Nine Lessons for National Health Reform

Nine Lessons for National Health Reform from the Failure of the 1988 Massachusetts Universal Health Insurance Law INTRODUCTION The 1988 Massachusetts universal health care law called for all residents of the Commonwealth to be offered health insurance coverage by 1 April 1992. This was to be accomplished through a combination of a mandate on employers with more than five full-time workers to either provide insurance or pay a new tax, and a residual state program for unemployed workers and those employed in firms with five or fewer full-time workers. Early in 1991, the Massachusetts legislature voted to delay implementation of the employer mandate for three years. The legislature has never acted to provide the funds required to finance the state's obligation. It is widely believed, as noted shortly, that the combination of the deep recession, the state's fiscal crisis, and the election of a new fiscally conservative Republican governor hostile to the 1988 law together explain its non-implementation. We conclude otherwise. The law failed because it did not even attempt to finance its coverage improvements with funds liberated by establishing effective cost controls. Its universal access problems could not have been kept because their design was unaffordable-- to business, government, and citizens alike-- in any conceivable economic and political circumstances. Massachusetts-- even more obviously than the nation as a whole-- already spends enough to care for every resident. Yet the 1988 law's design made existing high spending levels a barrier to universal coverage, rather than an opportunity. Although the economic, state budget, and political explanations are helpful, we contend that the specific provisions designed into the 1988 law are by far the most powerful explanations for the law's failure to-date. We draw nine specific lessons for national health reform. LESSONS 1. The 1988 Massachusetts universal health care law failed in part because it provided only promises of cost control, not guarantees. Managed competition appears to be another promise, and the backup regulations in the Clinton plan are flimsy and would be hard to enforce. Many European nations, by contrast, guarantee health care cost control by setting advance limits on how much can be spent on care. The promises of cost control in the Massachusetts law rested on hospital closings and bed reductions, inter-hospital competition, and managed care. The law changed hospital payment formulas to reward hospitals for successfully competing to attract more patients while punishing financially hospitals with low 2 occupancy and those losing patients, in apparent hope of forcing some institutions to close. It also provided new regulations obliging hospitals to retire a number of unoccupied beds. Unfortunately, both provisions rested on the incorrect assumption, often voiced loudly, that excess beds explained a significant share of the state's extraordinarily high hospital and other health costs. This is not true. The state's bed-to-population ratio has remained close to the national average in recent years, and has actually fallen below since 1988. Yet our hospital costs per capita remain 35 to 40 percent above the national average. Indeed, a better explanation for these new financing provisions is that they were sought by hospitals gaining patients, to provide them with higher revenues. In any case, the Massachusetts law's cost controls were speculative at best. They have been counter-productive in several respects. To the considerable extent that they have accelerated the closing of many smaller hospitals, they have probably tended to harm access to care. To the considerable extent that they have resulted in the closing of several relatively efficient hospitals, they have tended to increase cost. To the considerable extent that they provided financial incentives for more inpatient and more outpatient care overall, they are further associated with cost increases.1 2. Garnering narrow majorities for the universal health care law in the Massachusetts legislature helped to produce a bill that could not be implemented in any conceivable economic or political climate Certainly, it is useless to insist that legislation be perfect, but it is senseless to fight for the only law that could pass if it is a law that cannot be implemented. To do so will only further erode faith in government. This activity resembles that of the person looking for his keys under a street lamp (even though he dropped them around the corner and dozens of yards away) because "the light is better here." Securing passage for the 1988 Massachusetts universal health care law required abandoning any effective controls on hospital or other health costs, giving hospitals a great deal of new money during the law's first four years, deregulating hospital volume (as just noted), capping private sector contributions to the hospital free care pool, having the main employer mandate take effect only after four years, and forgiving firms employing five or fewer full-time employees any obligation to help finance the law. The law provided new revenues for hospitals, sums which kept growing larger to grease passage through the legislature. These would ultimately total $3 billion in higher payments during the law's first four years, for business-as-usual care of already insured patients. 