Managed Care at the Millenium:
Scenes from a Maul
David A. Hyman
How should one assess the managed care backlash? Is it a case of the medical empire striking back against attempts to constrain professional autonomy and the ever-increasing flow of funds into the health care sector of the economy? Is it a rational response to the failure of more global reform? Is it more or less what managed care deserves, given its performance to date? Is it a problem of perception, as people generalize from unrepresentative horror stories? Is it simply political opportunism? Is it a combination of all of the above--and if so, in what proportions? Unfortunately, the general approach is to respond to such questions with platitudes that managed care is "unaccountable" and subscribers lack "voice," or that the proposed legislation is "reactionary" and "big government." It is difficult to obtain a valid assessment of the merits of the managed care backlash on the basis of such sweeping generalizations. A better strategy is to systematically examine the specifics of the reported problems (including an assessment of their severity and frequency) and the costs and benefits of the various proposed reforms.
To be sure, it is a daunting task to approach the managed care backlash in this way. A virtual avalanche of bills have been proposed in the past few years, and the nature of the legislation has evolved, from obvious anti-managed care initiatives (e.g., "any willing provider" legislation) to direct "regulation by body part" (e.g., prohibitions on drive-through deliveries) to complete regulatory frameworks encompassing dozens of provisions (e.g., mandatory external appeals, liability rights, and making managed care organization [MCO] medical directors subject to the disciplinary [End Page 1061] authority of state licensing boards). However, the alternative--to take on faith the severity of the problem and its frequency, and the merits of the proposed reform--has little to recommend it. In a series of articles, I have evaluated a variety of consumer protection initiatives which enjoy overwhelming popular support (Hyman 1998a, 1998b, 1999a, 1999b). These articles demonstrate that it can be quite misleading to assess the merits of even sensible-sounding consumer protection initiatives from a distance.
General Considerations
There are a host of imperfections in the markets for health care and health insurance. Information on quality and coverage is hard to obtain, and harder to interpret. Consumers are difficult to organize, and their ability to get the coverage terms they want is limited by the aggregated nature of health insurance coverage, their bounded rationality, and their limited ability to "shop around" in an insurance market with restrictions on portability and preexisting conditions. An additional layer of complexity is added by the structural reality that most employed Americans secure their insurance coverage through their place of employment (and receive a significant tax subsidy for doing so). Finally, there are the distortions which result from the regulatory and common law vacuum created by the Employee Retirement Income Security Act (ERISA). These market imperfections significantly strengthen the case for some form of regulatory intervention.
At the same time, there are equally compelling reasons to be skeptical about the merits of the managed care backlash. To date, most of the targets of "reform" have been identified on the basis of bad anecdotes and the political appeal of the relevant patients and conditions--a strategy that is hardly a recipe for sensible public policies. Indeed, it is clear that designing the policy agenda on the basis of the claim that particular identifiable lives are in jeopardy creates major distortions (Havighurst, Blumstein, and Bovbjerg 1976). In addition, many of the problems with managed care that have been identified are far more complex than is commonly acknowledged, and the solutions are rarely self-evident. Few legislators have the necessary training or inclination to weigh the (often conflicting) evidence on the benefits of any given consumer protection. Evidence on the costs of any given consumer protection is frequently unavailable, and estimates are subject to considerable uncertainty. In a voluntary insurance market, cost-increasing consumer protections (which [End Page 1062] are what has emerged to date) will predictably price some people out of the market--and it is hardly self-evident where the cost/quality/access equilibrium should be set, let alone whether there should be a single standard for all coverage. The drafting of consumer protections is also readily hijacked by entrenched providers, who have their own interests at heart. Finally, there is the emotional overlay accompanying health care issues, and the off-budget feature of many of the reforms.
Thus, although general principles provide a prima facie case for regulation, there are good reasons to be concerned about the merits of the provisions likely to emerge from the legislative process. Indeed, such regulations should be closely scrutinized, to ensure that they are actually a public-regarding improvement on the status quo. The balance of this article briefly summarizes my findings with regard to three of these initiatives: the prohibition on "gag clauses," mandated coverage of certain visits to an emergency department, and the prohibition of drive-through deliveries. Each of these initiatives enjoys broad bipartisan support, but all are problematic.
Gag Clauses
Legislation in this area was prompted by several examples of physicians who were terminated or reproached after they had spoken out against particular MCO practices and by statements from physician groups that gag clauses were an unethical interference in the physician-patient relationship (Pear 1995; Woolhandler and Himmelstein 1995). Although managed care representatives denied that their contracts included gag clauses and voluntarily inserted antigag language, sweeping legislation quickly prohibited any retaliation by an MCO against a provider on account of any communications with a patient regarding health-related matters.
