Health Affairs

This Article Abstract Submit a response to this article Services E-mail this article to a friend Alert me to new issues of the journal

Read related perspectives by Jack Wennberg,
Clifton R. Gaus, and Carolyn M. Clancy.

H E A L T H  S E R V I C E S  R E S E A R C H : 
A H C P R  P O L I T I C S
W E B  E X C L U S I V E   

25 June 2003  
  

AHCPR And The Changing Politics 
Of Health Services Research

Lessons from the falling and rising political fortunes of the
nation’s leading health services research agency.


by Bradford H. Gray, Michael K. Gusmano, and Sara R. Collins

ABSTRACT:

The Agency for Health Care Policy and Research has had a turbulent history. Created with little opposition in 1989, it narrowly escaped being eliminated in 1995, only to be reauthorized (with a new mandate and name—the Agency for Healthcare Research and Quality, or AHRQ) with overwhelming support in 1999. In focusing on budgetary history, this paper sheds light on why health services research (HSR) has difficulty obtaining funding from a government that is willing to spend vast sums on basic biomedical research. The paper argues that three strategies—bureaucratic, marketing, and constituency building—that advocates adopted in the late 1980s made HSR more visible and consequential and were responsible for AHCPR’s budgetary successes as well as its near-demise.

Coherence and consistency are not hallmarks of the American health care system. Nonetheless, politicians’ resistance to spending money for research to increase understanding of the structure, process, and effects of health services while committing enormous resources to basic biomedical research is striking. An instructive window on the reasons can be found in the turbulent history of the Agency for Health Care Policy and Research (AHCPR), which existed between 1989 and 1999 as the leading government agency for health services research. It was preceded by one agency and followed by another—first, the National Center for Health Services Research (NCHSR), then the Agency for Healthcare Research and Quality (AHRQ). Lessons from the agency’s creation and its near-demise in 1995 provide some insight into why health services research (HSR) has such difficulty securing ongoing support.

The political problem of health services research. The Institute of Medicine (IOM) defines health services research as “the interdisciplinary field that investigates the structure, processes, and effects of health care services.”¹ Exhibit 1 compares the budget of the federal government’s lead HSR agencies in 1980, 1990, and 2000 with the budget of the National Institutes of Health (NIH), whose mission is basic biomedical research.² The orders of magnitude are completely different, with HSR at about 1 percent of the NIH budget. Not only have the HSR agencies struggled to grow, but they have at times been threatened. The fiscal year 2003 budget is the most recent example, with President George W. Bush proposing a 16 percent cut in the AHRQ budget and a 16 percent increase in the NIH budget.

It may seem counterintuitive that basic biomedical research has garnered eager bipartisan support over several decades, while agencies whose central mission is research on such practical matters as the cost, quality, use, and outcomes of health services perennially struggle in budget battles. The reasons for NIH’s success are reasonably clear. NIH benefits from (1) a purpose that is understood in Congress and identified with a widely held concern—serious diseases; (2) strong support from powerful constituencies from the prestigious academic medical centers that are at the cutting edge of patient care, as well as from disease-oriented associations whose spokespersons (including celebrities) often speak from personal experience; and (3) a mission that both stimulates economic growth (basic research, leading to patient care technologies) and collides with few vested economic interests. NIH commonly does well in presidential budget requests, and Congress sometimes appropriates more funding than the president requests. The relationship between NIH and Congress is well symbolized by the seven buildings on the sprawling NIH campus in Bethesda, Maryland, that are named after former key members of the Senate or House appropriations committees.³

Health services research has struggled in all stages of the budgeting process—sometimes within the Department of Health and Human Services (HHS); often at the Office of Management and Budget (OMB), where the president’s budget is put together; and almost always in the appropriations process, particularly in the House. Exhibit 2 illustrates the funding history of the lead HSR agencies over the past two decades. Interviews with HSR advocates and congressional staffers in the early 1990s provide a plausible account of the field’s political problems before the creation of AHCPR.⁴ HSR was not easy to explain, and it rarely solved problems that afflict people who could appeal for more funding at appropriations hearings. Its main advocates were researchers and their professional association, the Association for Health Services Research (AHSR, now known as AcademyHealth), which was established in the early 1980s to advance the field. A lack of visible external constituencies lent a self-interested flavor to researchers’ efforts to make a compelling case for support, and they were not themselves a sufficiently large or wealthy constituency to command favorable treatment on those grounds. Through the 1980s NCHSR’s funding was stagnant, except for the 1987 increase for the National Medical Expenditure Survey (NMES).

A change in the politics. The field’s fortunes changed in 1989 with the passage of legislation that established AHCPR. The effect can be seen in Exhibit 2, with the near-doubling in the FY 1990 budget. The agency was created with little opposition within Congress, yet within six years it was threatened with extinction in the congressional budgetary process. It survived, but with a sharp budgetary reduction in FY 1996. Four years later it was reauthorized with a new name and mandate by an overwhelmingly positive vote.

In this paper we examine the ten-year history of AHCPR, focusing particularly on congressional support, which translates into budget. The paper is based on almost 100 interviews with people who were involved in different aspects of the agency’s history on Capitol Hill, in the White House, within the agency, and outside of government.⁵ We suggest that AHCPR’s creation fundamentally changed the politics of HSR, changing it from an obscure activity of concern to a narrow group into a visible and potentially important activity affecting politically divisive topics. This change, we further argue, was the result of bureaucratic, marketing, and constituency-building strategies adopted in the late 1980s by leading HSR advocates to change the field’s seemingly hopeless budgetary stagnation. As a consequence, the politics of HSR became much more complex and volatile, turning more heavily on its practical applications.

The Creation Of AHCPR

The Agency for Health Care Policy and Research was established as a direct result of the efforts of AHSR. We have previously described the circumstances, but here we highlight the strategies involved because they transformed the politics of health services research.⁶

During the 1980s the leaders of NCHSR and AHSR struggled unsuccessfully to convince budgetary decisionmakers in the administration and Congress of HSR’s merits, often citing the development of Medicare’s diagnosis-related group (DRG) system for hospital payment as an example. The budgeteers proved resistant to arguments for substantial increases in HSR funding, whether the rationale cited evidence of health system shortcomings or the amount the country was spending on health care or biomedical research. A change in strategy was adopted; this change led to the creation of AHCPR.

The key development was finding an opportunity to link the agenda of an improved funding base for HSR to issues for which an important constituency already existed on Capitol Hill. The opportunity arose in 1989. Several members had become aware of John Wennberg’s research on practice variations and RAND studies showing widespread inappropriate use of common surgical procedures. They were receptive to proposals for a new program of research on the outcomes and effectiveness of medical treatment. William Roper, the White House health policy adviser who had become an advocate for “effectiveness research” while head of the Health Care Financing Administration (HCFA, now the Centers for Medicare and Medicare Services, or CMS), was responsible for the fact that such a program was called for in President George H.W. Bush’s proposed FY 1990 budget.⁷ In addition, serious concerns on Capitol Hill about health care costs and Medicare’s financial viability created receptiveness to suggestion that outcomes research, technology assessment, and the development and dissemination of practice guidelines would produce cost savings.

This budding interest provided a vehicle to which other ideas could be attached. An outcomes research initiative did not require a new agency, but AHSR and allies on the staff of the House Energy and Commerce Subcommittee on Health saw an opportunity to make some bureaucratic and marketing moves that, it was hoped, would improve the position of HSR. The first bureaucratic move was to build the outcomes/effectiveness agenda into an amendment to the Public Health Service (PHS) Act that would elevate NCHSR’s activities and personnel into a new PHS agency alongside NIH, the Food and Drug Administration (FDA), the Centers for Disease Control and Prevention (CDC), and the Health Resources and Services Administration (HRSA). This new agency would carry out new outcomes research and practice guidelines responsibilities, while continuing NCHSR’s previous HSR programs. The elevated location was expected to lead to better budgetary treatment.

