Scientific Opportunities and Public Needs: Improving Priority Setting and Public Input at the National Institutes of Health (1998)

The fourth charge to the committee was to "assess the impact of statutory directives on research funding decisions." The committee broadened the charge to address the role of Congress in setting NIH research priorities through various means, including the use of report language, which is often used instead of statutory language to communicate congressional priorities to NIH.

The members of Congress who serve on one of the four main committees affecting NIH—the House and Senate authorization subcommittees and the House and Senate appropriations subcommittees—are more familiar with how science works than Congress as a whole and are aware of the problems caused by earmarking or other overprescriptive directives that reduce NIH's flexibility to set priorities informed by scientific opportunity as well as need. The health committees have been less prescriptive in recent years, but those who want Congress to specify funding of research on particular diseases can go, and recently have gone, to other places in the system. In recent years, for example, there have been earmarks for breast cancer research and prostate cancer research in defense legislation and an appropriation for diabetes research in the Balanced Budget Act.

Legislators on the NIH-related subcommittees also often have their own ideas about how NIH might better carry out its mission. Some become interested for one reason or another in a particular disease or medical condition. Some become interested in the health of particular groups, such as women, members

of minority groups, elderly people, and children. Some worry that NIH is not paying enough attention to or investing adequately in particular kinds of research—the state of clinical research is currently a big issue—or crosscutting activities (such as prevention or nutrition) or important fields, technologies, and approaches (such as bioengineering, medical imaging, and alternative medicine).

For these reasons, authorization bills and appropriations committee reports have until recently contained some fairly specific directives. These have included mandates to establish particular offices or centers in institutes or in the Office of the Director, the use of certain mechanisms (such as centers, clinical trials, or specific requests for applications), or the use of set-asides of specified amounts of funding for a particular activity.

Many Congressional leaders would prefer to rely on NIH to allocate funds among research programs, but first, they are mindful of their important oversight responsibility. NIH is part of a very important public function and has a very large budget. Second, members of Congress are constantly approached by individuals and groups about the impacts of terrible diseases, and they want to show that they are trying to help. At the same time, many believe that NIH should be making the decisions about the allocation of funding among research areas, because they are aware that the opportunity for advances varies from one problem to another. Thus, the legislators want to be able to refer people to an identifiable process at NIH and be assured that NIH is in fact listening and taking what they hear into account, that is, that the inputs of all interests have been fairly and appropriately taken into account in program and resource allocation decision making.

The existence of such a process is not clear to all members. Members of Congress are constantly exposed to disease-specific interest groups that have statistics on disease burdens and lists of scientific opportunities and argue that they are neglected in terms of the amount of funding for research on their disease of interest as reported by NIH. They wonder: How does NIH know that an area of research is emerging and when and how to respond? How are measures of disease burden and costs taken into account and how does the public have input? Does the NIH priority-setting process need to be changed, or is it working well and the need is just for better communication?

As noted, NIH operations, structure, and funding are affected most by four congressional committees: the authorization and appropriations committees in each house. The authorization committees are the House Committee on Commerce (Subcommittee on Health and Environment) and the Senate Committee on Labor and Human Resources (Subcommittee on Public Health and Safety).

The appropriations committees are the House Appropriations Committee (Subcommittee on Labor, Health and Human Services, and Education) and the Senate Appropriations Committee (Subcommittee on Labor, Health and Human Services, and Education).

Historically, the appropriations committees have played a major role in NIH priority setting. They usually appropriate millions more than the president's budget requests, which makes it much easier to influence, if not specify, the use of the appropriation than is the case with appropriations for other agencies in which Congress appropriates about the same amount or a little more or less than the amount requested by the administration.

Appropriations committees have a number of ways of communicating intent about executive agency priority setting, including bill language, report language, statements made on the House or Senate floor or at hearings, and informal contacts between committee members and staff and agency officials. All these methods are used in the case of NIH.

