Unintended Consequences of Health Policy Programs and Policies: Workshop Summary (2001)

Wendy B.Young, Rapporteur

The Robert Wood Johnson (RWJ) Health Policy Fellowships Program of the Institute of Medicine (IOM) conducted a workshop on August 22, 2000, on the subject of unintended consequences of health policy programs and policies. The purpose of the workshop was to gain a better understanding of why this phenomenon is a frequent and consequential byproduct of major health policy programs.

The workshop was based on a series of commissioned papers on the subject, written by current and former RWJ fellows. The papers, in the form of case studies, highlighted various examples of unintended consequences from the perspective of the authors’ expertise, special interest, or legislative involvement. The workshop brought together about 50 participants, most of them Washington-based policy experts working in government and leading health care organizations (see Appendixes A and C).

This workshop summary was prepared by an appointed rapporteur, Wendy B.Young, with the assistance of Marion Ein Lewin, director of the RWJ Health Policy Fellowships Program, and Kari McFarlan, deputy director. The summary was reviewed for accuracy in accordance with procedures approved by the National Research Council’s Report Review Committee guidelines.

Since 1973, the IOM, with funding from the Robert Wood Johnson Foundation, has directed the RWJ Health Policy Fellowships Program, which is designed to develop the capacity of outstanding mid-career health care professionals in academic and community-based settings to assume leadership roles in health policy and management. Each year, six fellows are selected to participate in a 1-year program of orientation and full-time work experience in the nation’s capital. As part of their working assignment, fellows usually select work for a member of Congress or a congressional committee office where they play a senior role in developing health legislation and policy. Over the years the work of RWJ fellows on Capitol Hill has been praised for contributing valuable clinical and health sciences expertise to the legislative process (Breaux, 2000).

The workshop dialogue, moderated by Lynn Etheredge, a nationally respected health policy analyst, focused on nine case studies covering a broad range of policy issues (Appendix B contains descriptions of the case studies approved by each speaker.). The rich discussion yielded several observations regarding unintended consequences in general, and consequences of spe-

cific health policy programs in particular. The views expressed are solely the opinions of the authors and other participants.

To set the stage for the discussion, Lynn Etheredge reviewed how valuable a historical perspective can be for revealing the outcomes of policies and for future policy planning. He reminded the group that, without conversations such as this, “the rich source of lessons from past health policy experiences tends to get lost and forgotten. This is unfortunate because so many of today’s priority health issues could be called ‘repeat encounters’. Whichever party is in the White House, whoever is in charge of Congress, and whoever the next round of RWJ fellows, they are inevitably going to face past issues and priorities in a different guise.”

Etheredge cited Richard Neustadt and E.R. May’s book, Thinking in Time, as useful for understanding what brought us to where we are and the real-world problems facing public policy officials as they contemplate difficult choices:

Etheredge challenged the authors and participants not only to create a record of the history of a particular policy development that will be useful to people who revisit the same kind of issue in the future but also to articulate general principles that can sensitize people to the value of discussions of how an understanding of the history of health policy in different areas can contribute to the future.

The case studies were grouped for discussion into three categories: Medicaid, Assessing Risks and Regulating Benefits, and Delivery System Restructuring. Each author briefly described how he or she became interested in the health policy issue and the major points of the case study. After a discussion of each case study, the dialogue shifted to general lessons for health policy that emerged from the nine case studies.

Medicaid’s Disproportionate Share Hospital Payment Program

David Altman’s interest in the Medicaid Disproportionate Share Hospital (DSH) payment program stemmed from his extensive exposure to the plight of highly vulnerable public hospitals in his home state of California.

Created under a Medicaid provision in the 1981 Omnibus Budget Reconciliation Act (OBRA 1981, P.L. 97–35), the DSH program requires state Medicaid payment rates to take into account the uncompensated care burden of hospitals that serve a disproportionate number of low-income patients. Other rulings over the years expanded Medicaid DSH payments to nearly $16 billion in 1997. A 1985 Health Care Financing Administration ruling allowed an additional federal match for donations that states received from health care providers. A 1996 OBRA provision allowed states to pay DSH providers above the Medicare upper-limit rate.

