From Medical Malpractice to Quality Assurance
A properly designed malpractice insurance system could actually decrease the prevalence of errors and enhance the overall level of care.
Every decade or so, the United States is seized with a fervor to reform medical malpractice. Unfortunately, this zest is typically motivated by circumstances that have little to do with the fundamental problems of medical malpractice, and the proposed changes to the system do not address the true flaws. A well-functioning malpractice system should focus not only on how to compensate patients for medical errors but also on how to prevent these errors from occurring in the first place.
The United States has faced a medical malpractice “crisis” three times since 1970. Each of these crises was precipitated by conditions that created a “hard” market: decreased insurer profitability, rising insurance premiums, and reduced availability of insurance. And each time the crisis became a polarized battle between trial lawyers on one side and organized medical groups and insurers on the other. On the one side, stakeholders link the crisis to “runaway juries” and “greedy lawyers.” On the other are those who blame interest rates and possibly insurer pricing practices. If one attributes the crisis to falling interest rates and bad investments in the stock market, the policy implications are markedly different than if soft-hearted and cognitively limited juries and ambulance-chasing lawyers are blameworthy.
In the end, calm is returned, but the situation of patients is not improved. We are left with a system in which most victims of medical error are not compensated for their losses and in which the overall quality of care is not what it might be.
As a first step in tackling the real problems of medical errors and mediocre quality assurance, we need to debunk the popular misconceptions about the problems with the medical malpractice system. Once these ferocious but ultimately pointless conflicts are defused, we can begin to think about fundamentally reconstructing the system with an eye toward improving the quality of care by giving practitioners effective incentives to deliver the services that people need. There are a variety of options for reform; one of them, called enterprise insurance, has the potential to provide the initiative for systemic change.
Many myths about medical malpractice dominate the public discourse. These myths reinforce misinformation and are used to justify statutory changes that benefit certain stakeholders but are not in the broader public interest. Five of the most common are: medical care is costly because of malpractice litigation; only “good” doctors are sued; there are too many medical malpractice claims; dispute resolution in medical malpractice is a lottery; and medical malpractice claimants are overcompensated for their losses.
The high cost of personal health services in the United States is frequently attributed to litigation and the high cost of malpractice insurance. This assumes that premiums and outlays for awards have risen appreciably and constitute a major practice expense. The data, however, do not show appreciable increases over long time periods. Between 1970 and 2000, mean medical malpractice premiums went from 5.5 to 7.5% of total practice expenses. This is not the case for damage awards; payment per claim has increased substantially since the mid1990s. However, relationships between medical malpractice premiums, claims frequency, mean payment size, and total payments are complex and assumptions should not be made based on a single indicator.
Some critics of medical malpractice contend that being at the cutting edge technologically makes a physician more vulnerable to being sued. There is no empirical evidence that being sued is an indicator of superior performance. However, there is evidence that physicians with no claims histories were rated by their patients as being, or at least appearing to be, more understanding, more caring, and more available. Overall, it is untrue that only good doctors are sued, but at the same time, being sued is not a marker of being a bad doctor either.
The myth that there are too many malpractice claims is a bit more complex. There are two path-breaking studies showing that there are both too many and too few malpractice claims. The first of these studies was conducted in California in 1974. The second, the Harvard Medical Practice Study, was conducted in New York in 1984. In both studies, surveys of medical records of hospitalized patients were conducted to ascertain rates of adverse events attributable to provision of medical care to these patients and rates of adverse events due to provider negligence, termed “negligent adverse events.” The California study revealed that of the 5% of patients who experienced an adverse health event while in the hospital, 17% suffered a negligent adverse event. In New York, the corresponding rates were 4% for adverse events, of which 28% were negligent adverse events. The authors found that “invalid” claims, those not matching the study’s determination of liability, outnumbered valid claims by a ratio of three to one. However, they also found that only 2% of negligent adverse events resulted in medical malpractice claims. There were 7.6 times as many negligent injuries as there were claims. Thus, there were errors in both directions: Individuals filed too many invalid claims and not enough valid claims.