Statutory Language The language used in NIH appropriations laws is usually general. Typically they read: "for carrying out section 301 and title IV of the Public Health Service Act with respect to cancer [or diabetes and digestive and kidney diseases, or allergy and infectious diseases, etc.], $x." However, statutory language is always an available route if Congress feels strongly about something and NIH has not responded to concerns expressed in report language.

For example, in the first session of the 105th Congress, Congress took the unusual step of including as an amendment to the FY 1998 appropriations bill a revision of the Public Health Service (PHS) Act of 1944 that authorized a program of research on a specific disease, Parkinson's disease (the Udall Bill). Although the appropriations committees did not give the amendment a specific appropriation, the National Institute of Neurological Disorders and Stroke is responding using the basic authority contained in the PHS Act (section 301). In November 1997, it issued an RFA for Parkinson's Disease Research Centers of Excellence.

Report Language Report language is the normal means of expressing congressional intent. Although directives in report language do not have the force of law, NIH takes them quite seriously. For one thing, if NIH does not respond, the directive could escalate into statutory language in a subsequent session of Congress. For another, for both Congress and NIH report language is a more flexible means of communication than bill language. The problem for NIH

comes when the language is too detailed and ends up forcing NIH to fund more research than there are good projects to support or to use particular mechanisms.

Examination of the FY 1993 and FY 1998 reports of the appropriations committees reveals examples of many kinds of directives, ranging from "the Committee is encouraged by the Institute's continued support of research on x" to "the Committee has provided $25 million for y disease research, $4 million more than requested and $5 million more than last year's level." The typical item in report language "encourages" or "urges" that NIH or the institute continue to make x a high priority or to expand support. There are many types of directives, however. The list includes the following:

• Specify a dollar amount for a particular program (Office of Alternative Medicine) or particular area of research (e.g., pediatric research initiative or Alzheimer's disease).
• Specify a mechanism, for example, saying that the committee has put in sufficient funds (or a specific funding amount) to support a new research center or a specific number of research centers addressing a particular disease, that NIH should issue an RFA, or that NIH should begin planning clinical trials.
• Call for a scientific conference on opportunities in a particular field or disease to be held and a report submitted before the next year's appropriations hearings.
• Require that a formal report be made on a topic of interest. This is partly done to express interest in that topic and is often done to look for suggestions for research opportunities that Congress can urge NIH to fund (if Congress has not done so already).
• Ask for a 5- or 10-year research plan to be submitted to the committee by a certain date and require consultation with the affected outside groups.
• Ask the NIH director to assign a clear lead agency (e.g., NIAID and Lyme disease in the FY 1993 Senate report).
• Suggest that NIH consider setting up a separate study section for a disease on which the committee wants to see more emphasis.
• Suggest a type of national advisory council member (the Senate committee suggested, for example, that the National Eye Institute add an optometrist).
• Direct that an intramural program be expanded or formalized.

The typical directive—"the Committee encourages the institute to expand its support of x research"—is usually not problematic because, given NIH's normal rate of budget growth, research on x is going to grow anyway, without any special steps being taken. In many cases, the institute can simply document in its next year's congressional justification budget and in testimony what would have happened anyway. That is not always the case, however. In the late 1980s and early 1990s, NIH budget growth slowed and the number of earmarks specifying amounts of spending on particular disease programs grew. The earmarks

sometimes added up to more than the budget increases, and so some institutes had to find ways to pay for them, which in turned squeezed other programs. In FY 1993, for example, the appropriations committees earmarked additional funding for research on breast, ovarian, cervical, and prostate cancers that was more than the total increase of $28 million that they gave to NCI for cancer research. The earmarks added up to $77 million, which NCI offset by cutting basic research and research on leukemia; non-Hodgkin's lymphoma; cancers of the colon, bladder, kidney, and brain; public information and education; and chemoprevention. Other examples include an earmark of $24 million for pediatric AIDS clinical trials in the FY 1991 House report. In that case, the conference committee agreed on a final overall amount for NIAID that was $34 million less than the House recommendation but specifically directed that the increase for pediatric AIDS clinical trials stay at $24 million, which resulted in cuts in vaccine development and other areas of AIDS research.