Based on the assumption that hospitals serving low-income Medicaid patients also serve a substantial number of indigent patients not eligible for Medicaid, DSH was intended to cross-subsidize the costs of uninsured care with the higher Medicaid DSH rate. The fragility of this assumption pointed out the risk of unintended consequences from programs that are not well targeted. With the introduction of Medicaid managed care, many hospitals for the first time competed for these publicly insured low-income patients, and public hospitals saw their Medicaid volume and revenues drop. As a result, Medicaid revenues were pulled out of public hospitals providing care for the uninsured, and hospitals that

provided care for Medicaid patients but not the uninsured were over-subsidized.

According to Altman, one major lesson from this case study is that an incremental policy change can affect the entire health care system; a second lesson is the value of listening to the key stakeholders in policy development.

“Whining on the part of interest groups is generally discounted in the legislative hearing and planning process but sometimes the whining is correct,” Altman observed. “In the case of DSH, the National Association of Public Hospitals and similar organizations were saying, ‘Wait, don’t do this. This is going to hurt us’.”

Title XXI: State Children’s Health Insurance Program (SCHIP)

Richard Bucciarelli’s extensive work in Florida on child health policy and the 1989 Florida Healthy Kids program (F.S. 624.91) led to his case study on the 1997 State Children’s Health Insurance Program (SCHIP), or Title XXI of the Balanced Budget Act of 1997 (P.L. 105–33). He reviewed SCHIP’s short and paradoxical history. SCHIP was intended to decrease the number of uninsured children, but the number has actually increased since the start of the program, even though lots of kids are being enrolled. According to Bucciarelli, this unintended consequence is not due to “crowd-out,” the much-anticipated risk that employers would drop health insurance coverage for the families of their employees upon the availability of a new public program. Although many uninsured children have received health insurance, Bucciarelli argues that SCHIP has also stimulated a lot of movement from one insured category to another. Bucciarelli labels this phenomenon “substitution” rather than crowd-out, due to parents electing SCHIP as a substitute for the more costly employee coverage with rising premiums, deductibles, and co-payments.

Many other dynamics have contributed to SCHIP’s unintended outcomes. Occupied with welfare reform and many other changes in the Medicaid program, states were slow to implement SCHIP. In addition, Florida providers were not quick to sign up, with SCHIP payment rates initially set at the Medicaid payment rate of 19 cents on the dollar. Enrollment was slow because many parents believe they do not need health insurance for their children.

One lesson here, as well as from many of the other case studies, is the difficulty of passing legislation without also creating unintended consequences because so many other dynamics accompany the implementation of a new program or policy. Bucciarelli suggests that the key is to recognize unintended consequences and work on legislation to fix those that may have a particularly negative impact.

Washington State Basic Health Plan: Safety Net Providers and the Uninsured

Robert Crittenden’s long interest in health care for the poor and uninsured and his work in the 1970s and 1980s on developing the Washington State Basic Health Plan (WBHP) framed his interest in studying the failure of Medicaid expansions to stop the erosion of health care for the uninsured. This case study tracked the impact of the 1989 WBHP on both the uninsured and safety net providers.

WBHP covers more than 130,000 uninsured persons and families with incomes below twice the poverty level, but the number of uninsured in King County (which includes Seattle) continues to grow and the safety net capacity to care for the uninsured has shrunk. As WBHP introduced new revenues to cover the cost of low-income patient care, safety net providers, responding to market incentives, increased their capacity to serve the growing insured population and shifted their payer mix to more insured than uninsured. As a result, traditional safety net providers are serving only 20 percent of the total uninsured population. Before the advent of Medicaid managed care and WBHP, safety net providers were serving more than a third of the total uninsured population. Crittenden said that the difficulties with incremental and often more politically acceptable policy approaches is a declining ability to consider and address the whole problem.

