Why can't managed care companies be held liable for the consequences of their decisions when they refuse to authorize payment for care? Physicians, psychologists, and other health professionals, after all, can be sued for malpractice when they make bad decisions, and if they are found to have been negligent, they are responsible for compensating injured patients or their survivors. These days, with decisions about admission, discharge, and even outpatient care more likely to be made by managed care reviewers than clinicians, the former often hold the key to whether patients receive care. If they decide wrongly, and patients are harmed, should they not be held responsible for their negligent decisions?
Richard Clarke's family has been asking this question since the night in November 1994 when he was found dead in a parked car, with a garden hose leading from the tailpipe into the passenger compartment. Clarke, who had a drinking problem, was covered by a Travelers' health insurance policy purchased by his wife's employer, American Telephone & Telegraph. Eight months before his death, he was admitted to a community hospital for alcohol detoxification. The facility's recommendation for a 30-day inpatient program was referred to Greenspring, the managed care company that oversaw the mental health and substance abuse benefits in the Travelers' health plan. Although Clarke's policy specifically included one 30-day inpatient rehabilitation program per year, Greenspring rejected the request, authorizing instead a five-day hospitalization for detoxification.
Before the month was out, Clarke began drinking again. By September 1994, Clarke, the father of four children, was willing to seek voluntary hospitalization at a specialized treatment facility. Despite clinical recommendations for an extended stay, Greenspring declined to authorize more than eight days of inpatient care. Twenty-four hours later, after downing a considerable amount of alcohol, cocaine, and prescription drugs, Clarke locked himself in his garage, with the car running. His wife found him in time, and he was airlifted to an out-of-state hospital to be treated for carbon monoxide poisoning.
When medically ready for discharge, Clarke was involuntarily committed for 30 days of alcohol detoxification and rehabilitation. When Greenspring once more refused to authorize payment for care, he was sent to a state facility that is the traditional refuge of skid-row alcoholics. To no one's surprise, Clarke started drinking again soon after release. Within three weeks, he was dead.
Clarke's wife, Diane, filed suit against Travelers and Greenspring, alleging that they had been negligent in failing to approve medically necessary treatment for her husband's alcoholism. Had a treating physician declined to authorize extended hospitalization under similar circumstances, the case would have gone to trial, and Diane Andrews-Clarke's chances for a favorable judgment would have been good. But because an insurer and a managed care company were the target of her complaint, no jury will ever hear the evidence regarding their behavior. The case was thrown out of court by a federal judge (
1).
Greenspring and Travelers, the judge ruled, were protected from liability by a 1974 federal law, the Employee Retirement Income Security Act (ERISA) (
2). The statute was originally designed to protect the pension plans of workers by establishing national standards for funding and payment. Such a process required preempting all state laws on the subject. Almost as an afterthought, Congress extended ERISA's preemption of state laws to all laws affecting employee-benefit plans in general, including health plans (
3). Preemption affected not just statutes but also common law remedies such as actions in tort, of which malpractice is a subcategory.
Thus employees unhappy over how their benefits were being administered could no longer bring suit in state court; only the federal courts had jurisdiction over such claims. And the only remedies available to them were those specified by ERISA.
These remedies are embodied in Section 502(a) of ERISA, which allows suits to be brought for recovery of benefits that were improperly denied, or to enforce other rights under the statute. But it does not permit awards of compensatory or punitive damages, as would usually be available under state law. In practical terms, this means that if Richard Clarke, after having been denied coverage for hospitalization, had paid for care from his own resources, he could have sued after the fact to recover his expenses, assuming the courts agreed that he had been entitled to coverage. Such a suit, however, would be his only remedy. In a case where care is denied and a bad outcome ensues, ERISA does not provide recovery for pain and suffering, additional medical costs, lost wages, or—in this case—wrongful death (
4). In other words, Richard Clarke's family is out of luck.
By the time Clarke's suit was filed, the result could not have been unexpected. A series of federal circuit courts have rejected similar claims (
4). There is an irony here that should not go unnoticed. People who obtain their health plans other than from their employers, that is, by direct purchase from an insurer, or who are otherwise “unprotected” by ERISA, such as federal or state government employees, retain their state law rights to sue for negligence by an insurer or managed care company. However, these people represent only a small minority of persons with health insurance in the United States today.