The public’s view of juries leads to the inference that outcomes of litigation are often random. Actual data, however, leads to the opposition conclusion: Outcomes are not random. There is a definite relationship, albeit an imperfect one, between independent assessments of liability and outcomes of legal disputes alleging medical malpractice. One study estimated that payment is made in 19% of malpractice claims when there is little or no evidence of errors. In contrast, when the evidence of an error is virtually certain, payment occurs 84% of the time. Using the results of this study, claims not involving errors accounted for 13 to 16% of the system’s total monetary cost. The way one views this percentage (substantial or small) depends on where one draws the line between error and no error. Unfortunately, the New York study conclusions do not stress or even mention that the estimates of error are subject to a very high degree of uncertainty.
Similar to the myth that malpractice claims are decided without regard to evidence of negligence, the myth that most plaintiffs are overcompensated for their injuries is pervasive. However, a comparison between the cost of injuries incurred by claimants and compensation actually received revealed that medical malpractice claimants on average are undercompensated. In one study, compensation exceeded cost by 22% for claimants who received compensation at verdict, whereas 26% percent received no compensation at all. On average, including those cases for which no compensation was received, compensation amounted to about half of monetary loss. Even including compensation for nonmonetary as well as monetary loss, compensation fell far short of injury cost. Nevertheless, this does not eliminate the possibility that compensation was excessive in selected cases.
Reconstructing the system
In principle, medical malpractice should be a quality-assurance mechanism; in practice, it falls far short of achieving this goal. For one thing, there is no empirical evidence that the threat of medical malpractice makes health care providers more careful. Also, meting out compensation is very expensive. Sadly, medical malpractice “tort reform” has aimed to save medical malpractice premium dollars rather than make it an effective mechanism for assuring quality and efficiently compensating injury victims. For example, a popular but misguided tort reform, caps on damages, has worked to reduce payments by medical malpractice insurers and create premiums below what they otherwise would have been, but caps have not altered the incentives, except perhaps to discourage attorneys from representing medical malpractice plaintiffs, even those with valid claims. If there is a benefit to caps, it is in redistributing income from injury victims and their attorneys to health care providers rather than in improving quality of care or markedly reducing rates of unnecessary tests and health care costs more generally. It seems unlikely that any savings in medical malpractice insurance premiums would accrue to patients as taxpayers and health insurance premium payers. Organized medicine plausibly supports caps primarily as a response to pressures from its constituency for financial relief.
Although the current system has many flaws, there is also a brighter side. First, contingency fees for plaintiffs’ attorneys give patients who are unsatisfied with outcomes a mechanism for addressing their grievances that may not be possible through other channels. The regulatory apparatus, which has a responsibility for safeguarding the quality of personal health services, is sometimes controlled or substantially influenced by health care providers and health care regulators who may be unresponsive to patients’ complaints. Second, the U.S. jury, despite its limitations, gives ordinary citizens a role in the dispute-resolution system. Although jurors are only rarely scientists, physicians, or other health care professionals, they reflect society’s values. Third, even during the crises when substantial increases in malpractice premiums occurred, the premiums remain a tiny component of total health care costs. Viewing long-term secular trends in medical malpractice payments and premiums rather than the short time periods during which there has been substantial growth in premiums reveals that increases in payments and premiums are rather modest, only slightly higher than the changes in prices in general. Finally, the current malpractice system does a good job of identifying some real errors.
However, the current system has serious deficiencies, just not the same as those typically depicted in the media. First, unlike other fields of personal injury tort, there is no empirical evidence that the threat of medical malpractice lawsuits deters injuries. This is a very serious deficiency, particularly because injury deterrence is typically listed as a goal, perhaps the primary goal, of tort liability. Second, tort liability focuses on the mistakes of individual providers, but errors frequently reflect simultaneous omissions or misjudgments on the part of several individuals. Third, most medical errors do not result in malpractice claims. As a result, the signal from tort to health care providers is insufficiently precise or even wrong. Fourth, compensation to injured patients is typically less than what they deserve based on the loss attributable to their injuries. Litigation is an extremely inefficient system for compensating injury victims. Various types of insurance, such as health and disability insurance, are much more efficient in distributing compensation to persons who have incurred a loss from receiving less than appropriate care.