Trends in Appropriations Reports The numbers of items in the reports accompanying the appropriations bills increased between the reports for the 2 years that the committee examined in detail (1993 and 1998), but the items included far fewer earmarks for specific levels of funding for particular programs or other detailed directives in the latter year (see Table 5-1). The Senate, for example, specified $133 million in FY 1998, whereas it specified $785 million in FY 1993.

Historically, NIH has benefited from having a permanent authorization (section 301 of the PHS Act of 1944 gives the secretary of DHHS broad permanent authority to conduct and sponsor research) and from the practice, begun in 1948, of authorizing ''such sums as may be necessary." Historically, the main action in the authorization process was whether or not to create new institutes. The pressures that this puts on Congress resulted in the 1984 study by an IOM committee of the organizational structure of NIH (Institute of Medicine, 1984).

Since then the number of institutes and other major units (centers and divisions) has increased to 21. Beginning with the example of the AIDS research buildup in the 1980s, the focus among disease-specific interest groups has shifted to tracking the amount of funding across NIH for their particular interest and trying to impose more crosscutting coordination of programs, for example, by establishing a tracking and coordinating office in the Office of the Director of NIH.

The 1993 Reauthorization The most recent reauthorization was in 1993, and a reiteration of its provisions indicates the trends (U.S. Congress, 1993):

• Reauthorized certain expiring authorities of NIH.
• Mandated establishment of the Office of Research Integrity in DHHS.
• Lifted the moratorium on human fetal tissue transplantation research.
• Mandated inclusion of women and minorities in clinical research protocols.
• Established the Office of Alternative Medicine in the Office of the Director.
• Established the Office of Research on Women's Health in the Office of the Director.
• Established the Office of Biobehavioral and Social Sciences Research in the Office of the Director.
• Put the National Center for Human Genome Research on a statutory basis.
• Mandated establishment of an intramural laboratory and clinical research program on obstetrics and gynecology within the National Institute of Child Health and Human Development (NICHD).
• Established the National Center on Sleep Disorders Research in the National Heart, Lung, and Blood Institute (NHLBI).
• Codified in statute the establishment of the Office of AIDS Research and strengthened and expanded its authorities, including the authority to receive all AIDS appropriations and disburse them to the institutes and centers.
• Authorized establishment of an NIH director's discretionary fund.
• Provided the NIH director with extramural construction funds (with a set-aside of $5 million for centers of excellence).
• Mandated establishment of the Institutional Development Award Program.
• Required NCI to conduct the Long Island breast cancer study.
• Authorized scholarship and loan repayment programs for individuals from disadvantaged backgrounds.
• Elevated the nursing center to an institute, the National Institute for Nursing Research.
• Elevated the NHLBI Division of Blood Research to a center.
• Provided a number of other new NIH authorities and directives.

The 1996 Senate Reauthorization Bill The Senate (but not the House) passed a reauthorization bill in 1996. The pressure to mandate programs and organizational units for particular purposes was intense. The subcommittee tried to avoid

specificity. According to its report on the NIH Revitalization Act of 1996 (U.S. Senate, 1996):

Despite the subcommittee's intentions, the report and accompanying bill in fact contained a number of new set-asides and other provisions that affected NIH priority setting by mandating programs for and directing resources toward specific diseases. Most of them were added as amendments. The bill did the following:

• Elevated the genome center to institute status, the National Human Genome Research Institute, and mandated a set-aside of at least 5 percent of the new institute's extramural research funds for research on ethical, legal, and social issues.
• Included provisions intended to increase support for clinical or patient-focused research by NIH.
• Established a Parkinson's disease research and training program with up to 10 core grants for Morris K. Udall Centers for Research on Parkinson's disease, with $80 million authorized for the first year and such sums as may be necessary for the next 2 years (added by amendment).
• Increased the authorization for diabetes research by 25 percent over the next 3 years (added by amendment).
• Created an office of pediatric research in the Office of the Director "to increase pediatric biomedical research," with increased funding to be allocated by the director after consultation with external advisers and NICHD (the bill authorized $50 million a year for the next 3 years) (added by amendment).
• Required NIH to report how it is going to implement an earlier report to Congress, Support for Bioengineering Research.
• Codified the Office of Rare Disease in the director's office.
• "Encouraged" the NIH director to establish a Pain Research Consortium involving all NIH units involved in pain research (in lieu of an amendment to establish a National Center for Pain Research at NIH and six regional centers for pain research).

Current Situation Some committee members (e.g., Senator Dan Coats [U.S. Congress, 1997]) expressed strong reservations about the wisdom of legislating in such detail. One result was the series of hearings that the Senate subcommit-

tee held on priority setting at NIH, beginning on May 1, 1997, in an effort to reach an understanding among members of Congress and between Congress and NIH on the desirability of letting NIH decide on most if not all allocations below the institute level (see discussion in Chapter 1).

Although most witnesses in the May 1, 1997, hearing on NIH priority setting and several that followed favored letting NIH set priorities and opposed detailed directives from Congress, the authorization committees face intense pressures to use the reauthorization bill to accomplish specific goals. Current proposals include

• a Center for Alternative Medicine,
• codification of the General Clinical Research Centers in the PHS Act,
• a new Institute on Biomedical Imaging,
• a Center on Bioengineering in NHLBI,
• a pediatrics initiative,
• a proposal to use $2.5 billion from a tobacco settlement and earmark one-third for behavioral research, and
• initiatives left over from the 1996 reauthorization, including the National Center for Pain Research and a diabetes program.

The U.S. Congress has always played an active role in setting research priorities at NIH. For example, it periodically decides to create new institutes or other organizational entities and routinely allocates funds among NIH institutes and centers and other units through the annual appropriations process. It may mandate the amount of spending or specify mechanisms of research on particular areas or diseases (e.g., mandating the establishment of centers or the issuance of RFAs) if it concludes that NIH is neglecting opportunities or needs.

Congress, of course, has the authority, which it has felt compelled to use from time to time, to intervene in NIH's affairs, for example, by mandating the creation of a center or office or by specifying a funding level for research on a particular disease. The committee agrees, however, with the sentiments of many legislators that Congress should rely as much as possible on NIH's own priority-setting processes because Congress generally lacks the expertise to judge the degree of scientific opportunity. The committee believes that implementation of its recommendations would go a long way toward ensuring a process that interest groups will find to be open and fair and, hence, would reduce the level of public appeals to Congress.

It appears to the committee that in general the Congress has eased up on its use of earmarks and other restrictive directives in recent years. Although the number of congressional directives has increased, they are much less specific. For

example, the number of earmarks for funding for research related to particular diseases has declined significantly. Congress's use of report language to convey concerns and priorities is positive, but the specification of budget amounts or specific mechanisms for funding (e.g., mandating that a certain number of centers be established or that a specific RFA be issued) should be done only as a last resort, because these approaches often have unintended effects. For example, in testimony to the committee, an advocate for AIDS research reported that earmarked funds, provided in response to the requests of advocates, had had unexpected negative impacts on other, at least equally important areas of AIDS research. Earmarking of funds for specific diseases also pits disease-specific interest groups against one another.

As science changes, however, and new health problems emerge, NIH must shift its priorities and make organizational changes to adapt. Such changes are taking place all the time. The establishment of the Center for Human Genome Research and its later elevation to institute status are examples. The creation of the Office of AIDS Research is another example. In these and other cases, if commitments of substantial new funding or major organizational changes are involved, Congress invariably becomes involved through the reauthorization process. Groups and organizations that believe that their interests will be helped by the creation of a new NIH unit (by increasing visibility and funding) will ask Congress to authorize such funding. Indeed, as mentioned above, there are a number of current proposals to mandate new organizational entities and levels of funding for specific diseases in pending NIH reauthorization legislation.