The agency’s name involved a marketing decision. Health services research, an AHSR lobbyist once said, was as difficult to sell as a dead fish wrapped in newspaper. The term “health care research” was adopted instead, and the word “policy” was added to convey that its mission went beyond research. (The initial name, Agency for Health Care Research and Policy, was changed late in the legislative process when a Senate staffer realized that the acronym would be “AHCRAP.”)

A second bureaucratic move concerned funding. The PHS Act amendment included a new source of funding in addition to an ordinary authorization to appropriate—a tap of 40 percent on the evaluation money that constituted 1 percent of funds appropriated under the PHS Act. In separate provisions with different origins, an amendment to the Social Security Act provided some budget authority under the Medicare trust fund to support the outcomes research program proposed in the president’s budget. Thus, the legislation creating the new agency gave it two new streams of funds.

The maneuvers to create the new agency succeeded in the legislative context of 1989 for several reasons. First, there was a much larger bill (the Omnibus Budget Reconciliation Act, OBRA, of 1989) into which the two sets of amendments could be quietly tucked. Second, outcomes and effectiveness research had well-placed friends in the Bush White House (Roper), the Senate (Majority Leader George Mitchell [D-ME] and David Durenberger [R-MN], ranking Republican on the Senate Finance Committee), and the House (Bill Gradison [R-OH], ranking Republican on the Ways and Means Subcommittee on Health). Third, the legislation attracted no opposition, because the funding for the initiative was in the president’s budget but also because it was handled skillfully (for example, the side-by-side document used in House-Senate Conference Committee to compare different versions of the legislation was innocuously titled “Health Care Quality”) and because the higher-stakes physician payment reform battle provided political cover.

In sum, several pieces of strategic thinking by AHSR leadership were incorporated in the legislation. First, the new agency was to address practical concerns that already had a constituency. The agency (and HSR) thus became identified with the quality and cost concerns for which a constituency had been built, principally through the influence of Wennberg and RAND’s Robert Brook. Second, the legislation incorporated bureaucratic strategy to improve the lot of HSR—its movement into an NIH-level PHS agency and the incorporation of two new sources of funding. Third, the legislation reflected marketing strategy, with the abandonment of the phrase “health services research” and the suggestion that the agency did more than research. These strategies all seemed successful at the outset, but their dangers became apparent within a few years.

The Impact Of The Legislation

The legislation had several important effects. First, the movement of health services research from NCHSR into the new agency did enhance its prominence and visibility. However, although the agency’s architects hoped that a prominent researcher would be given leadership of the agency, the Bush administration named Jarrett Clinton, the current NCHSR head, as the agency’s first director. Clinton was a career PHS physician who had been appointed to the NCHSR position for only a few months, transferring over from HRSA. Although an experienced federal bureaucrat, he had no research background. He retained two senior NCHSR administrators, Norman Weissman and Donald Goldstone, in the key directorships of the “extramural” and “intramural” research programs, a structure that also came over from NCHSR.

Second, the legislation substantially changed HSR’s budgetary fortunes. The funding moved to a new plane—from $53 million in 1989 to $97 million in 1990, a remarkable change in view of the field’s history. The agency had steady budgetary growth over the next few years, reaching $162,386,000 when the FY 1995 appropriation was approved in October 1994. In two of those years Congress actually appropriated more money than requested by the administration (in both cases, the Bush administration). The first Clinton budget proposed a 23 percent increase (for FY 1994), and Congress approved almost 97 percent of the requested amount.

Third was the creation of a new program for developing practice guidelines and an enhanced and focused program of outcomes research. Central to the legislative changes that had improved HSR’s fortunes had been advocates’ success in convincing key actors that the agency would actually have an effect on important problems in medical practice and costs. Thus, the agency’s responsibilities involved a degree of redefinition of “health services research,” at the border where research meets practice. The two main problems that engaged the agency in its first few years were on this borderline. Both became important in the 1995 attack on the agency, so they bear discussion in some detail. The topics were medical practice and health reform.

The Agency And Medical Practice

AHCPR’s statutory responsibilities regarding outcomes research and practice guideline development reflected an expectation that the agency could do something about the problems revealed in research showing large practice variations and extensive inappropriate use of services. The agency’s work on outcomes focused on the funding of Patient Outcomes Research Teams (PORTs)—multidisciplinary centers that would focus on particular medical problems and review and synthesize available research, analyze practice variations and patient outcomes using administrative data augmented by primary data collection, disseminate the results, and evaluate the effects of dissemination. Fourteen five-year PORTs were funded in the first two years, at about $1 million per PORT per year.⁸ The agency was also supposed to develop practice guidelines for at least three conditions by January 1991 and to disseminate them and report to Congress in early 1993 about their impact on the quality and cost of medical care. Notwithstanding the quality implications of the problems that stimulated the outcomes research and guideline programs, costs were clearly a major congressional concern.

The history of the agency’s work regarding medical practice is to be the subject of a separate paper, but two points are germane to the strategies that had brought AHCPR into being. First, because Congress expected that the agency would affect medical practice and health care costs through the outcomes research and practice guideline programs, several evaluations and hearings regarding the agency’s performance took place in the first few years. Second, Jarrett Clinton, the agency’s first director, recognized both the practical and the political need to work with specialty medical groups and invested heavily in such activity. This constituency building later proved helpful.

The Agency And Health System Reform

Health system reform provided an opportunity for AHCPR to be useful in addressing the administration’s concerns. Notwithstanding the word “policy” in the agency’s name, the only explicit policy roles in its legislative mandate were in developing practice guidelines and in making recommendations to the HHS secretary regarding whether specific medical technologies should be covered by federal health insurance programs. Our interviews indicate that the key actors in AHCPR’s creation did not foresee that the agency might be asked to play a role in developing health system reform, a topic that had not been on the agenda in 1989.⁹ Even so, as an executive-branch agency, AHCPR had two important assets that would be useful to the administration in the policy-making process. The first was the intramural research program that collected and analyzed data—most notably in the form of NMES—that could be used in analyzing the implications of different policy options. Second, it employed a substantial cadre of researchers who knew about those data and health policy issues more generally. These assets had been used somewhat in previous policy debates, as in predicting the cost of the prescription drug benefit in the Medicare catastrophic legislation of 1987.

AHCPR became involved in President Bush’s work on a health reform proposal, and it subsequently participated in the development of President Clinton’s Health Security Act. In the view of AHCPR leadership, having its personnel involved in major White House policy initiatives had substantial benefits, fostering personal relationships and demonstrating the practical utility of agency activities in the circles in which presidential budgetary decisions were made.

Bush health reform. The Bush administration’s health reform effort began with the assembly of a multiagency working group, after Harris Wofford’s 1991 election to the Senate signaled that health reform had potency as a political issue. AHCPR involvement emerged out of contacts between Stephen Bandeian, special assistant for health policy development at OMB, and Donald Goldstone, director of the AHCPR Center for Intramural Research. Bandeian came to believe that there was “no stronger group of health care analysts in the federal government” and that the 1987 NMES, despite important limitations, was the best source of the information about health services’ use and cost that was needed in designing major reform proposals.¹⁰ Working within a framework set by the domestic policy team, Bandeian led a small group that drafted President Bush’s reform proposal over the holiday break before its release in February 1992.

During this period, Bandeian and Goldstone hatched an idea for bringing OMB and AHCPR closer together. Bandeian saw that OMB lacked the modeling capacity needed to estimate the cost and impact of reform proposals, and he worried about reliance on analyses by proprietary firms. Early in 1992 he and Goldstone negotiated what Goldstone characterizes as a “nice political arrangement” whereby an AHCPR senior research analyst would be detailed to OMB for up to six months.¹¹ Economist Dean Farley played this role during the Bush administration, as did economists Pamela Short and Len Nichols during the Clinton administration.