An evaluation of this issue must begin with a taxonomy of the varying provisions which could be characterized as gag clauses. The purest form of gag clause is one which prohibits a health care provider from revealing potentially beneficial but uncovered treatments to the patient. A less exacting gag clause restricts the timing of discussion of treatment alternatives until authorization for a particular treatment has been secured from the MCO. Other contractual terms which have been described as gag clauses include general confidentiality provisions (which prohibit the disclosure of information regarding compensation arrangements, utilization data, quality assurance procedures, and the like), nonsolicitation [End Page 1063] provisions (which restrict the ability of the provider to advise patients to disenroll from the MCO, switch to another MCO, or approach patients if the provider is terminated by the MCO), and antidisparagement clauses (which purport to limit the ability of the provider to criticize the MCO).
For all the complaints that have been raised about gag clauses, providers have yet to present a pure gag clause, let alone proof that any of these provisions are being enforced in a way that systematically restricts communications between providers and patients. The U.S. General Accounting Office (GAO) determined that there were no pure gag clauses in any of the 1,150 contracts from 529 HMOs they examined (GAO 1997). The GAO noted that confidentiality, antidisparagement, and nonsolicitation clauses were found in a majority of the contracts they examined, but that such provisions are standard business terms--and a majority of the contracts that included such provisions also included express statements that nothing in the contract was intended to prevent or limit discussions between physician and patient. Physician groups have complained about the "chilling effect" of confidentiality, antidisparagement, and nonsolicitation clauses, but the GAO found that contracts from physician-owned-and-operated MCOs used such clauses with approximately the same frequency as non-physician-owned MCOs. The GAO concluded that these provisions "are not likely to have a significant impact on physician practice . . . because physicians are not fully aware of them, do not interpret them as hindering communication, or choose to disregard them" (ibid.: 13).
In reality, the gag clause controversy is a stalking horse for the unhappiness of providers with their loss of clinical autonomy, and their newfound realization of the extent to which they are economically dependent on MCOs. Indeed, the GAO's report on gag clauses concluded that it is the contractual relationship between MCOs and providers, with "its short duration and provision for termination without cause--that may make physicians feel constrained from speaking openly with their patients" (ibid:15). Even if one views this issue as one requiring legislative redress, a prohibition on gag clauses does not address it in any meaningful way.
Regardless of the evidence on the prevalence of gag clauses, it is hard to be against full communication between providers and patients. Indeed, such communication would seem to be a necessary precondition to an effective physician-patient relationship. However, there is ample precedent for such gag provisions. The Supreme Court has expressly authorized the use of pure gag clauses by the federal government to restrict the communication of information from health care providers to patients [End Page 1064] (Rust v. Sullivan 500 U.S. 173 1991). Providers are free to refuse federal funding, but if they accept the funding, the federal government has broad latitude to fashion an associated gag clause.
In addition, anti-gag clause provisions do have real costs, especially if they are drafted broadly. Such statutes may actually encourage strategic behavior by providers. Indeed, it is likely that future terminations of providers, as well as modifications in the terms on which MCOs are willing to deal with particular providers, will give rise to claims of retaliation for violation of a gag clause. The irony of such statutes is that they give providers an incentive to complain loudly and publicly about managed care, because doing so increases the probability that any adverse action by the MCO will be deemed a response to such complaints. The costs of sorting out such cases are ultimately reflected in the premiums that are charged for coverage, and the percentage of the premium spent on medical care. A prohibition on gag clauses also does nothing to address the more far-reaching failure of providers to disclose to patients the incentive structure under which they operate--and makes it less likely that problem will ever be solved.
Finally, even though gag clauses have been widely condemned, at some point in the future there may well be a place for them in the coverage market. There is a significant problem with inappropriate medical care in the United States (President's Advisory Commission 1998). Ex ante, if patients believe that their MCO is likely to do a better job than the average physician in identifying inappropriate treatments, a pure gag clause helps prevent the costs which would otherwise result from symbolic blackmail of the MCO by physicians and patients. As yet, there is no evidence that patients trust their MCO more than their physician, but patients may eventually decide that they have been symbolically blackmailed enough--and may come to embrace gag clauses as a "Ulysses contract" to lower the incidence of such conduct (Hyman 1998a).
Access to Emergency Care
MCOs have taken a variety of steps to limit subscriber access to emergency departments ("EDs"). Physicians and patients have complained that MCOs employed widely differing definitions of what constituted an emergency, required preauthorization even though that strategy made no sense in dealing with emergencies, retrospectively denied coverage based on discharge diagnosis, and generally second-guessed the judgment of ED physicians. A number of high-profile incidents in which catastrophic [End Page 1065] consequences reportedly followed MCO refusal to authorize a visit to the ED or denied coverage for an unnecessary visit effectively defined the issue (Hyman 1998b). The legislation which resulted at the state and federal level inter alia required MCOs to cover the bill if a prudent/reasonable layperson would have gone to the ED, and restricted the use of preauthorization.