Finally, health care providers in the United States largely reject the view that medical malpractice has a constructive role to play in health care delivery. Providers generally see no link between medical malpractice litigation and provision of low-quality care. Much commentary in assessments of medical malpractice and patient safety see medical malpractice as part of the problem rather than part of the solution. This misconception is an important roadblock because malpractice claims often arise from deficiencies in care.
Thus, medical malpractice does badly on injury deterrence, improved patient safety, and compensation of persons with medical injuries. Its strongest features are giving injury victims a day in court and making professionals accountable to ordinary citizens.
Patient safety and medical malpractice are inextricably linked. However, neither market forces nor the threat of tort liability seem to provide sufficient incentives for quality assurance. An important reason that the threat of lawsuits has not improved patient safety is that medical malpractice insurance shields potential defendants from the financial burden of being sued. Such insurance tends to be complete; there are no deductibles or coinsurance, and liability limits of coverage are rarely exceeded. Medical malpractice premiums tend not to be based on a physician’s own history of lawsuits. Thus, a physician with many past lawsuits may pay the same premium as a colleague who has never been sued.
Shortcomings in medical malpractice are not wholly responsible for shortcomings in the quality of health care in the United States. U.S. airlines have implemented very effective safety procedures. In other sectors, market forces provide some guarantee of quality. For example, there is no quality crisis in the hotel market. To the extent that consumers demand high-quality hotel rooms, this is provided by the market.
However, there are few means for consumers to inquire about the quality of a hospital or doctor, let alone demand high-quality health care. Employers often speak about quality assurance, but with few exceptions, medical care is not their principal business. Given the limitations of market forces in pressuring providers to supply high-quality care, there is indeed a role for government regulation and private regulatory mechanisms, such as peer review and tort liability. These mechanisms are not substitutes for the market but rather complements to it.
Options for reform
Meaningful tort reform should take account of the fact that many medical errors are not simply errors of individuals; they are errors of systems. Also, health care providers must have financial incentives to exercise care and implement meaningful quality-assurance mechanisms.
Overall, what have been called tort reforms have been short-term fixes, which do not improve system performance. In recent years, the reform most favored by physicians, hospitals, and insurers has been caps on damages. Caps have the effect of lowering payments per paid claim and probably discourage some trial lawyers from representing some medical malpractice plaintiffs. But they do not fundamentally change how medicine is practiced.
Scholars, other experts, and some policy analysts have proposed more sweeping reforms of the current system. They include no-fault insurance, health courts, alternative dispute resolution, private contracts, scheduled damages, enterprise liability, and enterprise insurance. Each proposal has advantages and disadvantages, and no one reform provides an exclusive remedy to the problems with the medical malpractice system. Of these options, however, enterprise insurance has, in our view, the potential to initiate systemwide change.
No-fault insurance. No-fault approaches are designed to be substitutes for tort, providing compensation regardless of fault. Currently, no-fault is widely used as a substitute for tort in auto liability and workers’ compensation. Medical no-fault has been implemented in only two states, Florida and Virginia, and for only a few medical procedures. The low administrative expense and faster payment of damages makes no-fault insurance an attractive alternative. But in the Florida and Virginia the programs were implemented to achieve savings in medical malpractice premiums rather than to distribute compensation to a larger number of medical injury victims. Revenue for these programs comes from physicians and hospitals. But if the system is truly no-fault, why should physicians and hospitals be the only parties taxed to fulfill a broad social obligation to compensate those with misfortunes? It seems more appropriate to tax the public at large, and no U.S. state has agreed to do this.
At least as interesting an alternative, not under active discussion, is private no-fault insurance. A hospital with an effective quality-assurance program could offer no-fault insurance to its patients for a reasonable premium, with an even lower premium for patients who agreed to forego filing a tort claim in the event of an injury. To the extent that the hospital had an effective quality-assurance program, the savings in premiums could be passed through to its patients.