A physician/attorney with years of experience with the Ways and Means Health Subcommittee before coming to OMB, Bandeian was knowledgeable in health policy. And, although disappointed in AHCPR’s ability to meet many data requests, he believed that agency personnel did their best; he described them as “excited to be involved in something of national importance” and having a “workmanlike, professional attitude.”¹² Goldstone and the staffers who were involved described their approach in similar terms.¹³

Gail Wilensky, who later became one of the agency’s most important advocates when it came under congressional attack, holds a more critical view of the agency’s performance during this period. When the Bush health plan was being developed, Wilensky was head of HCFA, before moving to the White House in 1992 as deputy assistant for policy development. She said this of her efforts to get information from the agency late in the Bush administration: “Based upon my experience as a researcher and as someone who knows how bureaucracy worked, the agency was very unhelpful in supporting the analysis…I was asking for some very specific numbers and estimates—not to have the staff doing wholesale modeling, just estimates, distributions, fields that hadn’t already been coded properly and analyzed. And I was getting very little cooperation. Lots of talk but very little action, a pretty intransigent response.”¹⁴

There are several possible explanations for Wilensky’s experience. Unrealistic expectations or misinterpretations seem unlikely, since she was a health economist who had helped design NMES while at NCHSR. Her negative experience also could have resulted from bureaucratic rivalry stemming from her position as HCFA administrator. This is not implausible, since HCFA was well established as a source of budget estimates, and AHCPR was something of an outsider.

Wilensky’s interpretation is shared by some, but not all, Republican observers of the agency’s performance during the Bush and Clinton administrations. Wilensky believes that AHCPR had a political bias—that it was less responsive to data requests from the Bush health team than it was, later, to the Clinton team, for which it provided analyses that were influenced by political considerations. Almost everyone else we interviewed who was involved in the agency’s work on health reform in the Bush and Clinton administrations, including some others who served in the Bush White House during that period, rejects this view. However, Wilensky believes that had President Bush won the 1992 election, the agency would have suffered in the next Bush budget. Although this is speculative, it does point to the risks that can arise when a research agency receives legitimate requests for assistance in a politically charged context.

Clinton health reform. AHCPR played several roles during the Clinton administration’s eighteen-month effort to develop and pass a comprehensive health care reform plan. First, it was likely the federal government’s largest supplier of personnel for the task force and working groups that developed the ideas from which the components of the Clinton health reform proposal were selected. The agency did not become identified with any specific ideas, however. In the view of then director Jarrett Clinton (a Bush appointee), the agency was “expected to be helpful,” but “most people from the agency were not involved in the debates about policy.”¹⁵ In his view, the agency’s work on health reform was “not political” but was a “response to the administration” and “a reasonable contribution on our part.” In any event, the task force’s work lasted only four months.¹⁶

Second, the agency played an analytic role in a direct extension of work done during the Bush administration. Using NMES data and the HSR literature, agency personnel carried out microsimulations of the economic consequences of policy options being considered for the Clinton health reform proposal both while the task force worked and over the next year as legislation was developed.¹⁷ The agency’s role was facilitated by Goldstone’s assigning economist Len Nichols to work with Bandeian at OMB during the transition after the 1992 election.

Nichols was assigned to several working groups under Ira Magaziner, who had overall responsibility for developing the Clinton proposal. One was the cost estimation group. In Nichols’ account, Goldstone wanted AHCPR’s health economists to be able to provide cost estimates to the president’s policymakers, based on microsimulations using NMES data. The usual authoritative source of cost estimates on such matters was the HCFA Office of the Actuary, which had the “brand name” for cost estimates on Capitol Hill and in the Department of the Treasury. Nichols, with the backing of OMB deputy director Alice Rivlin, convinced Magaziner early on of the advantages of having two sets of economic estimates. As Nichols put it, “That gave a fair shot to Goldstone and made AHCPR an equal of HCFA in health plan development.”¹⁸

Space constraints preclude a detailed account of AHCPR’s analytic role in the first eighteen months of the Clinton administration, but several points are notable. First, the agency was well positioned to contribute because of the OMB relationship and modeling work begun in the Bush administration. In Nichols’ words, the Clinton plan required “worldwide flying,” and the agency had already at least “built a plane that could fly to Philadelphia.” Second, once the AHCPR model was fully operative, it could respond quickly to changes in assumptions and options. Magaziner’s modeling and data analysis group (Ken Thorpe from the office of the HHS assistant secretary for planning and evaluation, David Cutler from the Council of Economic Advisers, and Nichols) worked for a “substantial period” on the economic effects of different policy options.¹⁹ They would meet in the morning to review new data and then would give new assumptions to Farley and his AHCPR colleagues in the afternoon; new runs would be done overnight for discussion the next morning; and the process would begin again. Third, there was indeed a sense of competition with HCFA, but this was a competition that the agency could not win, because the most important judges outside of the White House—the Congressional Budget Office (CBO), Senate Finance Committee, House Ways and Means Committee, and Treasury Department—had long-term relationships with the HCFA actuaries and trusted their analytic approach and their independence from White House influence. The arguments about the budgetary impact of the proposal were ultimately resolved by building in a hard budgetary cap, making choice among competing estimates unnecessary. In any event, of course, the whole proposal failed.

AHCPR’s work on the Clinton health reform proposal provided powerful evidence of the benefits of having constituencies beyond the HSR community. The agency would have benefited greatly had the Health Security Act been enacted. The act’s public health section would have expanded the agency’s role to conduct and support research on the reform of the U.S. health care system, including research on new topics (consumer choice, workplace injuries, risk-adjustment methods, and factors affecting access to care for underserved populations). The act included much larger authorizations—$400 million in FY 1996; $500 million in FY 1997; and $600 million in the next two years. (The agency’s current budget was just over $128 million.) The bill specified that this authorization was in addition to any other authorizations that might be available.

AHSR’s Strategy Comes Inside The Agency

AHSR had achieved most of what it sought when AHCPR was created, having been instrumental in drafting the Waxman bill that emerged from the legislative process with all key components intact. The major disappointment had been that a leader of the field had not been appointed as the agency’s director. This changed in early 1994, when Clifton Gaus was appointed to replace Jarrett Clinton. Like assistant HHS secretary Phillip Lee, who recommended his appointment and who was an AHSR founder, Gaus was in the inner circle of AHSR leadership.²⁰ He had served as its first president after heading HCFA’s Office of Research and Demonstrations for many years. Most recently he had worked on the Health Security Act.

A close observer of the agency (and its predecessor, NCHSR) and a longtime participant in the struggle to obtain additional HSR funding, Gaus had ideas for change that were aligned with the strategies developed at AHSR in the late 1980s. In an early AHCPR senior staff meeting, Gaus posed the question, “Who is our customer?” He received the answer he expected: health services researchers. Gaus disagreed, explaining that the agency had to consider its customers to be those who would make use of the products of its work and that the research community should be seen as being the agency’s partners in meeting the needs of the customers on whose goodwill the agency’s support would depend. This was a fundamentally different way of thinking about setting priorities and relating to the world. By seeking to understand and meet the needs of potential users of its work, the agency could build broader and deeper support than HSR had ever enjoyed, Gaus believed. He also had marketing ideas: He believed that the traditional organization of research activities into “intramural” and “extramural” centers did not facilitate political support because it failed to convey any substance about the work being done. Accordingly, he reorganized the agency’s activities into research centers that had a substantive focus (such as the Center for Outcomes and Effectiveness Research) and that involved researchers from both of the former intramural and extramural programs. He also brought new leadership (Ross Arnett from HCFA) for the agency’s main data collection effort: the collection of national health care use and spending data. The agency’s survey, NMES, was redesigned to make the data more complete and more current; it was renamed the Medical Expenditure Panel Survey (MEPS). Finally, Gaus ratified an emerging decision to move away from the agency’s signature activities: the PORT grants and the guideline development efforts, both of which, evaluations suggested, were neither effective nor cost-effective.

