American Journal of Law & Medicine

Toward a workable model of "no-fault" compensation for medical injury in the United States.


Tort "crises" and serious interest among policy-makers in tort reform are interrelated, cyclical phenomena in the United States. In the mid-1970s, and again in the mid-1980s, medical malpractice markets across the country experienced rapid increases in the frequency of claims, rising premiums, and in some places, shrinking options for professional indemnity coverage.(1) A wave of tort reform measures followed, the majority of which were designed to chill the interest of plaintiffs' attorneys in bringing suit.(2)

Although claims severity appears to have crept steadily upward,(3) most industry indicators suggest that medical indemnity markets have held fairly steady over the past decade.(4) The extent of this placidity may be somewhat misleading, with the revenue lines of many insurers apparently buoyed by a cushion of savings from overly pessimistic premium hikes in the late-1980s.(5) Nonetheless, the predictions of some pundits about a third malpractice crisis, 1990s-style, have not proved accurate.(6)

The decade proved to be a period of relatively little agitation for system reform among the key actors--physicians, plaintiffs, attorneys, and insurers. In response, state legislatures generally avoided consideration of any substantive changes in the regulation of medical malpractice; they were apparently content to leave the courts to tinker with and dilute some of the more staunchly pro-defendant measures that had been enacted in crisis years.(7)

Despite the lack of any widespread dissatisfaction with the malpractice system today, empirical evidence of its lackluster performance continues to mount.(8) Concerns persist about the system's ability to promote high quality health care and to distribute compensation fairly to patients who suffer injury caused by negligence.(9) Nonetheless, devoid of stimulus from market shocks for pursuing significant system reform, the status quo looked, until recently, comfortably outside the purview and interest of policy-makers.

Over the past year, tremors have emerged from a most unlikely epicenter. The Institute of Medicine's ("IOM") 2000 Report, To Err is Human, brought together in startling fashion data on the burden of morbidity and mortality attributable to medical injury in American hospitals.(10) The Report also focused attention on the principal factors that feed the problem of medical error and iatrogenic injury--most notably, a failure to observe and learn from errors (partly attributable to providers' reluctance to report them), and a continued focus on individual rather than systematic causes of error. The IOM traced the roots of these endemic features of the health care system to medical malpractice regimes, a connection that the media and policymakers readily grasped.(11) Consequently, for the first time in more than a decade, serious discussion about experimentation with alternative approaches to patient compensation has begun.(12)

Because of their compatibility with patient safety agendas, "no-fault" approaches present the most intriguing possibility from within the list of alternatives.(13) Experimentation with no-fault compensation in the United States has long been debated.(14) We have previously demonstrated that a no-fault regime would be affordable.(15) However, a number of thorny legal and logistical difficulties would need to be resolved before a no-fault program for medical injury could be tested. In this article, we outline those difficulties and suggest a number of solutions that may be suitable for incorporation through legislation that enables experimentation with no-fault regimes in health care.


The IOM Report brought to the attention of a wide audience what has become increasingly clear through epidemiologic research in recent years: with modern medicine's capacity to alleviate disease, injury, and disability comes significant risks that it may independently cause harm. Data on the incidence of harmful mistakes suggest that the supposed deterrent effect of medical malpractice suits alone has not been sufficient to address the problem.(16) On the contrary, litigation may well stifle efforts to reduce error.(17)

Steps to redress medical injury depend on detailed information on the incidence, typology, and etiology of errors. Recognizing this, proposals ranging from voluntary reporting of error to mandatory reporting of unintended harm to patients are already in the public domain,(18) as is an awareness of the need to consider systemic factors that contribute to the occurrence of errors.(19) To be successful in the current environment, however, the push to gather more detailed information on errors and their causes must contend with practitioners' wariness of reporting events that may leave them open to accusations of negligence. Both anecdotal and empirical evidence suggest that providers are less willing to disclose information about errors they make or see when a punitive atmosphere prevails.(20) Fear of blame among those individuals closest to errors thus poses a major obstacle to design and implementation of patient safety initiatives.(21) This cost, together with ongoing concerns about the performance of the medical malpractice system,(22) underscore the need to consider alternative systems that are better able to compensate injured patients and promote high quality healthcare.

