On Government’s Role in the Crossing of Chasms
One does not generally view the Institute of Medicine (IOM) as a source of amusement. But there is something comical (albeit tragicomic) about the series of IOM reports produced since 1999 that address quality shortfalls in American medicine. It’s not their titles, which are more evocative than amusing: To Err Is Human, Leadership by Example, and Crossing the Quality Chasm. And there is certainly nothing funny about the shortcomings they document: “tens of thousands of Americans die each year from errors in their care, and hundreds of thousands suffer or barely escape from nonfatal injuries that a truly high-quality care system would largely prevent” (IOM 2000:2).
The (sadly) amusing aspects of these reports involve the sorts of remedial policies that they discuss. Or, more precisely, those that they don’t discuss. It is as if the committees involved in these reports were given a common charge: See how long you can write about dramatic and persistent failings of performance in the health care system without ever mentioning government regulations that might remedy those failings. It’s like some perverse game of policy charades: the reports can act out the need for regulatory intervention, as long as they never actually mention regulation directly.
The IOM committees prove to be quite astute at this game. Although they call for government action, they would limit its role to acting as a purchaser (i.e., “Leadership by example”), collecting information, or disseminating reports on system performance. They do go so far as to recommend that government actually mandate that health care providers report performance measures, but only after carefully weighing the potential for voluntary reporting. All told, out of thirty recommendations in the three reports, only one even mentions regulation. And this recommendation is a carefully worded call for additional study about the ways in which regulations governing professional practice might either “facilitate or inhibit” improvements in quality (IOM 2001: 20).
If it is amusing to see how hard the IOM committees worked to not discuss regulation, it is far sadder to consider the consequences of this omission. Although the reports document “the absence of any real progress toward restructuring health care systems to address both quality and cost concerns” (ibid.: 3) and warn that “the current set of activities has not closed the quality gap and is unlikely to do so in the future” (IOM 2002: 4), this purblindness regarding regulatory interventions means that none of the reports even considers changing the basic legal rules under which the health care system operates. Because the IOM was unwilling to endorse any exercise of government authority, the potential for fundamental change is lost. A series of modest incremental recommendations take its place, yet these incremental reforms lack the scope to effectively address the pervasive factors that currently undermine quality in American medicine. To recall the advice of Winston Churchill: one does not cross a chasm in two strides.
Barriers to More Effective Government Regulation in Medical Care
The Institute of Medicine is not alone in its unwillingness to seriously consider expanding government regulation. Particularly within the Beltway, policy makers have grown quite averse to the notion of regulating medical care. This is illustrated by the failure of Congress to pass any form of a patient bill of rights, even though comparable legislation has been enacted in a number of states.
Obviously, this antiregulatory sentiment reflects the current dominance of conservative ideology in federal policy making. But there is more to it than ideological or partisan framing among policy elites. Some observers attribute this turn against regulation to trends in prevailing public opinion. For example, one assessment of changing public opinion toward medical care over the past fifty years concluded that health care reform had become more difficult because “Americans are clearly less willing to see expanded government regulation in general than they were during the 1960s” (Blendon and Benson 2001: 43). But this conclusion ignores evidence that the public is far more supportive of government regulation in health care than in other aspects of society. Blendon and Benson (ibid.: 42) report that in the year 2000, roughly 60 percent of the public thought there was “too much” government regulation of business generally. Other surveys conducted during this same time suggest that only 28 percent of the public felt that there was too much regulation of health care, compared to 45 percent who favored more extensive regulation.1 Indeed, the public is more supportive of government regulation of medical care today than it was of government regulation generally in the 1960s. Given this evidence, public attitudes can hardly account for the resistance of federal policy makers to even consider expanding government authority over medical care.
Although not all were written with this intent, each of the articles included in this issue of JHPPL provides some additional insights about the prospects for more extensive or effective regulation of medical practice. The first two of these articles share a common substantive focus: the quality of care in American nursing homes. The third article examines how physician groups have responded to increasing competition among health plans. The final article explores the data that are available to monitor variation over time and across states in the number of Americans without insurance. Although only the first two articles are explicitly about quality of care, all four speak to the scope and limits of government involvement in medical care and thus the potential for government intervention to improve the overall performance of the health care system.
Regulating Quality in Long-Term Care
When Bruce Vladeck (who would later head the Health Care Financing Administration) wrote about nursing homes almost twenty-five years ago, he (1980: 147) noted that “direct regulation of the quality of nursing homes and the services they deliver has been an important government activity for at least a generation.” Although these regulations were subsequently challenged by Republican officials intent on deregulation, Congress retained and even strengthened this extensive regulatory role over the subsequent two decades. This was in no small part because the public saw these regulations as essential and rallied in opposition to proposals reducing government oversight. In a survey conducted in early 2001, for example, only 5 percent of respondents reported that government regulation over the quality of nursing home care was too extensive, whereas 61 percent favored expanding regulation (Kaiser Family Foundation 2001).