Thus, as 1994 unfolded, the bureaucratic, marketing, and constituency-building strategies that had led to AHCPR’s creation appeared to be a substantial success. However, the dangers that those strategies held for a health services research agency were thrown into sharp relief after the dramatic midterm election of 1994.

From New Life To Near-Demise In A Half-Dozen Years

In its first few years AHCPR carried out its new statutory responsibilities, continued the research missions it inherited from NCHSR, and became involved in the health system reform efforts of the early 1990s. The agency enjoyed steady budgetary growth—from $115 million in FY 1991 to $159 million in FY 1995. It underwent an apparently successful leadership transition from a quiet bureaucrat who had carefully managed the agency’s implementation of its legislative mandate to a gregarious organizational politician who had good connections in the Clinton administration, was well known in Washington, and had strong ideas about change.

The picture of success changed suddenly with the 1994 congressional election, in which the Republicans gained control of both the House and the Senate with a broad government-reduction agenda: the Contract with America. The federal budget for FY 1996 became the focus of an extraordinarily contentious battle between the administration and Congress that culminated in government shutdowns; the 1996 appropriations bill was not passed until April 1996, the seventh month of the fiscal year. The House of Representatives, led by Speaker Newt Gingrich (R-GA), set out to implement sweeping changes to roll back the welfare state, reduce federal regulation, and destroy the Democrats’ power base.

Should such an atmosphere have been dangerous for a health services research agency? The Republican agenda was not necessarily hostile to research. Indeed, when the interminable FY 1996 budget process ended, NIH had a 5.6 percent budget increase (more than $700 million) to reach $11.9 billion. However, from the beginning of the budget process, AHCPR was in trouble.

That process began in the House Budget Committee under Rep. John Kasich (R-OH). The agency’s name appeared on that committee’s “hit list” of 140 discretionary programs to be eliminated. In the joint House-Senate budget resolution agreed to in May, the agency was to be cut by 75 percent, with only MEPS surviving. (A “rescission” for the FY 1995 budget also passed in May and took back $3 million from the agency’s budget for the fiscal year that was then in its eighth month, reducing the 1995 budget to $159,386,000). The House/Senate budget committee conference report in June called for complete elimination of the agency’s funding. In a Ways and Means subcommittee hearing in July, Rep. Sam Johnson (R-TX) led an attack on the agency for wastefulness and unwarranted interference with the practice of medicine. The House Appropriations Committee approved an appropriation of $125 million for the agency in July, but when the appropriations bill went to the floor, Representative Johnson proposed a floor amendment to reduce the appropriation to zero. After intense behind-the-scenes negotiations, that amendment was withdrawn, but a modified version was passed by voice vote reducing AHCPR’s appropriation to $65.5 million in the appropriations bill passed by the House. The agency fared better in the Senate, where the Appropriations Committee approved a budget of $127,310,000, but the appropriations bill stalled in the Senate in the budget stalemate. When, after a Senate-House conference, the appropriations bill finally passed in April 1996, the agency ended up with an appropriation of $125,169,000, a 21 percent cut from FY 1995.

This précis of events tells nothing about reasons for the attack or how its most severe potential effects were moderated. These are crucial to the lessons to be drawn from the events.

Why Did The Budgetary Attack Occur?

The change from Democratic to Republican control of Congress in 1995 set the stage for the agency’s near-demise, but it does not suffice to explain it. After all, AHCPR had been created during a Republican administration; its initial funding came in part from provisions (regarding effectiveness research) in President Bush’s FY 1990 budget; and Republicans had been key figures behind the legislation in both the House (Gradison) and the Senate (Durenberger). The transformation on Capitol Hill put the agency into harm’s way for four reasons, all of which relate to the strategies behind its creation.

Reason no. 1: too few friends. AHCPR had never had a broad base of support. As the 1995 session began, Senators Mitchell and Durenberger and Representative Gradison were no longer in office, and most staffers who had been involved in the AHCPR legislation were also gone. Importantly, no new constituency had been built among other members and staffers during the agency’s early years. Jarrett Clinton had not relished the political aspects of his job, and although Gaus had some relationships on Capitol Hill, the turnover of chairmanships and staffs with the change in party control in early 1995 rendered everyone’s Rolodexes obsolete. When the budget resolutions signaled serious trouble, the agency (and AHSR) had few established relationships on which to call, and most of the people who valued the agency’s work were gone.

Reason no. 2: vulnerability to criticisms of waste and inefficiency. Congressional staffers seeking targets for funding cuts in 1995 found reports from three congressional agencies that criticized the agency on grounds that resonated with the budget cutters’ concerns. A 1995 U.S. General Accounting Office (GAO) report, although praising the agency’s guideline development program for its rigor, found the guidelines weak and not user-friendly.²¹ A Physician Payment Review Commission (PPRC) report criticized the effectiveness of the agency’s practice guideline program, arguing that guidelines by themselves don’t generally improve practice and suggesting greater collaboration with professional associations.²² Finally, an Office of Technology Assessment (OTA) report concluded that the agency’s effectiveness research would probably not save money; that AHCPR’s PORTs, as well as their use of administrative databases, had produced disappointing results regarding the comparative effectiveness of alternative treatments; that the agency was but one of “many” federal agencies that issued clinical practice guidelines; and that these guidelines might conflict with those developed by “private groups” and might not suffice to change clinical practice.²³

These reports were initiated before control of Congress changed and were not intended as attacks, but the criticisms were deadly in the political atmosphere of 1995. They were used to support Republican arguments that the agency was inefficient, ineffective, and nonessential. For example, after targeting AHCPR for elimination, Representative Kasich’s House Budget Committee pointed to the OTA report as evidence that its work was duplicative and unnecessary.²⁴ According to staff, the Budget Committee’s decision to eliminate the agency was based in part on the findings of these three reports. Later, Representative Johnson cited the OTA report in a July 31 “Dear Colleague” letter in which he proposed elimination of the agency (mocking it as the “Agency for High Cost Publications and Research”) and again when he offered amendments on the House floor to eliminate, and later, to reduce the agency’s appropriation.²⁵ When Rep. Henry Bonilla (R-TX) spoke in favor of the Johnson amendment, he argued that the PPRC “reported to Congress that the guidelines produced by AHCPR are having little impact on clinical practice, are difficult to implement, and are used infrequently by the private sector.”²⁶ As one congressional staffer put it, the agency became part of the “make-fun-of-government-rhetoric…It was almost the old Bill Proxmire Golden Fleece Award,” with Representative Johnson ridiculing, as a classic example of government waste, an AHCPR study showing that cardiologists are better than primary care physicians in treating heart disease.²⁷

Reason no. 3: identification with partisan politics and the Clinton plan. In the 1994 election, the defeated Clinton health reform proposal best symbolized for Republicans all that they opposed. Few things could have been more hazardous to a federal agency in 1995 than to be linked to that proposal by Republicans on Capitol Hill. Some Republicans whom we interviewed believed that the agency’s role in health reform had been appropriate and was not a major source of its 1995 political difficulties. However, others did make the link, and there is evidence that the health reform experience contributed to the agency’s 1995 travails.

In that context, several aspects of the agency’s activities raised problems. First, the word “policy” in its name implied that the agency had a role in health policy, a term that some members identified with the Health Security Act. Second, the Clinton proposal had assigned important monitoring responsibilities to AHCPR. When Sen. Pete Domenici (R-NM), chair of the Senate Budget Committee, proposed a 75 percent cut in the AHCPR budget, his justification was that the agency “was to be the primary administrator of comprehensive health care” and, thus, would not now be needed.²⁸ This was one of many indications that the agency suffered from a lack of understanding of its activities. A perception that AHCPR had played a partisan role in health reform was evident in the House Budget Committee’s justifications for eliminating it, where it was alleged that the agency had “performed an advocacy role in the health care debate the previous two years while its funding increased from $125 million in 1992 to $163 million in 1994.”