At first glance it is intuitively appealing to make ED coverage contingent on prudence and reasonableness. Unfortunately, words like prudence and reasonableness require the MCO to apply a largely subjective standard, which is only second-guessed when coverage is denied. Although this solution may appear to finesse the coverage issue, its indeterminacy actually has the potential to cause severe secondary disputes. If patients want increased ex ante certainty about which visits to the ED will be covered, the prudent/reasonable layperson standard will probably not make things better--particularly when the enforcement of the standard must be left in the hands of the MCOs in the first instance. In addition, the prudent/reasonable layperson standard is an extremely overinclusive coverage term if the current high level of inappropriate ED utilization is any guide (Hyman 1998b). Coverage which includes this term will be more expensive than more restrictive alternatives --especially if coupled with limitations on the use of copayments and deductibles. Finally, if reasonableness is such an excellent and cost-neutral standard, then why did Congress abandon its use of reasonable cost reimbursement under Medicare in 1982 by adopting the prospective payment system, and why did the nation's governors lobby Congress for years to eliminate a similar term forcing state Medicaid programs to reimburse the reasonable costs of providers? The reluctance of the states and Congress to accept a reasonableness standard when they are footing the bill counsels caution when they propose imposing a similar standard on private parties.
In like fashion, at first glance, it seems to make little sense for an MCO to require preauthorization in the case of an emergency. Yet, as noted previously, many people go to the ED for nonemergencies. Requiring advance authorization discourages such conduct and allows the MCO to arrange for an alternative source of treatment for patients who do not require emergency care, or at least forces members to self-insure for the resulting expense if they do not obtain such approval. Many patients get better care at lower cost as a result of such practices.
Some statutes allow the MCO to require preauthorization, but only if [End Page 1066] the MCO responds to calls within a certain amount of time, or maintains twenty-four-hour coverage. Such provisions have the effect of conferring a significant advantage on larger and more established MCOs. Perhaps people only want to purchase coverage from such MCOs, but there is no a priori reason to think so. Each MCO offers its own mix of providers, restrictions on access, coverage, and cost. Against this diversity of arrangements, there seems little reason for "consumer protection" laws to be used to dictate the terms on which ED care is provided--unless, of course, the goal is to use such laws to restrict competition from smaller and less-well-established MCOs.
Drive-Through Deliveries
By the end of 1996, approximately thirty states and the federal government had enacted laws mandating coverage of forty-eight hours of hospitalization following a vaginal delivery and ninety-six hours of hospitalization following a delivery by Cesarean section (Hyman 1999a). A shorter stay was possible if the attending provider, in consultation with the mother, agreed--but positive and negative inducements encouraging early discharge were prohibited. As with gag clauses and access to emergency care, some highly unrepresentative anecdotes defined the issue for the public (ibid.).
Despite wide popular and legislative support, the case for an extended postpartum stay is actually extraordinarily flimsy. There is little or no evidence justifying extended postpartum stays, or indicating that postpartum stays of the specified length provide a significant benefit (Braveman et al. 1997; GAO 1996; Hyman 1999a). As one prominent commentator noted, "It is reasonable to conclude that discharging apparently well newborns from the hospital before the third day of life, at least in the absence of documented substitute services at an alternate site, is likely to result in moderately but not dramatically increased risks of hospital readmissions . . . [but] the difference between a postpartum stay of 24 hours and a stay of 48 hours is unlikely to be a critical determinant of newborn or maternal health outcomes" (Braveman et al.: 334).
Even if postpartum stays of the specified length provide a benefit, it does not follow that the benefit justifies the associated cost, or that the same results cannot be achieved in some other way at lesser cost. It is quite revealing that approximately two-thirds of the states which enacted legislation prohibiting drive-through deliveries excluded state employees [End Page 1067] and Medicaid recipients (who account for approximately 40 percent of births in the United States) from its protections. The only thing these groups have in common is that their medical expenses are an on-budget expense. The federal legislation included state employees but excluded Medicaid recipients, although subsequent legislation added Medicaid recipients so long as they were participating in a managed care arrangement. Thus, legislative opposition to drive-through deliveries disappeared the moment it became apparent that fixing the problem would result in on-budget expenses. The legislation also created perverse incentives for the coverage of postdischarge services and coverage decisions in general, and does not appear to coincide with the expressed preferences of many postpartum women (Hyman 1999a).