Nor are negligence claims the only type of actions impeded by ERISA. The scope of most state efforts to regulate health care benefits—including the much ballyhooed state parity laws for mental illness—is substantially limited by ERISA's restrictions. For ERISA-related purposes, there are two types of health insurance plans: those purchased by employers from insurance companies (as in Richard Clarke's case) and those “self-funded” by the employer, an increasingly attractive option.
Self-funded plans, estimated to have more than 44 million members (
3), are immune from almost any form of state regulation. Mandated benefits, whether for mental health or postpartum care; antigag-rule laws; patients' rights statutes, including review and appeal procedures—all of these and more are precluded from application to self-insured plans by ERISA. Indeed, even the highly publicized new state laws allowing patients to sue health maintenance organizations (HMOs) for malpractice when coverage is denied are now being challenged under ERISA (
5). Plans purchased by an employer from an insurer differ only in that mandated benefits can be required, because such requirements are considered regulation of insurance, an area not preempted by ERISA.
Despite ERISA's impact on health care, most health professionals and patients are unaware of the statute. Federal laws with strange-sounding acronyms rarely become the focus of public attention. But it is difficult to argue with the conclusion that ERISA both deprives injured patients of reasonable recourse to those responsible for their situation and provides perverse incentives to managed care companies to deny care, secure in the knowledge that they cannot be sued for their actions, no matter how outrageous. Indeed, even the judge who threw out Diane Andrews-Clarke's claim against Greenspring and Travelers noted that “in the health insurance context, ERISA has evolved into a shield of immunity which thwarts the legitimate claims of the very people that it was designed to protect” (
1).
What can be done? Congress, which created ERISA, could amend the law to permit negligence actions against insurers and managed care organizations. As sensible as that might seem, a powerful coalition of groups is blocking ERISA reform on Capitol Hill. Large employers, whose plans are now covered by ERISA, fear that costs will increase if they or their agents can be sued for negligence. The insurance and HMO lobbies have similar concerns. Although proposals of this sort are made periodically, ERISA will have to be on the lips of voters in every congressional district in the country before Congress would be likely to act.
Wendy Mariner, a health law expert who tracks ERISA issues, has suggested that a more limited reform might be more palatable to business interests (
4). ERISA's own remedy in Section 502(a) could be expanded to allow recovery for compensatory damages when wrongful denials of coverage occur. This strategy would keep the actions in federal court and maintain uniformity of approach across the country, without denying injured patients recourse from the negligent parties. It is doubtful, however, that most industry representatives would be willing to embrace even this reform.
With Congress unlikely to act until a fire is lit under it, the courts are the forum of last resort. Some courts have been chipping away at ERISA's shield by allowing non-staff-model HMOs to be sued under theories of vicarious liability for the negligence of the physicians or other professionals who provide care (6-8). In essence, these courts have distinguished between decisions about whether benefits should be authorized—challenges to which are preempted by ERISA—and actions affecting the quality of benefits once it is agreed that they will be provided. Allowing suits addressing the latter issue has the paradoxical effect of continuing to immunize managed care organizations when they deny care altogether, while subjecting them to the risk of liability as soon as they agree to provide some coverage. The incentives here are hardly rational.
Mariner, however, has suggested that this approach could be expanded so that decisions about whether a particular form of treatment—most often hospitalization—is indicated would be considered decisions about the quality of care, rather than benefit determinations. “A decision that a service is (or is not) medically necessary…is also a decision that the service is medically sufficient for purposes of good patient care” (
4). This approach would allow the courts themselves to take the initiative in extending liability to managed care entities, even under the current version of ERISA. Whether the courts will be inclined to deviate from existing precedents and to stretch the statute quite so far remains to be seen.
For now, it seems clear the most important step toward change is to educate health professionals and the public alike about the importance of ERISA. That a statute passed almost a quarter-century ago for entirely unrelated reasons should be allowed to deprive states of the power to regulate health care and deprive patients of fair compensation for the harms they suffer is a parody of the legislative process. Something has to change.