Republican critics were also aware that the agency had become the place of employment for several people who had been involved with the Clinton plan, as well as several former Democratic staffers on Capitol Hill. They knew that before Gaus was named director of the agency, he had had lead responsibility in HHS for developing the PHS sections of the Health Security Act. It was noticed when the agency hired Christine Williams, who had been Senator Mitchell’s legislative assistant for health before that Democratic leader’s retirement, as well as Walter Zelman (who had prominently worked on health reform in the White House), Jamie Reuter (a top Democratic staffer from the Ways and Means Health Subcommittee), and Peter Bouxsein (a well-known former staffer of Rep. Henry Waxman [D-CA]).

Reason no. 4: committed enemies. AHCPR was also confronted in 1995 with an advocacy organization’s active efforts to get it defunded. The source was an association of back surgeons who disagreed with conclusions reached by the PORT on low-back pain and with practice guidelines based on that work. The agency had had previous experience with criticism by medical groups. In 1993 a large prospective outcome study by the cataract PORT came under attack from the American Society of Cataract Surgery, the Outpatient Ophthalmology Society, and the American Board of Ophthalmologists. However, that attack never extended to the agency itself, and it ended when the PORT’s data were useful to ophthalmologists in (1) discrediting a GAO study alleging that inappropriate cataract surgery was widespread and (2) causing Milliman and Robertson, a consulting firm, to withdraw a practice guideline that had led some insurers to refuse to pay for some surgery.²⁹

The 1995 controversy over back surgery posed a much more serious threat because it focused on the agency’s appropriation. The surgeons found sympathetic ears among House Republicans who, for reasons already discussed, were prepared to believe the worst about the agency. The events of 1995 followed many years of controversy over the merits of surgical procedures for low-back disorders. AHCPR entered this fray when its PORT on low-back pain reviewed the research and concluded that there was no evidence to support spinal fusion surgery and that such surgery commonly had complications.³⁰ The North American Spine Society (NASS) created an ad hoc committee, which attacked the literature review and the subsequent AHCPR practice guideline on acute care of low-back pain. In a letter published in 1994 in the journal Spine, the committee not only criticized the methods used in the literature review and expressed concern that the conclusions might be used by payers or regulators to limit the number and types of spinal fusion procedures, but it also charged that AHCPR had wasted taxpayer dollars on the study.³¹ An entity known as the Center for Patient Advocacy was formed by Neil Kahanovitz, a back surgeon from Arlington, Virginia, to lobby on the issue. It organized a letter-writing campaign to gain congressional support for its attack on AHCPR. Kahanovitz used personal contacts to gain the support of Representatives Bonilla, who had a staffer who was Kahanovitz’s patient; Johnson; Gerald Solomon (R-NY); and Joe Barton (R-TX). Solomon, Bonilla, and Johnson led the effort in the House to end the agency’s funding, energetically promulgating the NASS/Kahanovitz argument that it was supporting unsound research and wasting the taxpayers’ money.

How Did The Agency Survive?

In the highly contentious political context of 1995, where fundamental issues were at stake and an extraordinary search for budget cuts was occurring, a small research agency that was vulnerable to charges of inefficiency and waste, had a partisan tinge, and had some committed enemies was in serious trouble. Potential defenders within the Clinton administration were engaged in other battles with higher stakes. For months the fight for the agency’s survival became the primary activity of Gaus and AHSR. They sought help wherever they could find it, argued for the value of the agency’s work in whatever forums were available, rebutted arguments that had been made against the agency and its work, and coordinated a diverse set of agency supporters.

AHCPR was politically weaker in 1995 than it had been in 1989 in three ways: It no longer had supporters in key positions; it now had active enemies; and it faced challenges to its rationale, legitimacy, and performance. In its defense, it had arguments, allies, and a threadbare base of support on Capitol Hill.

Support for continued existence. Two elements constituted the core arguments for the agency’s continued existence. First, it had been created with bipartisan support in a Republican administration, the ideas behind its work had been embraced by the previous Republican administration, and Republicans had been key sponsors in both the Senate and House. Second, the problems that had originally attracted congressional interest—the outcomes/effectiveness problem and the Medi- care cost problem—remained. The rationale for activities to address these problems was no less strong in 1995 than in 1989.³²

In terms of private-sector allies, AHCPR had grown stronger over the years, the back surgeons notwithstanding. Jarrett Clinton had involved many professional organizations in the agency’s work on outcomes research and practice guidelines. In the months after his appointment as AHCPR director, Gaus had undertaken a strategic planning process that involved organizations such as the American Association of Health Plans (AAHP), American Hospital Association (AHA), American Medical Association (AMA), and other trade and professional organizations.³³ His premise had been that if the agency was to grow, it needed stakeholders who would speak for it. In 1995, however, the question was survival, not growth.

Important advocates. The multistep budget and appropriations process was the locus for the struggle, and many individuals and organizations helped. For example, the AHA, AAHP, and the Health Insurance Association of America (HIAA) lent their lobbying resources, both in testifying and in keeping track of breaking developments. The Association of American Medical Colleges (AAMC) passed a resolution in support of the agency, as did the AMA’s House of Delegates. The American Nurses Association put in helpful testimony. When Gaus was invited to debate Representative Johnson on a cable television station in North Carolina, Robert Graham of the American Academy of Physicians arranged for a North Carolina physician to substitute, with great effectiveness. The Paralyzed Veterans of American came to AHCPR’s defense because of the agency’s guidelines and outcomes work on topics such as pain control and urinary incontinence.

Members of AHCPR’s advisory council also helped, as did some the people who played key roles at the creation: Roper, who was now a senior vice-president at Prudential HealthCare, as well as now retired Senators Mitchell and Durenberger and Representative Gradison, who now headed the HIAA (it was the HIAA’s “Harry and Louise” commercials that been prominent in opposition to the Clinton health reform proposal).³⁴ The agency’s active supporters included parties whose interests had frequently clashed—such as AAHP president Karen Ignagni and AMA executive vice-president James Todd.

One of the agency’s most important advocates was Gail Wilensky, who had good relationships with many Republicans on Capitol Hill, having accepted dozens of requests from senators and representatives to speak in their districts about Medicare while she was HCFA administrator in the Bush administration. Notwithstanding her earlier critical views about AHCPR, she found much of the attack on the agency to be unfair or based on serious misconceptions. She spoke with members and staff on the most relevant committees, particularly in the House, and urged them “not to participate in killing the agency.”³⁵

Support on the Hill. Two House Republican staffers who had played a part in passage of the legislation that created the agency had influential positions in the new Congress. They were the starting point for the effort to increase key members’ understanding of the nature and importance of the agency’s work. One was Charles (Chip) Kahn, who had worked in the 1980s on the outcomes/effectiveness research agenda with both Senator Durenberger and Representative Gradison and was now staff director of the Ways and Means Health Subcommittee. This provided a linkage to Medicare and Medicare expenditures, which Larry Patton, AHCPR’s congressional liaison, believed to be crucial. The other was Tony McCann, who had worked as a Bush administration budget official on the FY 1990 effectiveness initiative and was now staff director of the House Appropriations Subcommittee on Labor and HHS. The chairs of these two subcommittees, Bill Thomas (R-CA) and John Porter (R-IL), became the agency’s two most determined supporters in the House. Close observers agree that without their efforts, the agency’s budget would probably have been reduced to zero in the House. Their support included letters to colleagues and advocacy in their committees and within the House leadership.