This type of voluntary no-fault program would offer several important advantages to hospitals and their medical staffs. First, it might relieve providers of the threat of tort. Second, offering no-fault benefits would be a signal to consumers that the hospital has an effective patient-safety program and low rates of medical errors. Third, to the extent that injury victims value quick payments with little involvement of attorneys, this too would increase demand for the hospital’s services.
Although hospitals may anticipate some savings, it is essential that such no-fault coverage extend to a large number of conditions. When exclusions from coverage are necessary, they should be broad and easily understood by patients. Very narrow thresholds are difficult for patients to assess in advance of injuries. A few very costly procedures may be excluded from coverage, but it would be important that these be listed and described in understandable terms in advance.
Complete substitution of no-fault for tort is infeasible; however, a system in which patients would contract for no-fault coverage well in advance of receiving care at the hospital is more reasonable. Contracting in advance is essential to avoid situations in which a patient is faced with the option of contracting for no-fault at point of service, which could be interpreted as an adhesion contract, a standard form or boilerplate contract entered into by parties with unequal bargaining power. One way to partially avoid the unequal bargaining power is to allow employees to designate whether or not they wish to substitute no-fault for tort when they are choosing their health insurance plans. Surcharges for no-fault (if surcharges are imposed on patients) could then be built into the premium charged. In the case of voluntary no-fault, because insured patients would agree not to sue under tort, the savings in tort payments would offset at least part of the cost of the no-fault plan. No-fault plans would require prior regulatory approval, depending on the applicable regulatory authority. Regulators would pay attention to the method of enrolling persons into the plan, pricing, and issues bearing on plan solvency.
Health courts. Medical care is a technical subject, and proponents of health courts argue that judges and juries are often not well positioned to deal with the complexities. In addition to providing victims with consistent, fast, and relatively easily obtained compensation when warranted, health courts are also intended to reduce cost by streamlining the process, maintaining consistent medical standards, and capping or scheduling damages.
Full-time judges are a major feature of the health court proposals. The judges would deal only with malpractice cases, and there would be no jury. In one proposal, specialized judges would shape legal standards for medical malpractice, creating a body of science-based common law that healthcare providers could rely on when making treatment decisions. In theory, a body of science-based common law seems valid and useful, but it raises issues of its own. In the context of health courts, standards for medical practice would develop under state law. Yet, without federal regulation, each state would be free to develop its own standards, allowing for variable legal and medical practice standards among states.
Although concerns about the inadequacies of juries to decide technical matters and the inexperience of judges in medical matters provide the main rationale for health court proposals, what is not acknowledged in the policy debate is that the concerns about lay juries and judges apply to the much larger issue of the use of scientific evidence in the courts. Health courts represent only one of several overlapping alternatives for addressing this issue. Other alternatives include use of court-appointed experts, bifurcated trials, use of special masters, specially convened expert panels, blue-ribbon juries, and alternative dispute resolution.
We agree that health courts have attractive features but are reluctant to give this option our enthusiastic endorsement. Preserving juries in some form, even if they are blue-panel juries, would provide a broader representation of perspectives and values than would sole reliance on a narrow group of professionals to make judgments on specific cases. Even a judge with health expertise will not be able to be expert on the full range of issues health courts are likely to confront. In the end, it is important that plaintiffs as well as defendants view health courts as legitimate. If the court consists entirely of or is dominated by physicians and other health professionals, buy-in by plaintiffs, and society more generally, seems highly improbable.
Alternative dispute resolution. Dispute resolution under the trial-by-jury system is extremely costly. Thus, alternative approaches that streamline the process seem attractive. Generally, alternative dispute resolution is made up of any means of settling disputes outside of the courtroom. The two most frequently used forms are arbitration and mediation. Arbitration is a simplified version of litigation (there is no discovery and the rules of evidence are simpler). Mediation uses an impartial third party to facilitate an agreement in the common interest of all the parties involved.
The main advantage to alternative dispute resolution is that the process tends to be speedier than a trial. The advantages of arbitration include lower cost, private proceedings, more flexibility than in a court, and when the subject of the dispute is highly technical, the appointment of arbitrators with the appropriate degree of expertise. There are also disadvantages. The parties need to pay for the arbitrators, and the rule of applicable law is not compulsory. With binding arbitration, the decision reached is comparable to a jury verdict and can be overturned only if there is evidence of malfeasance in the process of reaching a decision. In mediation, sessions are not decided in favor of one party or another, and the parties are not bound to resolve their dispute and may pursue litigation if dissatisfied with the results of mediation. Although speed in dispute resolution and lower cost are advantages, there is some empirical evidence that this actually leads to more lawsuits.