The three Medicaid policy case studies elicited some immediate discussion which focused on three points. First, the policy making process for these programs did not adequately consider the impact of managed care on public programs, particularly safety net providers. Second, poorly targeted programs of indirect

subsidies carry greater risk of unintended consequences. The assumptions of the DSH program were invalidated when managed care competition was introduced into the Medicaid market and de-linked Medicaid providers from providers for the uninsured. These indirect subsidies set up a new dependency as hospitals used DSH monies to balance lost revenue from managed care cost containment and then fought to protect the current program even though they recognized its failures and the need for change. Third, the single-policy-oriented static perspective in which health policy programs are often constructed fails to take into account other important dynamics at play in the health care system. Better understanding and consideration by policy makers of the broader dynamic forces affecting the health care marketplace may help clarify and limit the potential risks of unintended consequences.

Benefits Mandates by Body Part

Andrew Balas’s interest in benefits mandates grew during his RWJ health policy fellowship congressional assignment in the office of Senator Bill Frist (R-TN). Benefits mandates “define a health care disease- or condition-specific health care requirement with the force of legislation,” and require the provision for coverage of health care services usually not managed or reimbursed by the government. These policies emerged from the societal context of a rising power of consumerism. In response, public interest focused on the quality of health care and the growing patients’ rights movement. At the same time, the world of medical science and clinical medicine focused on efficacy, risk, and cost-effectiveness of treatment options and viewed benefits mandates as an intrusion into the medical profession. Balas uses the policy mandating coverage for a minimum hospital stay for mastectomy as an example. This mandate grew from public demand for protection from cost-driven denials by managed care plans, media headlines of patients outraged over the shortness of hospital stays after mastectomies, and a personal story included by the president in the 1996 State of the Union Address.

The science refuting the efficacy of a 48-hour minimum hospital stay, including a National Institutes of Health consensus statement on breast cancer treatment and recent studies published in respected medical journals, was absent from the political debate. According to Balas, the scientific discussion and the public discussion proceeded on two parallel and independent tracks. A lesson from this policy case study is that science’s traditional methods of communication do not work well in influencing policy when public pressures politicize an issue. Scientists need to better understand the policy making process, learn how to communicate effectively with policy makers, and recognize that the weight of scientific evidence is just one of many factors that enter into the political decision making process.

Compensation to Veterans for Exposure to Agent Orange

Gerald Charles’s interest in the policies and provisions relating to veterans’ compensation for Agent Orange exposure (P.L. 96–151, P.L. 98– 542, P.L. 102–4, and P.L. 104–204) grew from his RWJ health policy fellowship assignment on the Senate Committee on Veterans’ Affairs. He brought to this congressional experience prior knowledge of these herbicides from his earlier career at the U.S. Department of Agriculture’s Forest Service.

Agent Orange, used during the Vietnam War to defoliate and deny cover and food crops to the enemy, grabbed public attention in the late 1970s with headlines of Vietnam veterans claiming a connection between illnesses they had developed and exposure to the herbicide. The controversy erupted overnight against a background of societal concerns about environmental issues following publication of Rachel Carson’s Silent Spring and other works, coupled with significant cynicism and suspicion about American involvement in Vietnam. As Congress struggled to develop a policy response with scant prior scientific studies on the long-term effects of Agent Orange, the issue quickly developed into a clash between the demands on science to conduct technically difficult epidemiological studies and the political pressure to do something in a timely manner.

Veterans’ compensation benefits had traditionally been based on two tenets: either that military service caused an illness or injury or that a veterans illness or injury was temporally related to military service. Because of the sense of urgency to act and because Congress could not get the scientific “cause and effects” relationships about Agent Orange it would have liked, it settled instead for “statistical associations.”