House of Representatives. The agency benefited from the generally favorable orientation toward research held not only by Representative Porter but also by Speaker Gingrich. The importance of the leadership’s support was particularly important in the lengthy and chaotic floor battle over the appropriations bill. Representative Johnson’s proposed amendment to take the agency’s budget to zero might have succeeded without the leadership’s intervention. Deals were being made to garner votes for the overall bill, and the agency’s supporters were in a position to negotiate. On the night of a crucial floor vote, Speaker Gingrich withdrew his support for zeroing out the agency, and Johnson withdrew his amendment.³⁶ He and his allies remained determined to impose big cuts on the agency’s budget, and the leadership and House agreed to a cut to $65 million in the House version of the appropriations bill. This last cut was illusory, however, because before the vote occurred, Representatives Thomas, Porter, and David Obey (D-WI), the ranking minority member of the Appropriations Committee, had negotiated an agreement with Appropriations Committee Chair Bob Livingstone (R-LA) under which the leadership would support the Senate’s budget number for the agency in the eventual House-Senate conference committee. The aforementioned handful of people who were able to make the agency’s case to members of the House Republican leadership prevented a greater disaster from occurring.

U.S. Senate. The agency faced a less difficult situation in the less partisan Senate. Several senators who had supported the agency’s creation, including Edward Kennedy (D-MA), Jay Rockefeller (D-WV), and Orrin Hatch (R-UT), were still in office. A letter drafted by Senators James Jeffords (R-VT) and Rockefeller and signed by ten senators (eight Senate Democrats, plus Senators Jeffords and William Cohen [R-ME]) was sent to Sen. Arlen Spector (R-PA), chair of the Senate Appropriations Subcommittee on Health. This letter called for the maintenance of FY 1995 funding for the agency. It observed that “there has been some misunderstanding about the role, purpose and mission of AHCPR” and argued that “it is essential to have a federal agency that works with the private sector to provide consumers with information to make informed choices, measure and improve the quality of care and improve the cost and effectiveness of our health care system.”

Even so, the Senate Appropriations Committee and the Senate itself approved a reduced budget figure for the agency ($127,310,000) that was similar to the amount that had been approved by the House Appropriations Committee. In the conference committee, the House leadership, as promised during the floor battle months before, supported the Senate number. On 26 April 1996, almost seven months into the fiscal year, President Clinton signed the appropriations bill that, after a final small across-the-board reduction, was only slightly less than the amount approved by the Senate. The agency had survived, but with a 21 percent budget cut.

Aftermath Of The Agency’s Near-Demise

AHCPR came under attack again in 1996 in the House of Representatives. Although the House voted to freeze the agency’s budget, the FY 1997 appropriations bill provided for $143,470,000, only $301,000 less than President Clinton had requested. Medicare trust fund monies were eliminated, but this was minor. The budget increase over FY 1996 did not indicate a new appreciation of the agency’s work but was attributable to the cost of the massive National Medical Expenditure Panel Survey. The agency’s struggles to gain increased budget were not over—indeed, the president’s budget requested only a $6 million increase the next year, of which Congress agreed to half. However, the question of AHCPR’s survival had been put to rest.

A change in leadership. The budget battle on Capitol Hill had been the focal point for much of Gaus’s work in 1995. The resulting budget meant that the agency could make almost no new grants in 1996. To Gaus’s further dismay, the agency failed to gain support from within the administration for substantial increases in the president’s budget proposals in subsequent years. Gaus concluded that new leadership was needed to move the agency forward, and he stepped down as director in early 1997. In a friendly transition, he was replaced by John Eisenberg.

Eisenberg brought new stature and a distinctive set of strengths to the job. He was a nationally known health services researcher who was chairman of the Department of Medicine and physician-in-chief at Georgetown University, an IOM member, and a former AHSR president. Having chaired the PPRC for several years, he had what he described as “trusting and friendly relationships with a number of key staff on both sides of the aisle.”³⁷ His being a physician gave the agency increased legitimacy on matters pertaining to clinical care. He was by consensus both brilliant and politically skilled. And he was extraordinarily well connected—having served as HHS Secretary Donna Shalala’s personal physician and as a consultant and confidant to Nancy-Ann Min, who oversaw the agency’s budget at OMB (and went on to head HCFA); and being a fellow Tennessean to Vice-President Al Gore (and a friend of Gore’s legal adviser, Charles Burson, from Memphis days) and Republican Senator Bill Frist (with whom he also shared Princeton roots), who became the agency’s key Senate supporter.³⁸ Eisenberg, who directed AHCPR and its successor, AHRQ, until his death from a brain tumor in 2002, had a transforming effect on the agency.

Changes in the agency. A full account of the Eisenberg era is beyond the scope of this paper.³⁹ Some of the important changes that took place were already under way before the attack on the agency. Eisenberg extended the major policy changes that Gaus had initiated, including the organization of research activities into topical centers and the focus on customers. Eisenberg saw three types of decisionmakers as the main customers for the agency’s work: people who make clinical decisions (patients, families, clinicians), those who make management and system-level decisions (purchasers and providers), and those who make public policy decisions at all levels of government. Eisenberg emphasized the development of partnerships with organizations of health care decisionmakers such as medical and nursing professional societies and AAHP, with which a National Guideline Clearinghouse was developed.

Some changes grew directly from Eisenberg’s stature and leadership. As concerns about health care quality and patient safety came to the fore in the late 1990s, AHRQ became the lead agency on quality, and Eisenberg became the operating chair of the Quality Interagency Coordination Task Force, which brought together all of the government agencies that deliver or purchase health care (Veterans Affairs, Defense, Office of Personnel Management, Medicare). Partnerships to improve health quality were developed between AHRQ and these agencies.

Some developments could be interpreted as reflecting lessons learned from the 1995 attack, but they mostly either had roots that antedated 1995 or are consistent with what a politically savvy director would do. Recognizing the importance of engaging in activities that are valued by those who directly or indirectly might affect the agency’s resources, the agency undertook energetic efforts to establish ongoing contacts and liaisons to learn what activities and types of information might be important, to whom, in setting priorities. A closely related activity was to make key constituencies aware of the agency’s activities. Dissemination in its various forms became a higher priority. Eisenberg’s bridge-building orientation was reflected in extensive consultative activities and speaking engagements before audiences ranging from Capitol Hill to the annual meeting of the North American Spinal Society, whose members had so recently worked for the agency’s demise.

The agency also moved away from activities that could generate negative political fallout. Its leadership, and that of HHS, drew two implications from the agency’s experience in 1995. The first was that having the word “policy” in its name invited misconceptions about the agency’s activities and risked identifying the agency with unpopular health policies of the current administration, whatever it might be. Many in Congress saw the conduct of research and the development of policy as distinct activities. The second was that the development and dissemination of practice guidelines would always carry a substantial risk of generating opposition from affected providers, who might, under the right circumstances, be able to jeopardize the agency’s funding in the appropriations process. In 1996, under Gaus, the agency replaced its guideline development and dissemination program with program to support external “evidence-based practice centers,” which would be responsible for compiling and organizing data to be used by other (mostly private-sector) organizations to develop practice guidelines. (Such a change had been recommended in the 1995 report by the PPRC, which Eisenberg chaired.) In the 1999 reauthorizing legislation, which was developed with the agency’s leadership and which changed its name to the Agency for Healthcare Research and Quality, the phrase “practice guidelines” disappeared.