3 The main political jobs of negotiating the shape of -the new law were to satisfy hospitals and businesses already providing health insurance. The core provisions benefiting both took effect soon after the law was implemented; both then withdrew effective political support for the remaining provisions, particularly the employer mandate to pay or play. The Clinton administration seems to have concluded that its first task is to secure enough votes to pass health care reform legislation. It may be repeating the .Massachusetts error of focusing on politics rather than on building a bill that can work after enactment. It has concluded that single payor financing is politically infeasible owing to public reluctance to convert $400 to $600 billion in existing insurance and out-of-pocket spending to taxes. The Clinton administration has concluded, therefore, that the only alternative is an employer insurance mandate to raise money, combined with managed competition to contain cost. Ignored entirely is the breadth of evidence from most of the Western European and other industrial democracies. There, solid cost controls typically do not require single payors, and HMOs, competition, and regulatory micromanagement are uniformly absent. Present are universal coverage, spending at half of U.S. levels (on average), and generally superior health outcomes and citizen satisfaction. 3. Health care for all requires real cost control, not new money, along with methods of redistributing existing spending It is vital to link realistic cost controls directly to new programs to improve coverage in part because the obligation to find money to help people is the best motive to save money. The Massachusetts universal health care law's cost controls were at best uncertain and at worst counter-productive. As well, the law saw no necessity to employ a mechanism to capture any savings that might be won and recycle them to improve financial protections for patients lacking coverage. For these two reasons, the Massachusetts law required large and visible increases in total health care spending in order to protect previously uninsured residents. The Clintons' health reform proposal's pooling of most funds in the financial reservoirs of the health alliances certainly provides a useful foundation for redistribution. Even though the nation, like Massachusetts, is clearly spending enough, it is doubtful that managed care, competition, and premium caps will squeeze out enough clinical and administrative waste to win savings for redistribution. 4 4. It is vital to provide everyone-- or at least a very large number of people-with valuable and tangible benefits very soon after the reform law passes The 1988 Massachusetts universal health care law was scheduled to provide very few benefits until it had been in effect for about four years. Only then would the main universal health insurance mandate take effect and provide coverage to previously unprotected workers and their families. There was so symmetry and synchronicity of pain and gain; this parallels and in part reflects the failure to coordinate effective and guaranteed cost controls with coverage improvements. Until 1992, only relatively small groups of people were scheduled to be helped-disabled working adults, disabled children, and families leaving welfare for their first two years of work, beginning in July of 1988; workers collecting unemployment insurance, beginning in April of 1990; and workers enrolled in "phase-in" demonstration programs. The promises and programs were poorly publicized, and enrollments built very slowly. Few people identified themselves as potentially eligible for the unemployed and phase-in programs and so could not support them. The largest group, those receiving unemployment insurance, knew that their benefits were temporary only. Because so few people benefited from the 1988 law soon after implementation, it engaged the attention of only a minute fraction of Massachusetts residents in the months and years after passage. The law never created a substantial constituency of people with something to lose from its repeal. Each of the discrete programs' constituents have had to fight alone against cutbacks. Because the uninsured people who were to benefit from the universal health care law's main play or pay coverage provision are such a diverse and difficult group to organize, few people were motivated to defend that provision from repeal or delay once small businesses, ideological conservatives, and other groups sharpened their knives. The Clintons' proposal promises few early benefits. Medicare's new outpatient drug benefit would begin in July of 1996. All states would have to participate in the proposal's core programs by 1 October 1997 (though states could begin participating 1 January 1995 and up to ten could be involved in 1996).2 Yet the new national cigarette tax would take effect in 1994 and the substantial Medicare and Medicaid cuts would be felt in 1996. To win support for passage and implementation of a new national health reform law, it might be useful to include something like a new outpatient prescription drug benefit that would take effect within 90 days of passage. It might cost as little as $25 billion after modest and income-scaled deductibles and co-payments.3 The net cost of this new benefit would be much less, since it would preclude covering outpatient prescription drugs through job-based insurance. The new benefit could be financed in part through a modest new tax, one that bought concrete and visible gains. Most people do expect to have to pay-- and 5 declare themselves willing to pay-- higher taxes for improved coverage for all Americans. In part, it would be financed by federal actions designed to hold down the real burden of prescription drug costs for all Americans. These actions could include establishing the federal government as the sole buyer of prescription drugs, federal ceilings on drug prices, and limiting insurance coverage to the lowest-cost appropriate medication through reference pricing. These are all steps that other nations have employed successfully. Again, guarantees, not promises. 