Applying statistical associations to decisions about veterans’ compensation marked a radical departure from the traditional “caused by or temporally related to military service” framework and incrementally altered the compensation system by creating a new precedent. Similar to Balas’s case study on benefits mandates, a lesson from the Agent Orange policy case is that unintended consequences arise from political decision makers’ pressures to make decisions even when there is inadequate scientific evidence. This case also serves as an example of the risks of unintended consequences when decision-makers establish precedents that are not well grounded in science.

Federal Preemption Provision of the Employment Retirement Income Security Act

Karen Guice’s case study stemmed from her RWJ health policy fellowship congressional assignment in the Office of Senator Jim Jeffords (I-VT), at that time chair of the Committee on Health, Education, Labor and Pensions, where she “learned ERISA primarily for self-defense reasons.” The 1974 Employee Retirement Income Security Act (ERISA) (Title 29, Sections 1001–1461) was passed in response to concerns about the management of employee benefit plans, including pensions and health care coverage. Her case study focused on the preemption language added by the conference committee: ERISA preempts states from regulating health insurance plans that are self-funded by employers. Guice’s paper traces the dynamic history of the preemption language, triggering events, policy makers’ assumptions, and unforeseen changes in the marketplace, including the impact of managed care.

Congress passed ERISA to protect employee pension benefits. After 8 years of debate, floor speeches at that time recorded Congress’s expectation that ERISA was one of a series of incremental policies and that a national health insurance program would soon follow. While the bill was in conference to reconcile separate House and Senate versions, three states took action to regulate self-insured plans. The conference committee added the preemption provision to fend off state jurisdiction until Congress could turn to universal health insurance coverage.

After passage, the political climate changed, public attention and the political agenda shifted to other priorities, and national health insurance never came to be. Exempt from state insurance regulations, self-insured businesses developed more innovative employee benefits plans. As health care costs rose over the next 2 decades, benefits managers opted for managed care programs to reduce insurance costs, but employees felt restricted from accessing the health care services they thought they needed. Clearly, policy makers in 1974 did not foresee the introduction of managed care and the utility of state-level oversight of these more cost-conscious plans. The ERISA case is another lesson in the difficulties of fixing unintended consequences, given the complexity of issues, the public’s short attention span for any one issue, and the fast-paced dynamics of the health care system.

Impact of 1998 Medicare Restructuring on Nutritional Care for the Elderly

As a nutritionist, Johanna Dwyer has closely followed the changes in Medicare requirements for and billing of nutrition services. Her case study, done in collaboration with Ann Coulston, Mary Fitz, and David Pelletier, described the fragmentation of nutritional therapy for frail elders following new Medicare regulations in 1998 (Federal Register, 62 (244):66726–66763) that use an outcomes approach for monitoring the quality of care at participating hospitals.

The 1998 changes in Medicare Part A (the program for hospitals) eliminated the standards that included staffing requirements for a qualified dietitian. The intention of the 1998 Part A shift to an outcomes standard was to hold hospitals accountable for the nutritional health of their patients rather than for the number of hospital

dietitians, but necessary specifications and quality assurance mechanisms are not yet in place.

Adding to the dynamics, Medicare Part B (the program for ambulatory services) has failed to keep up with the shift of chronic care from inpatient to outpatient care. The 1998 law excluded nutritional services from Part B billing and failed to recognize malnutrition as a comorbidity or complication. These incomplete Medicare policies set up the unintended consequence of inadequate nutrition therapy for frail elders both in inpatient and outpatient care. Whereas, linking payment to outcomes could be viewed as improved public policy, in this case no specific outcomes or quality assurance guidelines related to nutrition for frail elder Medicare patients were sufficiently detailed. The lesson here is that incomplete incremental policy development spawns unintended consequences.

Social Security Exclusion of Patients in Institutions for Mental Diseases

Jeffrey Geller has long been interested in the state-federal struggle over who is responsible for the care of people with chronic mental illness. He has searched for a mental health policy that would “do what is right and might actually cost some money,” rather than mental health policies that follow the money and reconstruct a policy ideology after the fact. Geller’s case study traces the dynamics of incremental provisions that led to a vast mentally ill homeless population.