Among those whom we interviewed, the abandonment of the practice guideline program is the most lamented change, because of the high quality of the agency’s guidelines, its unique ability to bring the right parties to the table, and its ability to bring both objectivity and the appearance of objectivity to the task of assessing evidence. However, as Eisenberg wrote in 2001, “The strategy of AHRQ partnering with professional groups and others to use evidence reports that we have sponsored to write guidelines is the way to get them written well, and it is a model more likely to succeed than the old AHCPR model. There may be those who regret our not writing guidelines any more. I am definitely not one of them.”⁴⁰

The agency’s budget grew under Eisenberg, reaching $300 million in FY 2002. A close working relationship was developed with Senator Frist and his staff. The 1999 reauthorization bill that changed the agency’s name to AHRQ passed unanimously in the Senate, and the House vote was 417–7. This was, in Eisenberg’s words, “a remarkable attestation that the dark days of a few years back were gone.”⁴¹ “Policy” had disappeared from its name, and “practice guidelines” were no longer in its mission. But as in 1989, when practical concerns about practice variations and costs provided impetus for creation of AHCPR, the 1999 legislation also embodied the hope and expectation that the agency would address problems of great current concern: in this case, quality and patient safety. Once again the agency was given a mandate that went beyond research.

Conclusions

We have described the creation of AHCPR as the product of an interrelated set of strategies—bureaucratic, marketing, and constituency building—used by the professional association of health services researchers to try to enhance the status of and support for HSR. We have argued that these strategies help account not only for the creation of the agency but also for its near-demise in 1995.

What overall conclusions could we draw about the effects of these strategies from the perspective of some thirteen years after their implementation led to the legislation that created the agency? It is difficult to parcel out the effect of the agency’s five years of leadership (1997–2001) by John Eisenberg, who brought a unique set of personal and professional qualities to the position. His extraordinary impact on the agency and the larger field of health services research is documented elsewhere. That impact alerts us to the difficulty of parsing the effects of a structure that was put in place, as with the 1989 legislation that created AHCPR, and the specific people who played key roles in how that structure was realized. Even so, the question remains about the effects of strategies to tie HSR to the engine that was outcomes and effectiveness research, to put it into a higher-level agency with new streams of funding, and to sell it on its usefulness in dealing with important concerns of policymakers or their constituents.

Effects of increased prominence. The increased prominence that resulted from agency status was a mixed blessing, as we have shown.⁴² In budgetary terms, the strategies were a qualified success. Going back to Exhibit 2, the jump in the funding level that began in FY 1990, when the agency was created, clearly carried through the 1990s, even with the 1995 cut. It seems unlikely that these funding levels could have been achieved without the 1990 change in baseline. However, the circumstances under which this baseline increase occurred—with the Bush administration recommending the $52 million effectiveness research initiative and Wennberg convincing key congressional leaders that a major outcomes research effort was needed—suggest that the creation of the new agency itself did not account for the shift in the HSR funding baseline. The 16 percent cut that President George W. Bush proposed in AHRQ’s 2003 budget perhaps confirms this.

Mixing research and policy. There is also room for debate about the benefits of the marriage of outcomes research and health services research. On the one hand, it is not implausible that the idea of Patient Outcomes Research Teams would have enjoyed more sustained support had the research initiative been located in NIH rather than in an agency that was steeped in traditional health services research and that was struggling for legitimacy. And the creation of the agency, with its heavily clinical agenda, arguably shifted the field of HSR away from traditionally important topics concerned with the organization and financing of health services. But there was no budgetary constituency for research on those topics.

The 1989 legislation blurred the line between research and the application of research to practice and policy, with AHCPR expected not just to conduct and support research but also to change medical practice and moderate costs. The goal of achieving cost containment was vested heavily in the agency’s mandate to develop and disseminate practice guidelines. An assessment of the outcomes research and practice guideline programs is largely outside the scope of this paper’s focus, but it is clear that the expectation that this small research agency would have measurable real-world effects on medical practice was a double-edged sword. On the one hand, it attracted the congressional support that led to the creation of the agency, the budgetary expansion, and all the rest. However, it posed two dangers. First was the risk of failing to meet expectations, even if those expectations were arguably naïve. Second was the fact that activities that could result in changes in the flow of dollars had the potential to create enemies for the agency. By the mid-1990s the practice guideline program had been abandoned, along with any explicit hopes that AHCPR’s work would generate measurable cost savings for Medicare.

It is also clear that being involved in the application of research to an administration’s policy development is a double-edged sword. It provides an opportunity to demonstrate the practical utilities of dollars spent on research and a research agency, but there is a risk of being seen as either insufficiently or overly responsive to the administration. With health policy being an arena of deep political divisions, Eisenberg said, “I have felt it critical that AHRQ be seen as a non-partisan agency that helps decisionmakers with evidence. I felt it critical that AHRQ not get wrapped up in partisan issues and that I personally be seen as one who could work well with both sides of debates.”43 Eisenberg was successful, but issues of major health reform did not arise on his watch.

Three dilemmas. AHCPR’s experience points to three dilemmas in the politics of health services research. First is what might be called, somewhat oddly, the research dilemma. Congressional interest in supporting research for its own sake is bounded. The support enjoyed by NIH shows a willingness to support basic research in the belief that the long-term result will be new ways of preventing or treating disease. The “research” aspect of HSR has proved to be difficult to sell, and the agency’s experience has not changed that. Funding is more forthcoming for a research agency if plausible arguments are made that it can do something to address real-world problems of current concern—that is, that it will do something that is not research. Second and closely related is the relevancy dilemma. If it is expected to produce work that will affect health care practice or health policy, an agency can attract enemies who are vested in the status quo.

Third is the constituency dilemma. The agency’s work is significant to many different parties—to policymakers whose interests may conflict, to private decisionmakers (providers, purchasers, patients) whose interests may conflict, and to researchers who have competing ideas about how the agency’s limited resources should be spent. Having multiple constituencies can be a source of strength for a federal agency, but conflicting constituencies can be a source of vulnerability. The agency’s ability to manage the relevance and constituency dilemmas will be key to its future.

This research was supported by grants from the Robert Wood Johnson Foundation and the Commonwealth Fund. The authors are grateful to all who provided information in interviews: Ross Arnett, Andrew Balas, Stephen Bandeian, Marc Berk, Brian Biles, Linda Bilheimer, Robert Blendon, Peter Bouxsein, Robert Brook, Peter Budetti, Carolyn Clancy, James Cleeman, Jarrett Clinton, Howard Cohen, Steve Crane, Mary Cummings, David Cutler, Helen Darling, Karen Davis, Gordon DeFriese, Linda Demlo, Nancy-Ann Min DeParle, Richard Deyo, Jack Ebeler, John Eisenberg, Arnold Epstein, Anne Esposito, Bill Evans, Dean Farley, Gary Filerman, Michael Fitzmaurice, Richard Frank, Deborah Freund, Cliff Gaus, Allison Giles, Sherry Glied, Donald Goldstone, Bill Gradison, Richard Greene, Greg Hampton, Michael Hash, David Helms, Ada Sue Hinshaw, Julie James, Chip Kahn, Robert Keller, David Kindig, Roz Lasker, Risa Lavizzo-Mourey, Jason Lee, Philip Lee, John Liu, Steve Long, Brian Luce, Hal Luft, Nicole Lurie, Tony McCann, John McManus, Barbara McNeil, Daniel Mendelson, Gregg Meyer, Kevin Moley, Alan Monheit, Judy Moore, Joseph Newhouse, Len Nichols, Larry Patton, Mary Ella Payne, Edward Perrin, Rep. John Porter, Sue Ranthum, Jack Rodgers, William Roper, Louis Rossiter, Thomas Scully, Pamela Short, Stephen Shortell, Lisa Simpson, Michael Stafford, Barbara Starfield, Earl Steinberg, Donald Steinwachs, Michael Stevens, Robert Streimer, Albert Siu, Richard Tarplin, Ken Thorpe, Sean Tunis, Margaret VanAmringe, Daniel Waldo, Marina Weiss, Norman Weissman, John Wennberg, Gail Wilensky, Christine Williams, and Donald Wilson. Because many asked not to be quoted and because some information was provided by multiple sources, not all information is specifically attributed.