Private contracts. The rationale for private contracts between health care providers and patients, as a substitute for tort, is that tort liability determines compensation based on standards of care that may differ from those that patients might prefer. Private contracts might set out specific circumstances in which providers might be responsible for compensating injury victims, scheduling damages, and specifying alternative dispute resolution mechanisms when disputes arise.
The strength of private contracts is that they can reflect preferences of individuals. Individuals with higher willingness to pay for safety pay more for such care. However, individual choice opens the door to adverse selection. That is, persons who are prone to suffer an injury because their health is more fragile may be more willing to pay for contracts offering extra precautions.
Opponents of private contracts point out that the relationship between the patient and provider is not one of equal power. A hospitalized patient or even an outpatient may not be well positioned to negotiate with a physician. Courts have overturned contracts reached at the point of service for this reason. But this is not when contracting would occur. Rather, as with a voluntary no-fault plan, contracts could be options offered to persons at the time they enroll in a health plan. Agreement to a lower standard of care or a less generous schedule of damages would result in a lower premium.
Scheduled damages. Rather then set a limit on the maximum size of an award, establishing a schedule of damages sets payment criteria for all awards, not only the large ones. Because scheduling affects the whole distribution, it is conceptually superior to flat caps on grounds of equity of payment to claimants with very severe injuries relative to those with less severe injuries. The trial bar opposes this approach because it would limit the ability of plaintiffs to make a case for their special circumstances. Such flexibility, however, must be measured against the vertical inequities of caps; that is, caps limit large awards for the relatively serious injuries but do not directly affect payment for more minor ones.
An anticipated objection to scheduled damages is that they limit jury discretion in awarding damages. However, there is a tradeoff between complete individualization of awards and reducing volatility and increasing predictability of awards. It would be appropriate for states to review the instructions that are provided to juries in order to assess whether guidelines for determination of monetary loss should be developed. In the end, however, even though scheduled damages are preferable to caps, the link to quality assurance is at best indirect.
Enterprise liability. Enterprise liability is a means of aligning the incentives of providers and of accounting for the fact that many errors arise from defects in systems rather than in individual providers. Because many medical injuries occur during receipt of hospital care, it makes sense to start the alignment process with hospitals and the physicians who work there. Under enterprise liability, when the receipt of care is in a hospital setting, the hospital would be named as the defendant in medical malpractice lawsuits. Separate suits against individual physicians would not be filed. If the hospital were the only named defendant, it would have a greater incentive to adopt quality-assurance measures, including for outpatient care.
Left unsaid in general discussions of enterprise liability is how the burden of hospital premiums would be shared. It would be advisable that physicians bear some part of the premium burden to provide some incentive to avoid claims. Hospitals could implement their own systems of surcharging physicians with many medical malpractice claims. Of course, hospitals, or medical staffs operating on their behalf, would retain the option of removing from their staffs physicians with adverse claims experience or those who do not comply with hospital patient-safety regimens. In fact, hospitals would have a greater incentive to monitor physician performance and remove physicians with adverse claims experience.
With hospital enterprise liability, the deterrent would be internalized to the hospital, establishing a clear financial incentive for quality improvement and error reduction. These organizations could impose a combination of financial and nonfinancial incentives for individual physicians to prevent injuries, coupled with increased surveillance measures. Also, the hospital and physicians at the hospital collectively would have an incentive to promote patient safety, because the enterprise’s premiums would depend on future anticipated losses from medical malpractice claims.
There are several possible objections to enterprise liability. First, plaintiffs might view hospitals as rich and faceless institutions with deep pockets, thus increasing plaintiffs’ demands for compensation. Of course, under the current system, insurers presumably could be said to have deep pockets as well.