5. It is vital to avoid regressive financing, especially that originating in punitive and moralistic outlooks or misdiagnosed problems One fairly large component of the 1988 universal health insurance was the requirement that all full-time (more than three-quarters time) undergraduate and graduate students purchase health insurance for themselves and their dependents if they were not already covered. This originated in two beliefs, that this was a healthy population and so would not be costly to cover, and that many uninsured students were from out-of-state and were taking advantage of the state's free care pool for costly hospital care. The first belief was accurate; the second was not. Out-of-state students disproportionately attend private colleges and universities, many of which already required proof of health insurance coverage or purchase of protection. Massachusetts residents disproportionately attend public colleges and universities. Although from families with lower average incomes than private college students, the public college students were therefore particularly burdened by the new requirement, which could be seen as a new regressive tax on people pursuing higher education. With this new tax, ironically, the group perhaps least in need of insurance has been the only one required to buy it (rather than simply having it made available) under the universal health care law. Unhappily, it appears that the Clintons' proposals for premium payments, deductibles, and co-payments (for physician services and for outpatient prescription drugs) will cost many low-income people substantially more than they can afford. This regressive financing proposal seems to arise from the mistaken belief that our health care costs are so high in large part because Americans seek too much health care. (The far bigger problem is that caregivers provide too much, especially to people with good financial coverage.) The proposal also pursues the moralistic goal of making patients more cost-sensitive. Tragically, instead of relieving many poor people about worries over whether to pay for health care or to buy food, this will require them to forgo the food. 6 6. If the states are to function as laboratories of democracy under any national health care reform plan, it will be vital to secure accurate information on what is attempted, how well it works, and why. This has certainly not been the case in Massachusetts for the past five years. The state has not been able or willing to finance objective analysis of what was attempted. For example, the state has not collected and reported insurance claims data from the "phase-in" demonstrations of small business health insurance, even though fear of a surge in claims was ostensibly at the root of assertions that we must phase in coverage and study its effects before risking universal insurance. Similar shortcomings have plagued the programs for disabled workers and children, for unemployed workers, and for college students. As legislative and gubernatorial support for mandated universal insurance weakened over time and elections, so did national interest and the likelihood of outside support for analysis of Massachusetts health policy. As a result, much of what has been reported about Massachusetts reflects widely and comfortably held beliefs that have not been carefully examined, assertions that work to protect the interests and reputations of the 1988 law's authors and proponents, and often grossly inaccurate characterizations of the law by its critics. Some examples: In December of 1990, a report in the Boston Globe asserted that "Although Massachusetts' universal health care law has been widely written off as a doomed remnant of the Dukakis era, there are growing signs the landmark experiment will be retooled, but not dismantled, by the incoming Weld administration."4 This report discussed at length how practical and affordable it would be to implement the 1988 statute. It included several optimistic and unchallenged assertions from Dukakis administration officials in the Department of Medical Security. In April of 1991, a lengthy New York Times analysis blamed "a soured economy, a hostile new Governor and the fierce opposition of small-business owners" for proposals to repeal the main employer mandate. A supporter of the law blamed not its design but the recession and politics. But, as discussed above, it is entirely likely that the law's design was largely responsible for its problems. It contained large revenue increases for hospitals and feeble cost controls. Its failure to reallocate any existing spending to fund universal coverage meant that coverage improvements would have substantially increased the cost of health care in Massachusetts, already the highest in the world. Former-governor Dukakis claimed late in 1991 that "Massachusetts is about half-way toward implementation [of universal coverage], but we have a new Administration that is seeking to delay or kill the employer mandate."6 By no substantive measure could Massachusetts have been deemed close to half-way to implementation. Further, it seems unfair to blame delay or repeal of the all-important employer In mandate entirely on the new Weld administration given the Democratic legislature's readiness to cooperate (both houses actually passed delay or repeal measures even while Dukakis was still governor) and given the law's effective abandonment by its former hospital and business supporters. 