The key requirements of an alternative compensation model have been exhaustively reviewed by other commentators.(23) It must be able to channel compensation to eligible patients in a manner that is predictable, timely, affordable, and fair. In order to target safety improvement activities, the system must generate or at least be compatible with the generation of detailed information about errors without fear of reprisal against those who report that information. At the same time, the system must find ways to incent health professionals and organizations to work to improve quality of care. In addition, the compensation system must emphasize what the IOM Report concluded: most of the preventable injuries in our hospitals are not due to bad medical professionals, but rather to the imperfect systems in which these professional work. Although the system must have mechanisms for identifying and protecting patients from those few truly incompetent practitioners, it must avoid adhering to what Berwick has called "The Theory of Bad Apples"--the seductive (but erroneous) notion that significant advances in quality are achievable by discovering aberrant behavior and punishing individuals who are "guilty" of it.(24)

The major alternative to the current tort system is the no-fault model.(25) The central premise of this model is that patients need not prove negligence to access compensation. They must prove only that they have suffered an injury, that it was caused by medical care, and that it meets whatever severity or other threshold criteria that are in place. A major no-fault scheme already exists in the United States in the form of workers compensation.(26) In the medical arena, no-fault in varying forms is firmly established in several health care systems abroad: New Zealand and Sweden have operated no-fault systems for compensating medical injury for more than 25 years,(27) and Finland, Denmark and Norway for more than a decade.(28) Several small medical no-fault schemes have also been implemented in the United States to compensate specific injury types, including the Florida and Virginia schemes for birth-related neurological injury and the National Vaccine Injury Compensation Program.(29)

The theoretical benefits of a no-fault scheme measure up impressively against the key system prerequisites noted above, and there is some empirical support for the notion that those benefits are attainable.(30) By minimizing court time, lawyer expenses, and lengthy inquiries about whether or not defendants are negligent, no-fault should provide faster, more efficient compensation. The evidence emerging from the "bad baby" schemes in Florida and Virginia suggests cause for optimism on these fronts.(31) Such operational strengths are bolstered by new arguments that the no-fault model may offer an environment that is more conducive to discussing and reporting error, collecting data that can inform error prevention strategies, and generating incentives for physicians and organizations to strive to make health care safer.(32)


The two most experienced no-fault schemes in operation abroad, New Zealand's Accident Compensation Commission ("ACC") Scheme and Sweden's Patient Insurance Compensation Fund, both have a simple administration system.(33) Streamlined adjudication pathways handle straightforward claims; expert panels convene to hear the more contentious ones.(34) Claimants in New Zealand enjoy rights of appeal to the courts, while in Sweden they may request a panel of physician experts and proceed to independent arbitration.(35) Physicians in Sweden actively participate in the filing of more than 60 percent of claims.(36) Both schemes function with relatively little attorney involvement, and available evidence on the administrative costs suggest that these no-fault programs absorb dramatically lower administrative costs than their tort counterparts.(37)

If it were to fit within the existing health and legal environments in the United States, a no-fault scheme would not be able to operate in this fashion. Two structural departures would be particularly crucial. First, in order to minimize conflict with state and federal laws, the scheme would need to proceed at the state level. Second, it is legally and politically unrealistic to anticipate that any state would undertake wholesale replacement of tort law with a no-fault scheme, as Sweden and New Zealand have done. The scheme would need to be voluntary, in the sense that patients had some opportunity to opt-out, either at the point of receiving care, selecting their physician or hospital, or when obtaining insurance coverage through a health plan or hospital. In addition, given a variety of unknowns about how no-fault compensation for medical injury would work in the United States, we believe that the most prudent approach to implementation would be a series of pilot schemes with selected participants in a limited number of states, followed by rigorous evaluation.

As part of a four-year project to assess the cost and design of no-fault programs in Utah and Colorado, we investigated some of the structural details that would permit development of a state-based, voluntary no-fault scheme.(38) Through extensive discussions with policy-makers in those states and close study of existing foreign and domestic no-fault systems, we identified the following core features.