Conclusion
It has been said that a good drama requires victims, villains, and heroes. If so, the managed care backlash has been a most excellent drama, with consumers and providers playing the victims, MCOs playing the villains, and legislators and regulators playing the heroes. The script has not varied, regardless of the specific issue. All was well in Healthland until managed care appeared on the scene and destroyed civilized health care. A legislative posse is summoned, and incited with a few horror stories from carefully selected victims. The posse tracks down the MCOs and provides a few general directives about the boundaries of acceptable behavior (be nice, no cattle rustling, and absolutely no tying of victims to the railroad tracks). At first, the MCOs argue that they have not engaged in such conduct or, if they did, it was just an accident. The posse threatens to string up a few MCOs, and eventually the MCOs agree they will sin no more. A member of the posse is left behind to supervise, but now that the MCOs know the boundaries of appropriate conduct, civilized health care is quickly restored. Everyone in Healthland lives happily ever after.
If these three high-profile examples of consumer protection initiatives are any indication, this script is overly optimistic. This is not to say that managed care is perfect; it is not, by any stretch of the imagination. But, however bad the status quo, things could easily be worse. Focused initiatives are rarely what they seem, and more open-ended aspirational language ("reasonable" access to specialists, a "fair and efficient process" for appeals, "culturally competent" care) has its own problems. It is important to remember that the choice is not between imperfect markets and perfect regulations, but imperfect markets and imperfect [End Page 1068] regulations--and the case for regulation cannot be made simply by pointing to deficiencies in the market.
In short, consumer protection against managed care is harder than it looks. Even when a real problem is identified, the issue is invariably more complex than it first appears, and the proposed reforms suffer from their own shortcomings--even without factoring in the (usually carefully ignored) economic implications. Something more than "mom-and-apple-pie" rhetoric is necessary if consumer protection legislation is to have a beneficial impact. The temptation to micromanage in response to anecdotal horror stories is understandable, but legislative caution is likely to be the better part of regulatory valor. It was not for nothing that Alfred Kahn (1998:27), former head of the Civil Aeronautics Board (CAB), recently observed, "Those whom the gods would destroy, they first make regulators."
University of Maryland School of Law
David A. Hyman is a professor at the University of Maryland School of Law. He has a B.A. (1983), a J.D. (1989), and an M.D. (1991) from the University of Chicago. Hyman was a fellow in the Pew Program in Medicine, Arts, and the Social Sciences at the University of Chicago. He has published articles on procedure, narrative, consumer protection and managed care, patient dumping, the nature of professionalism and medical ethics, tax exemption for hospitals, and the legal issues raised by patient referrals. His research interests are principally in the regulatory and financing areas of health policy.
References
Braveman, P., W. Kessel, S. Egerter, and J. Richmond. 1997. Early Discharge and Evidence-Based Practice: Good Science and Good Judgment. Journal of the American Medical Association 278(4):334.
Havighurst, C. C., J. F. Blumstein, and R. Bovbjerg. 1976. Strategies in Underwriting the Costs of Catastrophic Disease. Law and Contemporary Problems 40(4):122-195.
Hyman, D. 1998a. Consumer Protection and Managed Care: With Friends Like These . . . Health Law Handbook, ed. Alice Grossfield. St. Paul, MN: West Group.
-----. 1998b. Consumer Protection in a Managed Care World: Should Consumers Call 911? Villanova Law Review 43(2):409-466.
------. 1999a. Drive-Through Deliveries: Is Consumer Protection Just What the Doctor Ordered? North Carolina Law Review 78(1).
------. 1999b. Accountable Managed Care: Should We Be Careful What We Wish For? University of Michigan Journal of Law Reform 32(4).
Kahn, A. E. 1998. Resisting the Temptation to Micromanage: Lessons from Airlines and Trucking. In Regulators' Revenge: The Future of Telecommunications Deregulation, ed., Tom Bell and Solveig Singleton. Washington, DC: Cato Institute.
Pear, R. 1995. Doctors Say HMOs Limit What They Can Tell Patients. New York Times, 21 December, A1.
President's Advisory Commission on Consumer Protection and Quality in the Health Care Industry. 1998. Quality First: Better Health Care for All Americans.
U.S. General Accounting Office (GAO). 1996. Maternity Care: Appropriate Follow-Up Services Critical with Short Hospital Stays. Pub. no. GAO/HEHS-96-207. Washington, DC: U.S. GAO.
------. 1997. Managed Care: Explicit Gag Clauses Not Found in HMO Contracts, But Physician Concerns Remain. Pub. no. GAO/HEHS 97-175. Washington, DC: U.S. GAO.
Woolhandler, S., and D. U. Himmelstein. 1995. Extreme Risk: The New Corporate Proposition for Physicians. New England Journal of Medicine 333(25):1706-1708.