7 A 1992 paper analyzing rights to health care asserted that the main Massachusetts play or pay mandate "was insolvent by 1989."7 It is not clear how something could be considered insolvent three years before it was originally scheduled to exist. 7. A number of other lessons from the attempt to promise universal health insurance in Massachusetts arise from the interaction between the 1988 law and the state's health care delivery and finance. It seems clear from the Massachusetts experience to-date that health maintenance organizations, competing or otherwise, have not yet succeeded in containing health costs. Massachusetts has now assumed first place in the share of its population enrolled in health maintenance organizations. It seems just as clear that competition among hospitals and other caregivers cannot be trusted to yield up an acceptable, accessible, and affordable configuration of hospitals, doctors, and other caregivers. Over one-third of acute care hospitals in the state have closed since 1970, with no discernible reductions in cost. 8. The Massachusetts experience suggests that merely manipulating financial incentives is not an effective cost control technique. Beginning in the 1988 universal health care law, Massachusetts provided financial incentives to hospitals to compete for patients. The result was, not surprisingly, a rise in the volume of care provided.8 These incentives were strengthened by a 1991 law providing for virtually complete hospital price deregulation.9 We suggest that financial incentives usually and inevitably overshoot the mark-both the financial incentives to overserve embodied in fee-for-service and cost reimbursement, and the financial incentives to underserve embodied in reliance on competing HMOs to contain cost. Both kinds of financial incentives require promulgation of offsetting or modulating regulations, but these can probably never be strong enough to stem the huge financial pressures they are designed to counter. This illustrates the way in which more competition breeds more regulation. Competing HMOs will be financially motivated to try to: a) game any nascent risk-adjusting technology to cream profitable patients; b) avoid patients they think will be more costly or difficult to serve, such as low-income citizens or people vulnerable to deprivation of care; and c) systematically underserve all patients when they can. It will be difficult to contend with these forces. Far better, we believe, to design payment mechanisms that are financially neutral because they allot finite sums to serve defined populations, and far better to pay this money to organizations that can be trusted to spend it carefully.10 8 9. It is not feasible to fill in the remaining gaps in insurance coverage by designing special, small targeted insurance programs. Some have argued that since private insurance has apparently succeeded in covering the great share of the population, the remaining task of government is to design, subsidize, and/or encourage insurance coverage for the remaining people. This may seem logical, but it downplays the very possibility that there are reasons why private health insurance has failed to cover some people. a. The Massachusetts experience with patchwork fill-in programs reinforces this concern. It has proven complicated, administratively costly, and often ineffective to rely on traditional insurance mechanisms to cover special populations at special times. The Health Security Plan, for example, is intended to provide or subsidize temporary health insurance coverage for workers receiving unemployment insurance cash benefits. Because workers and families are eligible for short times only, the program faces many of the challenges of the Star Wars programs, with many fast-moving targets to hit. This means not only that administration is very costly, but that many who could benefit do not learn of or enroll for their benefits while still eligible. The need to enroll people individually, rather than through automatic coverage, is one reason for low enrollment. Relying on a patchwork of plans leaves more seams and lots of small gaps, each requiring a new patch, a new program, and new coordination. The result is greater administrative cost and complexity, and greater barriers to negotiating the system by people in need. The Clintons' plan seems to reflect this lesson by avoiding the patchwork, fill-in approach. b. Special new voluntary "phase-in" demonstration programs for small businesses in Massachusetts proved unsuccessful. They were unattractive to insurors and employers alike, even with state subsidies. Nor has a more recent state initiative that