NOTES

1. Institute of Medicine, Health Services Research: Work Force and Educational Issues (Washington: National Academies Press, 1995), 1. The full IOM definition called health services research a basic and applied field that “examines the use, costs, quality, accessibility, delivery, organization, financing, and outcomes of health care services to increase knowledge and understanding of the structure, processes, and effects of health services for individuals and populations.”
2. Several other federal agencies, including the Centers for Medicare and Medicaid Services (formerly HCFA), the Centers for Disease Control and Prevention, the Department of Veterans Affairs, and NIH, conduct or support some health services research. In none of these is such research central to the agency’s mission.
3. The HSR agencies have occupied space in various larger governmental and leased buildings in Bethesda and Rockville.
4. This account is based substantially on the interviews conducted by one of the authors (Brad Gray) in 1991 with the leadership and staff of AHSR, former directors of NCHSR, and congressional staffers. A list of interviewees appears in an earlier paper based on those interviews: B.H. Gray, “The Legislative Battle over Health Services Research,” Health Affairs (Winter 1992): 38–66.
5. A list appears in the acknowledgments of this paper.
6. Gray, “The Legislative Battle.”
7. W. Roper et al., “Effectiveness in Health Care: An Initiative to Evaluate and Improve Medical Practice,” New England Journal of Medicine 319, no. 18 (1988): 1197–1202.
8. C. Maklan, R. Green, and M. Cummings, “Methodological Challenges and Innovations in Patient Outcomes Research,” Medical Care 37, no. 7 Suppl. (1994): JS13–JS21.
9. In an interview, Peter Budetti, who oversaw the drafting of the original legislation on the House Energy and Commerce Committee staff, explained that “policy” was in the agency’s name because it belonged with “research” in the academic research center from which he came. It explained the orientation of the research. He also noted that the shift from the office of the assistant secretary for health to a full PHS agency had represented a significant distancing of the agency from the policy development process.
10. Interview with Stephen Bandeian, 4 December 1999.
11. Interview with Donald Goldstone, 14 December 1999. The detailing of personnel across federal agencies was not unusual, but this particular arrangement was new. Personnel from several other agencies (HCFA, HHS Assistant Secretary for Planning and Evaluation, Treasury, Labor) were also brought into the policy development process at the White House, but Bandeian notes (personal communication, 5 October 2000) that AHCPR played a “particularly important role in the quantitative work that surrounded the development of health reform proposals.”
12. Bandeian interview, 4 December 1999.
13. Goldstone interview; and interview with Dean Farley, 28 December 1999.
14. Interview with Gail Wilensky, 29 November 1999.
15. Interview with Jarrett Clinton, 5 October 1999.
16. The task force was legally dissolved in May 1993—four months after its creation. Most task force members had completed their work several weeks earlier.
17. L. Nichols, “Health System Reform Debates and Medical Expenditure Surveys,” in Informing Health Care Policy: The Dynamics of Medical Expenditure and Insurance Surveys, 1977–1996, ed. A. Monheit, R. Wilson, and R. Arnett (San Francisco: Jossey-Bass, 1999), 213–232.
18. Interview with Len Nichols, December 1999.
19. Interview with Dean Farley, 28 December 1999.
20. Interview with Philip Lee, 7 September 1999.
21. U.S. General Accounting Office, Practice Guidelines: Overview of Agency for Health Care Policy and Research Efforts (Washington: GAO, 1995); as cited in Sarah F. Jaggar, U.S. General Accounting Office, testimony before the Ways and Means Subcommittee on Health, U.S. House of Representatives, 25 July 1995.
22. Physician Payment Review Commission, Annual Report to Congress (Washington: PPRC, 1995), 371–398.
23. U.S. Congress, Office of Technology Assessment, Identifying Health Technologies That Work: Searching for Evidence, Pub. no. OTA-H-608 (Washington: U.S. Government Printing Office, September 1994).
24. Medicine and Health Perspectives, 17 July 1995.
25. Congressional Record, 3 August 1995.
26. Ibid.
27. In his 31 July 1995 “Dear Colleague” letter, Representative Johnson mocked the agency for publishing studies concluding that cardiologists know more about heart attack treatments than primary care doctors; that the doctor-patient relationship affects whether patients sue for malpractice; and that Medicaid patients prefer AIDS specialists to general practitioners after an AIDS diagnosis. He often mentioned the study comparing cardiologists to primary care physicians when meeting with seniors’ groups in his Texas House district as a prime example of government waste.
28. Medicine and Health, 17 July 1995.
29. Interview with Earl Steinberg, 19 January 2000.
30. J. Turner et al., “Patient Outcomes after Lumbar Spinal Fusions,” Journal of the American Medical Association (19 August 1992): 907–911.
31. A. White et al., “Letter to the Editor,” Spine 19, no. 1 (1994): 109–110.
32. One Republican staffer recalls going with his boss, the chair of a committee that had passed the original authorizing legislation for the agency, to visit the House Budget Committee chair and his staff when they were proposing a zero budget for the agency. He paraphrased the argument they made as follows: “Are you guys nuts? Republicans created this. It is allowing us to figure out what we are spending money on. [Zeroing it out] is just crazy.”
33. The American Association of Health Plans (AAHP) was formerly known as the Group Health Association of America (GHAA).
34. On 11 August Mitchell, Durenberger, and Gradison issued a paper that was later introduced on the floor of the Senate by Sen. Tom Daschle (D-SD), titled, “The Agency for Health Care Policy and Research: A Beacon for Policymakers” (Health News Daily, 14 August 1995). They argued that AHCPR’s research had improved the quality of health care in the United States, helped maximize the value of government spending on medical care, and represented “an extremely sound investment for American taxpayers.”
35. Wilensky interview. She further noted, “I had never lobbied on behalf of a cause, except to keep this agency from being killed. And I was thinking that these [the agency] were the people who had stiffed me on legitimate [data] requests.”
36. We were unable to interview Johnson about his reason for withdrawing the amendment. We heard two speculations from interested observers at the agency. One was that he saw that he lacked the votes. The other was that Johnson, an ex–prisoner of war, was successfully lobbied by a disabled veterans group, which supported the agency.
37. John Eisenberg, personal communication, 29 January 2001.
38. Eisenberg and Frist also shared the same mentors in college, Anne and Herman Somers, whom Eisenberg credits with stimulating his interest in “scholarly exploration of health policy issues.” Eisenberg, personal communication, 8 February 2001.
39. See C.M. Clancy and H. Burstin, “John M. Eisenberg: Telling the Story,” Medical Care (November 2002): 1012–1013; W. Tierney et al., “A Remarkable Legacy of Science: Review of AHRQ-Funded Articles Published in Medical Care during John Eisenberg’s Directorship of AHRQ,” Medical Care (November 2002): 1003–1011; N. Foster et al., “In Memoriam: John M. Eisenberg, MD, MBA,” International Journal of Quality in Health Care (June 2002): 169–172; N. Foster, C. Clancy, and G. Meyer, “In Appreciation: John Eisenberg of the Agency for Healthcare Research and Quality,” Medical Decision Making (July/August 2002): 373–374; and G.D. Lundberg, “Obituary for John M. Eisenberg, America’s Top Doc,” MedGenMed, 15 March 2002, www.medscape.com/viewarticle/429997_print (20 May 2003).
40. Eisenberg, personal communication, 8 February 2001.
41. Ibid.
42. Notably, the Office of Assistant Secretary for Health, where AHCPR’s predecessor organization (NCHSR) had been located, was eliminated during the 1995 budget battle.
43. Eisenberg, personal communication, 29 January 2001.

Brad Gray is director of the Division of Health and Science Policy, New York Academy of Medicine. Michael Gusmano is a senior health analyst and associate director at the World Cities Project, International Longevity Center—USA. Sara Collins is a senior program officer at the Commonwealth Fund. All are in New York City.

Read related perspectives by Jack Wennberg, Clifton R. Gaus, and Carolyn M. Clancy

©2003 Project HOPE–The People-to-People Health Foundation, Inc.