The foundation of the Institution for Mental Diseases exclusion came as a result of the 1950 Social Security Amendments (P.L. 81–734), designed to ensure that states continued their traditional responsibility for long-term care of the mentally ill in state psychiatric hospitals. This provision, with later modifications, ruled that an individual in a psychiatric facility with more than 16 beds, or in a facility in which over 50 percent of the beds are occupied by patients with mental illness, was not entitled to Supplemental Security Income.

When the 1965 Medicaid program (P.L. 89– 97) passed, the federal financial participation rules created an opportunity for states to shift the cost of care and treatment of the mentally ill to the federal Medicaid program. In response to these new financing incentives and significant lobbying by patient advocates, thousands of chronically mentally ill patients were moved from state psychiatric hospitals to Medicaid-eligible community-based services, a policy later labeled “deinstitutionalization.” In addition, the 1997 Balanced Budget Act (P.L. 105–33) began a phasing down of the percentage of a state’s DSH payments that could go to state psychiatric hospitals, eliminating the last source of federal support for long-term inpatient care of the mentally ill.

The lesson from this policy case study is that incremental policies passed in the interest of short-term cost savings may neglect or disregard potential long-term consequences.

According to Geller the unintended consequences are:

EACH-RPCH Demonstration Grant Programs and Rural Health Care

Barbara Langner offered an example of a positive result from unintended consequences. During her RWJ health policy assignment with the Senate Finance Committee, Langner focused on issues of rural health care and worked on the Essential Access Community Hospital (EACH) and Rural Primary Care Hospitals (RPCH) demonstration projects.

The 1983 Prospective Payment System had set rural Medicare payment rates at 40 percent lower than urban rates, and within 5 years a record number of rural hospitals had been forced to close. Concerned that rural Medicare beneficiaries would lose access to care, Congress approved EACH-RPCH, a minor provision of the 1989 Omnibus Budget Reconciliation Act. Working on state health policy after her RWJ congressional experience, Langner had the

opportunity to closely examine implementation of this program in her home state of Kansas.

Communities were constructing effective local solutions for improving access to care for elderly citizens in rural communities through their local EACH-RPCH program. Langner described EACH-RPCH as a lesson about “getting it right.” In Kansas, EACH-RPCH allowed “the community to create a fiscal and community solution” for providing Medicare beneficiaries with access to adequate health care services. Placing responsibility at the local level was more effective in this case than a “one size fits all” federally directed approach. In addition, the EACH-RPCH legislation came at a time of crisis when many of Kansas’ small, rural hospitals were in danger of closing. This impending crisis motivated the major stakeholders to work together toward a solution.

During the day’s discussion of the history and dynamics of major health policy programs, several observations by participants emerged as recurring themes.

The Essence of Time

Time plays a role in how a policy and its unintended consequences can unfold. Stakeholders frequently warn of unintended consequences during the legislative process, but some time must pass to see which ones actually occur. Marsha Gold commented that some consequences are the playing out of known trade-offs made during the policy process. Some do not occur as predicted. For example, stakeholders were warned of the risk of crowd-out during the coalition process to develop SCHIP, and provisions were included to avoid enlarging the public-sector role in children’s insurance instead of just picking up the currently uninsured. The response turned out to be substitution, not crowd-out, as families elected to shift their children from private insurance to SCHIP or from Medicaid to SCHIP. The unintended consequence, visible 3 years later, is an increase in the public-sector insurance burden and a growth in the number of uninsured children.

The time over which policy implementation takes place spawns unintended consequences. Regulations to implement a new program are developed over time but must operate in different health policy environments that include dynamics not present when the legislation was passed. For example, SCHIP passed a year after the 1996 federal welfare reform, when states were consumed with overhauling the administration of their welfare and Medicaid programs. SCHIP implementation was delayed as states consolidated multiple social service units into single omnibus agencies, converted their welfare system to the Temporary Assistance for Needy Families (TANF) job placement program, constructed a new Medicaid eligibility system independent of TANF, and converted the Medicaid system from a state-managed indemnity program to a third-party managed care program.