permits stripped-down benefits for small business plans. Enrollment was excruciatingly slow, except in the one program that allowed workers to sign up directly, without employer involvement. Just 1,000 were enrolled in 10,000 funded slots; 40,000 had been expected originally. As has been the experience in other states, this points to the futility of voluntary programs to achieve universal coverage. The Clintons' plan reflects this lesson by proposing one regional health alliance as the enrollment agency and main financial reservoir. c. The Massachusetts law never attempted to eliminate the high administrative costs associated with insurance coverage and processing of individual claims. Indeed, insurors' interests in gaining business were catered to, in that most 9 citizens (except severely disabled people and welfare recipients) were to gain their new coverage through private insurors even when the financing was public. The danger in relying on private insurors is that high-risk people will be left uncovered or that government will have to pay excessively to get insurors to take that business. The limited interest and few bids from insurors to write policies for the Massachusetts phase-in demonstrations may have helped raise their cost and delayed their implementation. With the Clintons' plan, there is reason to fear similar obstacles to finding HMOs willing and able to serve low-income and other underserved communities. SUGGESTIONS FOR STATE HEALTH REFORM We should start by recognizing that there is no substitute for intelligent state action. Don't count on Washington to do it all. A national law aiming to cover everyone and contain costs is possible within a year or two. It might cover everyone, but it will almost certainly prove ineffective in containing costs until crisis creates an effective national political constituency for effective cost control. Therefore, the health care special interests-- hospitals and doctors, insurors, and drug and equipment manufacturers-- will fight for and win more money for business as usual. And if, at some point, Congress were to legislate effective national cost controls, it would try to shave the peaks of high health costs and fill in the valleys. This would, for example, decapitate health care in Massachusetts, where nearly two times as much per resident is spent as in Idaho, the lowest-cost state. Either way, high-cost states need to tailor home-grown legislative solutions. Face reality. Health care costs have got to be controlled, so they grow no faster than the economy, but nothing that has been done recently or is now contemplated seriously has a chance of working. Look at how the rest of the world has addressed its health problems. Ignore the surface issues, like Canada versus Germany versus Britain; like single-payor versus multiple payor. Recognize that no nation has contained health costs without covering everyone and that no nation has covered everyone without containing costs. And also recognize that both tasks require building trust among all parties-- payors, patients, and caregivers. Any state-level reform plan should probably address six key concerns: I. A ceiling on total health spending, so health keeps its fair share of the state economy, but no more. 2. Financial protection for everyone, a safe and solid foundation under each person, without worry of losing coverage when we change or lose jobs. 3. Methods of raising money and paying for services that separate the money from decisions about care, both for patients and for caregivers. This requires paying hospitals and doctors in ways that allow patients and payors to trust caregivers. It means avoiding financial incentives to overor under-serve. And it 10 means removing financial barriers to seeking needed care, in part by raising money fairly. 4. Professional re-orientation by hospitals and doctors toward patients and payors. Caregivers will need to `accept responsibilities to marshal inevitably limited resources to take care of everyone. 5. Freedom of choice of caregivers, including a well-paid family doctor for everyone who wants one, and freedom from worry that an employer's new negotiations with HMOs will force a change in family doctor. 6. The coverage and delivery systems must be organized in ways that hold them accountable for reaching and appropriately serving everyone. When multiple HMOs compete to serve fractions of a broad geographic area, there is no way to hold them accountable for under-care or over-care-either individually or collectively. It would be far better to see systems of service organized around geographically visible caregivers, with competition by quality and compassion across the borders of service areas (as most service areas overlap substantially), and with money following patient choice. 11 This package provides the foundations for the things we want from health care: freedom from financial worry, confidence that our doctors and hospitals will give us the care that works for us, and assurance that we will be able to reach a doctor who knows us any time of the day. It would be important for state governments to bring together all interested parties, set goals and timetables for reaching them, and broker a new health care peace treaty. States like Minnesota, Vermont, Florida, and Washington have legislated or are considering a range of health care innovations. States like Massachusetts cannot afford to lag. 11