But while public support for regulation of long-term care remains strong, two articles in this issue raise questions about how effectively these regulations have been enforced. The essay by Charlene Harrington and her colleagues documents that, despite this long history of regulation, serious violations continue to be found on a regular basis. In a given year, more than 30 percent of all facilities were identified as having quality problems that threatened the health or safety of a resident. Five percent of all facilities were deemed to be providing seriously substandard care. Despite an impressive array of regulatory sanctions, officials in many states expressed frustration at their inability to effectively address problematic facilities, in part due to the lack of adequate resources for regulatory enforcement. The researchers also found considerable variation in the intensity of enforcement efforts among the states, attributable in part to political considerations and in part to the structure of the nursing home industry.
The article by Jennifer Troyer and Herbert Thompson addresses a different aspect of the linkage between regulations and quality. They note that regulations tend to emphasize certain dimensions of quality, while neglecting others that are more difficult to measure in a reliable fashion. This leads to the potential neglect of factors related to “quality of life.” Troyer and Thompson suggest that these less readily measured dimensions of quality will consequently become the focus of consumer voice—efforts by residents or their families to express their grievances when they feel that some aspects of quality have been neglected. Although these grievances can be voiced through a variety of channels, including nursing home ombudsmen employed by each state, this article focuses on that time-honored American response to perceived mistreatment: sue the bastards.
The empirical analyses presented in this article suggest that, at least under some circumstances, there is a positive relationship between the probability of legal action against a particular nursing home and the number of times that it had violated nursing home regulations. But the direction of causality remains unclear. Does a lawsuit force the home to respond to quality of life issues, thus diverting resources that otherwise would be used to address the dimensions of quality measured by regulators, or does the failure of state regulators to effectively enforce quality standards allow facilities to operate with persisting quality deficiencies, which affect even those dimensions of quality that are not directly covered by the regulations?
For both these articles, the empirical analyses must be viewed as exploratory. In one case, they rely on data from a single year; in the other case, from a single state. But both suggest some provocative hypotheses about factors that may constrain more effective regulation of quality. The limitations on quality oversight that each article documents should give some pause to advocates of regulation. After all, the nursing home context would seem, for several reasons, to be a setting within which government regulation should be most effective. First, because states have been regulating quality of nursing homes for some time, they’ve had plenty of opportunity to learn the ropes and improve the ways in which they identify and address quality shortfalls. Second, public support for regulations remains high, giving these interventions considerable legitimacy. Third, health care professionals play a limited role in most nursing homes; there is therefore little countervailing authority to government action.
Given all these seemingly favorable circumstances, it is troubling that quality of care in many nursing homes remains so low. But there is at least one other factor at work in this setting, suggested by one of the books reviewed in this issue. Bo Burt’s treatise on public policy and death, Death Is That Man Taking Names, makes clear how deeply ambivalent Americans are about decisions in which death looms large. As Vladeck observed a quarter century ago, nursing homes are often seen as places that people go to die. Our collective reluctance to confront death, to deal with its moral quandaries, may well have spilled over to undercut support for effective regulation of these facilities, particularly when enforcement regulation requires expenditures of political and financial capital that might be seen as wasted on the facilities that Vladeck termed “warehouses for the dying.”
Burt’s analysis, however, also suggests yet another consideration that may constrain effective government intervention into medical care. He argues that government interventions (whether legal or regulatory in nature) are often intended to give primacy to a particular conception of ethical or high-quality care. As such, he sees them as suppressing legitimate disagreement about the goals of care and thus exacerbating tensions related to the treatment of individual patients. Although his argument is cast in terms of decisions involving the end of life, it could be readily applied to the broader set of considerations that Troyer and Thompson label “quality of life.”
Regulations of nursing homes have clearly affected these quality of life considerations. For example, when the life-safety codes were implemented, intended to ensure that bed-bound patients were not trapped in facilities from which they could not be removed in an emergency, a number of more home-like nursing facilities were forced to close. There was no opportunity for people to argue for the benefits of the homier settings, even at some additional risk from fire or other threats.
But regulations need not foreclose debate over the appropriate ways to define quality, given each individual’s preferences. As Burt notes in his analysis, some regulations, such as those associated with the Patient Self-Determination Act, seem to have actually fostered conversations among family members, and between nursing home residents, family, and staff, about the preferred forms of end-of-life care. Taking this approach, government regulation may be quite consistent with aspirations to greater empowerment for individual patients and their families.
Regulation and Market-Oriented Health Care Reforms
Stephen Page explores a different context in which regulations might enhance the performance of American medicine. He suggests that the market-oriented reforms typically labeled “managed competition” were ill-conceived because they failed to anticipate the most likely responses of physicians to the changing practices of managed care plans. More specifically, it has proven easier for physician groups to respond to the spread of managed care with strategies that neither enhanced efficiency nor improved quality of care. While physicians clearly adapted to changing market conditions, the strategies that were most viable for preserving professional autonomy and income were not those that best met other socially valued goals.