The scheme would require some central organization, situated at the state level, that would assume responsibility for receiving claims, adjudicating which injuries met compensation criteria, and disbursing compensation. In addition to these claims processing functions, the entity would maintain a database of injury types, together with factors determined to have contributed to the injuries, and it would disseminate these data in a redacted form to researchers and institutions throughout the state. The entity would establish benefit levels under direction from the legislature. For the mountain states, we predicted that it would be feasible to have a single large medical malpractice insurance company act as this central claims facility. Some initial public support may be necessary to set up a claims processing infrastructure, but the facility's core funding would come from premiums collected from institutions participating in the scheme.


We believe that the best means of communicating appropriate patient safety incentives to participants in a no-fault program would be to use institutions or "enterprises" as the unit of participation. The enterprise liability model and its potential advantages has previously been described in detail by Sage and by Abraham and Weiler.(39) The model's system orientation complements the broad view of error causation that leading commentators now promote.(40) Moreover, its capacity to impose experience-rated premiums promises to confront health care facilities with direct economic incentives to reduce their rates of medical injury.(41)

Although enterprise liability is thoroughly compatible with a no-fault approach,(42) introducing a no-fault program through this framework has clear limitations. Most notably, it will mean that participation in the no-fault system may not be immediately open to every kind of health care practitioner or facility. Assessment of the suitability of different institutional profiles to enterprise liability would need to be a goal of any full-scale demonstration project.

For initial experimentation, however, we propose that the logical group of institutions to test the potential benefits of a no-fault scheme would be large integrated medical centers. By covering the malpractice costs of affiliated providers through self-insurance, a number of these centers already have the core features of an enterprise model in place.(43) Of course, designating institutions as the premium-paying participant in a voluntary no-fault program does not vitiate the need to have individual patients in those institutions agree to forego tort remedies for future medical injury claims.


Workable compensation criteria are fundamental to any no-fault scheme. Adverse events occur in approximately three to five percent of hospitalizations;(44) the budget for compensating every one would be prohibitive.(45) In any system some uninsured costs will inevitably rebound to the injured patient. But whereas tort systems seek to confine compensation to events in which negligence caused injury, no-fault schemes offer compensation to a wider class of events.(46) In the considerable terrain of medical injury types that lies between all adverse events and only negligent ones, the threshold for compensation in foreign no-fault systems varies.(47)

We believe that the Swedish example of basing eligibility for compensation on the avoidability of the event is the most rational basis to compensate, and also the one that best facilitates quality improvement in medicine.(48) While "avoidability" is a subjective determination like negligence, the Swedish experience suggests that their decision-making processes have been successful in applying it to separate eligible claims from ineligible ones. With the benefit of many years of experience, the Swedes are able to identify many types of claims that should reasonably be compensated on the strength of basic information about the circumstances of the incident and the nature of the disability suffered.(49) Similarly the New Zealand ACC scheme, despite its contrasting criteria, has demonstrated an ability to provide speedy claims resolution, thanks in part to adjudicators' decision-making experience with the types of injuries that attract compensation.(50)


The New Zealand and Swedish schemes include a "disability threshold."(51) The New Zealand's ACC scheme demands fourteen days in a hospital or twenty-eight days of significant disability. Sweden's Patient Insurance Compensation Fund demands ten days in a hospital or thirty sick days.(52) The two no-fault birth injury schemes here in the United States target severely disabled infants.(53) Such thresholds help to direct available resources to those patients who suffer the most severe harm. Significant administrative expenses may otherwise be devoted to adjudicating a large number of relatively minor incidents. Modest disability thresholds are unlikely to rule out many claimants who would have gained compensation in a malpractice system because only significant injuries will tend to generate the sort of contingent fees that justify the plaintiff's attorney's costs of initiating the case.(54)


The central claims facility would need to develop a scale of benefits, appropriately calibrated to the extent of disability caused by the medical injury. Patients suffering similar harm and losses should get similar reimbursement, a feat that the malpractice system has straggled to accomplish.(55) For the core benefits of a compensation package, we have proposed that patients would receive some reasonably high proportion of their economic losses, including lost income, out-of-pocket health care expenses, and lost household production.(56) In addition, states could choose to add pain and suffering compensation based on pre-established schedules or indexed to the economic damage award.(57) Some uniform payment in cases of death would also need to be included as component of the compensation package. …

Log in to your account to read this article – and millions more.