Four years after welfare reform and three years after SCHIP, Bucciarelli reported that “30 percent of uninsured children are Medicaid-eligible and their parents are unaware of their eligibility.” The dynamics of welfare reform and Medicaid managed care produced the unintended consequences of SCHIP.

Barbara Langner also stressed the importance of timing in the success of EACH-RPCH. When EACH-RPCH passed, public concern had peaked in Kansas’ rural communities over the large and growing number of rural hospital closures. If EACH-RPCH had passed much earlier, before the Kansas state government, state hospital association, and rural communities believed they faced a crisis, these stakeholders would not have formed the partnership or agreed to the prescriptive requirements of these demonstration projects. EACH-RPCH found stakeholders highly motivated, and the partnership created a successful model for locally coordinated health care services.

Policy Arises from Politics

Roger Herdman, Beth Fuchs, William Brew, and other participants reminded discussants that policy arises from a political process in which the decision-making priority is reelection. In the political process there is an inherent desire to act. Political realities often deter policy modification and refinement to correct unintended consequences arising from imperfect legislation.

Larry Lewin reminded the group of the two “laws of unintended consequences”: (1) “incompleteness” stemming from numerous incremental policies that lack a systematic approach, and (2) “inflexibility” due to continuous lobbying by stakeholders to protect their gains. Faced with a zero-sum game to redistribute fixed resources, winners lobby hard to protect the status quo.

Marion Ein Lewin described the difficult debates regarding DSH during the recent IOM safety net study (Lewin and Altman, 2000). Although “stakeholders recognized that the current DSH payment program is ‘out of sync’,” and that it threatens public hospitals’ financial survival, any attempt to modify the unintended consequences of DSH brings with it a new and uncertain policy and political process. Intense competition for both the public’s and politicians’ attention to new crises distracts attention from fixing other policies. Without sufficient public demand for change, politicians and other stakeholders often face greater cost than reward in pursuing policy refinement.

Politics Trumps Science

In the political process, science is more often than not crowded out by politics, a lesson several RWJ alumni said they learned from their congressional experiences. In the benefits mandates policy case study, Balas observed that, “Conversations of politically correct versus scientifically accurate policy proposals are independent monologues that fail to converge through the entire policy debate.”

A few participants worried that unintended consequences of mandating medically unnecessary disease-specific management are (1) instilling in the public a false sense of security, and (2) placing an unnecessary burden on the health care system.

Politics trumped science in the Agent Orange debate. The issue rapidly became highly visible in an era of public cynicism and mistrust of government after Vietnam, together with growing environmental consciousness. Public sentiment for quick action grew faster than the scientific knowledge, as scant prior work on long-term effects translated into an inability to determine definitively whether exposure to Agent Orange could cause delayed illness or hereditary defects. In Charles’s view, the Agent Orange compensation policies passed, at least in part, because scientists did not understand the political process and Congress did not understand the scientific process.

Others added examples from their own experiences. Etheredge told of a colleague’s anxiety over reporting to a senator that the scientific results were inconsistent and lacked enough consensus to make a policy recommendation. To Etheredge’s colleague s surprise, the senator said, “That’s wonderful because it gives us even more room to do the politics.”

Balas told of a conversation early in his RWJ fellowship congressional assignment when he expressed exasperation over the lack of science entering the policy debate. When he told a prominent senator with major responsibilities for health care issues that, “The public is focusing on the wrong question, [that senator’s] reply was, ‘What the public asks is what we need to answer.’”

Beth Fuchs said she thought the tension between science and politics had increased: “As Congress has become more politicized and partisan…the interest in science and objective analysis starts to take a second place.”

Sally Satel said, “When ‘victim politics’ transcend science, facts can be useless in the debate,” giving the example of the Senate’s decision to mandate mammography coverage for women in their 40s despite contrary medical evidence.