Second, enterprise liability may restrict patient choice of provider. Physicians may have to limit their admissions to the one hospital at which they receive medical malpractice coverage. Physicians frequently have privileges at more than one hospital. This potential concern can be largely remedied by limiting physician coverage to care delivered within the walls of the facility under the hospital’s policy. Thus, if a physician practiced at three hospitals, he or she would be covered under three hospital policies. In addition, the physician would need to obtain medical malpractice insurance for care delivered in the office, but such coverage would be at a greatly reduced premium.
Third, physicians already complain about their growing loss of autonomy, and enterprise liability would probably exacerbate this trend. Physician autonomy is important because it allows providers to use their professional skill and judgment in particular situations. Outsiders, such as hospitals, may not be well positioned to know all the details and considerations of a physician-patient interaction.
Fourth, inpatient care is shrinking as a share of the total personal health care dollar. Because more care is being delivered outside the hospital, using the hospital as the locus of liability may not be ideal. But this concern disregards another trend. Hospital-provided ambulatory care is growing, and hospital enterprise liability would encompass care at all sites at which the hospital organization or system provides care.
Nevertheless, enterprise liability addresses many current deficiencies, especially the insufficient incentives providers have to invest in patient safety. A major barrier to implementation is the lack of a political constituency at the federal and state levels. Health care consumers are not well organized, and providers appear to be concerned with the “deep pockets argument,” as well as the loss of professional autonomy that may accompany enterprise liability.
Enterprise insurance. Another approach, enterprise insurance, does not change the cause of action against physicians and hospitals nor does it change the named defendants. Rather, physicians who render services to patients in hospitals would obtain their malpractice insurance through the hospital. Large organizations could self-insure for medical malpractice. Because all members of the pool would stand to lose from the provision of substandard care, there would be organizational incentives to monitor quality and implement quality-improving systems of care. For example, a hospital whose obstetric staff is sued repeatedly would have a direct financial incentive to take actions to deal with the causes of the lawsuits.
This approach seems very promising, but it too faces obstacles. In particular, in the United States, in contrast to other high-income countries and professionals in U.S. industries such as airline pilots, hospital medical staffs have been largely independent of hospitals. Physicians have resisted being under the control of hospitals, for financial reasons and out of concern for loss of professional autonomy. Any proposal from the “outside” that would cede control of medical decisionmaking to hospitals is likely to be resisted by many physicians. The key will be to have active physician involvement in hospital-based enterprise insurance. Smaller hospitals would face special challenges because they might be too small to operate a medical malpractice insurance plan on their own. Such hospitals might join regional compacts.
Finally, accountability incentives alone are not likely to provide sufficient motivation for hospitals to create systems management of medical injuries. Hospitals and physicians have many non-liability objectives and concerns. Implementation of enterprise insurance alone may not lead to optimal levels of patient safety in hospitals. Still, enterprise insurance is an attractive solution because it provides those in the best position to improve care with an incentive to introduce patient-safety measures.
Enterprise insurance creates efficiency by combining patient-safety measures and insurance, including premium setting. Because the insurer, in this case the hospital, is better able to “poke inside” the clinical organization to understand the source of errors, the insurer may be less likely to raise premiums dramatically because it has a better sense of what is going on. As with enterprise liability, hospitals would have added incentives to be selective about the quality of physicians they admit to and retain on their medical staffs. In turn, medical staffs have a much more direct incentive to support adoption of patient-safety measures in order to reduce medical malpractice losses at the hospital, especially if the medical staff is placed at some risk for losses above a threshold value.
Enterprise insurance has its limitations, but it also has the potential to provide the initiative for systemic change. By combining the function of preventing injuries with that of insuring loss if and when injuries do occur, the way to injury prevention is combined with the willingness to do so. Nevertheless, the medical malpractice apparatus, with or without enterprise insurance, should be seen as only part of the quality-assurance process. It cannot do the job on its own.
Frank Sloan ([email protected]) is J. Alexander McMahon Professor of Health Policy and Management and professor of economics at Duke University. Lindsey Chepke ([email protected]) is research associate at Duke’s Center for Health Policy. They are authors of Medical Malpractice (MIT Press, 2008), from which this is adapted.