Page’s proposed solution: have policy makers change the rules of the game under which physicians organize so that they have stronger incentives to develop the infrastructure within their medical groups to effectively oversee the practices of affiliated physicians. Some of these changes would entail more complete public reporting of characteristics of practice groups. Others would potentially involve regulations encouraging or requiring managed care plans to contract with physician groups that have particular organizational capacities.
Political scientists have long recognized that for market-based reforms to be effective, there would need to be an appropriately supportive role for government and thus the political circumstances and resources that make this role feasible (Morone 1993). But there has been little empirical research on the factors that support or impede state regulation of the conditions of professional practice.2 To his credit, Page acknowledges that it may prove difficult to formulate policies that are both acceptable to organized medicine and also promote pro-social responses among physician groups. But, however feasible, it seems likely that the political dynamics of this form of regulation will prove quite different from those shaping the regulation of nursing homes. The exact form of those differences, and their implications for policy enactment and implementation, remain in need of further study.
Assessing System Performance As a Prerequisite to Effective Regulation
Unwilling to push for government regulation, the IOM makes a strong case that government ought to collect information nationwide that would regularly assess the quality of care received by Americans, reporting “annually to Congress and the President on the quality of care provided to the American people” (IOM 2000: 7) As Harrington and colleagues discuss in the case of nursing homes, this sort of quality monitoring is also an essential prerequisite for consistent regulation.
The final essay in this issue, by Lynn Blewett and colleagues, at first seems unrelated to these considerations. It offers a summary and assessment of the various surveys that could be used for measuring the number of people who do not have health insurance in any given state. Because the number of uninsured Americans remains a pressing concern for policy makers, and a target for new policy initiatives at state and federal levels, this review of data sources is useful in its own right. But it can and should be also viewed from a larger context—as a case study of the challenges of providing reliable estimates of any characteristics in a population, particularly characteristics that are known to vary among subgroups of citizens.
In comparison to measuring quality of care, counting the uninsured seems an almost trivial task. Certainly it is complicated by the dynamics of insurance acquisition and loss. And it becomes more challenging when one needs to compare rates of insurance coverage among less prevalent groups of ethnic or racial minorities. But even with these challenges, the task of identifying people who are not covered by health insurance is at least an order of magnitude (perhaps several) easier than assessing, on a regular and consistent basis, the quality of care experienced by a representative group of Americans.
Yet Blewett and colleagues document that, despite these relatively simple aspirations for data collection, a number of states remain incapable of providing annual estimates of the number of their citizens who lack health insurance, let alone more detailed measures of the variation in insurance rates among different subgroups. Equally important, the surveys that have been fielded often collect information in ways that are inconsistent from one state to the next, making it impossible to reliably compare findings from the state-specific surveys. Even where appropriate data are collected, it has been sometimes difficult to make this information available to key public officials on a timely basis.
If this remains true even for the relatively simple task of counting the uninsured, a task that has been a clear priority for state governments for at least the past two decades, how optimistic can one be about the potential for collecting comparable data assessing quality of care? For this latter task, one would need to collect sufficient data not simply to compare across demographic groups, but to compare across health conditions. And because research suggests that quality varies across groups and conditions, there would need to be a sufficiently large and representative sample such that both these sources of variation could be assessed. In light of the experiences summarized by Blewett and colleagues, it is hard to be optimistic about the prospects for a data collection enterprise of this scope, absent quite substantial investments of additional resources by the federal government.
of Regulation Involving Medical Care
In light of the issues identified in these articles, it would be easy to be pessimistic about the potential for more effective regulation of medical care. So for purposes of balance, let me conclude by identifying just a couple of countervailing considerations. The first is derived from the experience of nursing home regulation. Despite our discomfort with death and our propensity to ignore those at the end of life, and despite the concerted efforts by several different presidential administrations to deregulate this industry, the regulation of nursing homes remains highly legitimate for the public and an ongoing concern even in the most conservative of states. This is in many ways a remarkable testament to the perceived need for regulation as a safeguard for the average citizen. These protections could be more effective, and they could be implemented more equitably. But they clearly have a deep-rooted and lasting place in the contemporary American health care system.
A second, more optimistic perspective can be obtained simply by looking outside the Beltway. Although many policy pundits remain skeptical about the potential benefits of managed care regulation, for instance, and congressional initiatives in this area seem locked in political stasis, state policy makers have proven far more responsive to calls for action. Over the past decade, there has been a remarkable spread of regulations governing the relationship between health plans and their affiliated health care professionals. In this sense, regulatory interventions appear alive and well in the American polity. But it should also be recognized that we know virtually nothing about the extent to which these regulations have been enforced and thus know little about their consequences for the delivery of medical care. It is quite possible that these state initiatives have been largely symbolic. Whether they actually alter the care that is received by the average patient is a question that demands the attention of researchers interested in quality matters.
1. Data from the Roper Center for Public Opinion Research. Question ID number USWASHP.00EHLTH.R16C.
2. An exception involves some studies of the diffusion of so-called freedom of choice laws that allowed nonphysicians to bill insurance companies for covered services.