Arguably, the threat of litigation has promoted better communication with patients and more collaborative decision making. In Britain the NHS Litigation Authority requires hospitals to appoint risk managers with some responsibility for reducing risk as well as managing claims. There are, however, many negative consequences to set against these positive influences.

With the rise of patient safety and systems thinking about the causes of adverse events, the tort system is looking increasingly anachronistic and an obstacle to progress on patient safety.³ The system has been criticised as costly, slow, inequitable in various respects, and blame orientated. It can be traumatic for those involved—patients and professionals alike—inducing much bitterness on both sides. The system is inherently adversarial and, although much of the trauma can be reduced by sympathetic and effective lawyers on both sides, patients still have to fight for compensation at a time when they really need to be looked after. Reforms being considered in Britain include fixed tariffs for specific injuries, alternative dispute resolution, structured payouts instead of large “one off” lump sums, as well as “no fault” systems.⁴

Tort is built, in practice at least, on the notion of individual fault. Typically, specific lapses in the standard of care provided by individuals are identified as the grounds for compensation. While not denying that clinicians do make specific errors, recent approaches to understanding adverse events emphasise that there is usually a chain of events leading to an adverse outcome, each of which may be influenced by a wide range of contributory factors.⁵ There is therefore a tension—in fact, a fundamental incompatibility—between a judicially orientated enquiry and the systemic investigations required for improving safety. Furthermore, fears of litigation are a frequently voiced obstacle to open reporting of errors and adverse events. While some clinicians’ fears can be offset by confidential or anonymous reporting, in a tort system there is always a trade off between the interests of patients and patient safety and the interests of clinicians and healthcare organisations.

An optimal system must act to reduce errors and harm and yet also compensate injuries once they occur. Studdert and Brennan³ argue that such a system should have five broad characteristics:

• the programme should encourage healthcare professionals to report errors;

• it should send strong quality improvement signals with financial incentives for safety and quality;

• it must include mechanisms to deal with the small number of rogue or reckless clinicians who harm patients;

• the compensation programme should act to reinforce rather than undermine the honesty and openness of the clinician-patient relationship; and

• where appropriate, patients should be compensated in a manner that is speedy, equitable, affordable, and predictable.

Studdert and Brennan consider that a “no fault” system can achieve all five goals, and argue that such systems should now be piloted in the United States.

Several “no fault” compensation systems with different characteristics operate internationally in Denmark, Sweden, Finland, and New Zealand. The Swedish approach, for instance, has proved to be efficient and affordable, although backed by a strong social security system. Physicians in Sweden appear to regard assisting with compensation claims as a continuation of the duty of care and as a natural part of their responsibility to their patients. Studies by Brennan and colleagues suggest that implementing the Swedish system of compensation, which employs a criterion of avoidability of injury as grounds for compensation, would not lead to greatly increased overall costs in the United States, but would compensate many more injured patients. Incentives for enhancing quality and safety of care can be built into “no fault” systems in several ways, usually in the form of “enterprise liability” by which a hospital or other healthcare organisation becomes liable for the costs of compensation, rather than individual clinicians. In addition to providing incentives to safe care, this approach is also consistent with a systems approach to the understanding and prevention of adverse events.

“the idea that [the tort system] acts as an effective deterrent is now bankrupt”

While the tort system can still be defended in some respects, and might be more acceptable in some countries than others, the idea that it acts as an effective deterrent is now bankrupt. As Davis and colleagues suggest, we now need to look more directly at the operation of “no fault” systems, both in terms of compensation arrangements and their impact on quality of care. There is, for instance, an assumption that “no fault” systems encourage reporting of errors but there is little evidence to support this view. The most important criterion for assessment of any compensation system should be its impact on injured patients and their families, not just in providing appropriate financial recompense where necessary but in ensuring that explanations, apologies, and long term support and care are regarded as the expectation rather than the exception.⁶ Compensation would ideally be a gesture of reconciliation and an acknowledgement that a healthcare organisation has a special duty of care to those it has harmed.