What guidelines can the committee offer Congress to help it determine whether to specify a new program, center, or institute? In 1984, an IOM committee recommended criteria to be used to evaluate proposals for new institutes or other major organizational changes (see Box 5-1).

Perhaps more importantly, the 1984 IOM committee also recommended that NIH rely on a range of activities short of establishing new institutes to respond to health needs and opportunities. That committee suggested that NIH have "a continuum of possible responses to needs and opportunities it identifies, matched with the magnitude and urgency of the demand" (Institute of Medicine, 1984:19):

The situation has changed since 1984. There are still calls for new institutes and centers. Today, however, the main focus of interested groups is on having Congress mandate NIH-wide programs or funding levels, or both, for specific diseases or other activities.

The present committee finds that the approach of the 1984 committee is still a good one. However, it sees the need to elaborate it to address other demands or responses beyond the creation of new NIH organizational entities. Additional responses that NIH could take short of congressional action include inclusion of an area in the NIH Director's Areas of Research Emphasis, designation of lead institutes, and establishment of a coordinating office in the Office of the Director of NIH.

Recommendation 10. The U.S. Congress should use its authority to mandate specific research programs, establish levels of funding for them, and implement new organizational entities only when other approaches have proven inadequate. NIH should provide Congress with analyses of how NIH is responding to requests for such major changes and whether these requests can be addressed within existing mechanisms.

NIH has an obligation to engage in periodic reviews of its organizational structure and planning and budgeting systems and to explain the results to Congress and the public, if it is to manage its own priority setting rather than react to directives from Congress trying to respond to requests from disease-specific interest groups. Such reviews would result in NIH making organizational and management changes, including the creation or disestablishment of institutes and centers and the reorganization of existing ones, or requesting new or expanded authorities from Congress, when needed. The reviews should also include evaluations of past organizational and management changes to see if they have been successful. This flexibility would help NIH remain organizationally dynamic and would help it incorporate changing scientific knowledge and meet health needs in a well-considered and planned way.

Recommendation 11. The director of NIH should periodically review and report on the organizational structure of NIH, in light of changes in science and the health needs of the public.

Through the appropriations process, Congress directed NIH to reduce the budget for research management and support by 7.5 percent in FY 1996 and did not allow increases in FY 1997 or FY 1998 (although activities designated as related to public health education were exempted from the cap in FY 1998). Those cuts came after reductions had already been made in response to the administration's Reinventing Government initiative.

Congress's intent was to reduce administrative overhead. However, as NIH is currently organized, research management and support includes a number of important program-related functions. It includes, for example, the personnel and other expenses (e.g., travel) of reviewing extramural research proposals and

managing the grants that are subsequently funded by NIH (it does not include the intramural research program). The extramural grant program is the largest and fastest-growing part of NIH's research effort. Research management and support also includes the capacities that need to be expanded to improve research priority-setting activities at NIH, such as new or expanded Offices of Public Liaison, the new Director's Council of Public Representatives in the Office of the Director, increased consumer participation in all NIH advisory bodies, and improved collection and analysis of disease-related data.

Resources invested in these underfunded functions not only should help NIH to fulfill its mission of improving the nation's health but should also improve the effectiveness of public oversight of its activities, thus enabling Congress and interest groups to observe and participate in a process that is more transparent and more satisfactory. This in turn may catalyze a change in which NIH and the consumers of health research work together rather than against each other and in which Congress lets NIH (informed by stronger public input) set research priorities.

Recommendation 12. Congress should adjust the levels of funding for research management and support so that NIH can implement improvements in the priority-setting process, including stronger analytical, planning, and public interface capacities.

Any additional resources needed to implement this recommendation do not necessarily have to affect the total amount appropriated to NIH. The recommendation is meant to affect the cap on research management and support funding within NIH budgets, if Congress elects to impose caps in future appropriations as it has in recent years.