Rita Vandivort said that, in the zero-sum game for redistribution of fixed resources, heightened public sympathy for some medical conditions over others creates a debate over the deserving and undeserving. She asked, “How do you infuse science into this?”

The workshop discussion closed with comments on how unintended consequences might be minimized during and after the development and implementation of new health policy programs. Peter Hasselbacher warned that avoiding the risk of unintended consequences is the ideological reasoning of libertarian and conservative think tanks opposed to new legislation or regulation and a tool for avoiding doing anything at all. With the understanding, as Marion Ein

Lewin said, “that unintended consequences are an inherent part of the political and policy process,” participants also explored how to improve the chances of later policy refinements.

Discussants looked at the role of timely information in reducing unintended consequences. Clyde Evans asked if public forums and focus groups made up of those groups potentially most affected by a proposed policy could identify the unexpected responses described in the policy cases such as SCHIP.

Etheredge supported this idea with the story of a former assistant secretary of labor who, faced with several proposals and no consensus, went to the custodian’s office on lunch break. After reviewing the basics of all the proposals with the custodian, who represented those most affected by the proposed plan, he asked, “So which would you like?” and was told “none.” The assistant secretary returned to his office to draft new plans.

Crittenden said that information is more valuable the earlier it is available. Information from pilot projects and demonstrations has more impact if results inform the debate early on. “If people don’t believe (the science), you need three years of changing ideas, attitudes, (and) thoughts (of) the key constituencies. If you don’t do that, all this good information (from pilots and demonstrations) goes for naught.”

Small demonstration projects can limit unintended consequences and confine the risks while seeing whether expected benefits will be realized. Langner thought the EACH-RPCH demonstration project was small enough to allow the local experimentation necessary for success. The medical savings accounts (MSA) demonstration provision of the 1996 Health Insurance Portability and Accountability Act (HIPAA) limited the anticipated harm to insurance risk pools while allowing for the observation of whether the public would opt for MSAs as an affordable insurance option.

Checkpoint mechanisms built into health policy programs reopen the door to imperfect policies. Jason Lee reminded discussants that legislation requiring demonstrations and annual reports is intended to create opportunity to revisit the policy debate. Beth Fuchs added that most major health policy legislation calls for a report back to Congress to monitor for harmful outcomes. These reports, from the secretary of the Department of Health and Human Services, the Medicare Payment Advisory Commission, or, in the case of MSAs, the General Accounting Office, can prompt review without waiting for public demand. “Sunset” provisions with automatic expiration dates create additional opportunities for refinement of imperfect policies. Balas commented that a sunset rule would at least limit the lifetime of a bad mandate for a medically unnecessary treatment.

Discussants reviewed how to better integrate the parallel monologues of politics and science. Jeffrey Biggs suggested that the scientific community needs to educate the public as well as Congress. Balas suggested that the legislative process might benefit from a scientific analysis somewhat akin to the Congressional Budget Office financial analysis of health proposals. Etheredge concluded that the RWJ health policy fellowships program has played a significant role in interjecting clinical science into the policy process, as evident from this workshop discussion. Beyond their year in Congress, he noted, RWJ fellows continue to contribute to health policy when they go back to their communities and careers and, as here, construct “a dialogue with national policy makers about intended consequences versus how policies work in practice.”

Breaux, J. 2000. Foundation fellows bring unique perspective to health policy. Washington Outlook.

Carson, R. 1964. Silent Spring. New York: Fawcett Crest.

Institute of Medicine. 1997. Information Trading: How Information Influences the Health Policy Process. Lewin, M.E., and E. Lipoff, eds. Washington, DC: National Academy Press.

Institute of Medicine. 2000. America’s Safety Net Providers: Intact but Endangered. Lewin, M.E., and S.Altman, eds. Washington, DC: National Academy Press.

Neustadt, R.E., and E.R. May, 1986. Thinking in Time: The Uses of History for Decision-Makers . New York: Free Press.