Vol. 333 No. 12 SPECIAL REPORT SPECIAL REPORT 81 I SHATTUCK LECTURE - BIOMEDICAL RESEARCH ENTERS THE STEADY STATE HAROLD VARMUS, M.D. Two and a half years ago, when I was still a pri- vate citizen working at the University of California, San Francisco, my colleagues, Mike Bishop and Marc Kirschner, and I offered advice to our new President in Science magazine.' The first of our 11 recommenda- tions was to increase funding for the National Institutes of Health (NIH) by 15 percent per year, in order to double the NIH budget in five years and restore the success rate for grant applications to at least 30 per- cent. A year later, just after I had been called to Wash- ington to work for that President as &rector of the NIH, I had to admit, in a speech to the American Soci- ety of Cell Biology, that "our proposal to double the budget by fiscal year 1998 is simply not real is ti^."^ By then, I was already reconciled to a more modest goal: "to stay ahead of inflation [for] the next few years." Now, little more than a year later, we are faced with a new Congress intent on a balanced national budget at all costs. As a result, I have been trying to stave off cuts in the NIH budget - some nearly as large as the in- creases we proposed not so long ago - that would re- duce our buying power by about one third early in the next century. How can we explain such reversals of mood and for- tune? From a historical perspective, they reflect three stages in the development of biomedical research: a past that was expansive and successful, a present torn by the demands of adjusting to an end to growth, and an uncertain future that provokes anxiety and some- times despair. The rapid changes also embody three concurrent but divergent themes in our enterprise: the exhilarating, sustained effects of a wave of scientific achievement made possible by nearly 50 years of con- tinually increasing investment; the sobering effects of a more recent need to adapt to budget deficits, the na- tional debt, and public suspicion of government; and the alarming effects of budget-balancing efforts that would send biomedical research (and many other en- deavors that benefit from federal support) into a rapid decline. THE HISTORICAL CONTEXT To think constructively about these divergent forces, we must try to put the current problem in a historical context. The golden era of the NIH lasted over 40 years, beginning just after World War 11 and ending with the downfall of the Soviet empire. At the end of World War 11, scientists were regarded as heroes who had provided many of the factors that contributed to Presented as the 105th Shattuck Lecture to the Annual Meeting of the Massa- Address reprint requests to Dr. Varmus at the National Institutes of Health, chusetts Medical Society, Boston, May 20, 1995. Bldg. 1, Rrn 126, 1 Center Dr., MSC 0148, Bethesda, MD 20892-0148. victory: chloroquine and synthetic quinine to combat malaria, penicillin for infected wounds, and, of course, the atomic bomb. The passage of the GI Bill ensured that colleges and universities would prosper, that much larger numbers of people would study beyond high school, and that education would be valued. The econ- omy was thriving, the country was optimistic, and the federal government was viewed as the right place for leadership. No wonder, then, that the NIH was encouraged to expand from a small group of laboratory buildmgs in Bethesda, Maryland, dedicated by President Franklin Roosevelt just before the war, into a much larger or- ganization that mushroomed on the NIH campus and funded research throughout the country, at universi- ties, medical centers, and private research institutions. Prodded by public enthusiasm for biomedical science, as enunciated by a few famous spokespersons (congres- sional leaders, such as John Fogarty, Lister Hill, War- ren Magnuson, and William Natcher; the philanthro- pist Mary Lasker; NIH Director James Shannon; and others), the Congress responded with new funds and new health institutes (Table 1). The end of the Cold War in 1989 found the country in a mood very dfferent from that of 1945. The out- come of this war was seen more as a defeat of the ill- designed economic and political system of our oppo- nents than as a triumph of our own values. Scientists, even those whose research had been handsomely sup- ported by expenditures for defense, were not consid- ered major contributors to our success with the Soviets; after all, nuclear deterrence was based largely on scien- tific discoveries from an earlier era. Expensive missile- building technology was viewed this time as the source of other woes - most obviously, a national debt that had grown to an unimaginable size in the 1980s. Far from being an economic stimulus that might benefit Table 1. Congressional Appropriations for the NIH from 1950 to 1995. TOTAL RSUL YEAR ($MILLION) COMPONENT$ APPROPRIAnON No. OF NIH 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 43 67 332 713 1,061 2,093 3,429 5,149 7,576 11,306 6 9 9 12 19 19 20 20 23 26 812 THE NEW ENGLAND JOURNAL OF MEDICINE Sept. 21, 1995 peaceful enterprises, the end of the Cold War was ac- companied by an increasingly desperate economic situ- ation. At the same time, other research institutions were showing signs of stress and loss of direction: col- leges and universities, some overbuilt and undersub- scribed, were having trouble balancing their books, and some large national laboratories were without the clear mission or the review processes needed to maintain a standard of excellence. These factors contributed to the wave of sentiment against the federal government that was dramatically expressed in the recent congressional elections. BENEFITS OF FOUR DECADES OF NIH GROWTH Perhaps it is not surprising that biomedical scientists were among the last people to understand that we had come to the end of an era. When the social and politi- cal themes of the 1990s first emerged, they seemed de- tached from the prosperous condition of biomedical science. Forty-five years of steadily increasing financial support had produced an enviable record of scientific accomplishment and remarkable improvements in the nation's health (Table 2). This partial list of the chang- es wrought by investments in biomedical research - by the NIH and its many partners, here and abroad - is stunning. These advances also contributed to the na- tion's economy, through the growth of the pharmaceu- tical, biotechnology, and laboratory-supply industries, and encouraged many of our brightest students to train for careers in biomedical research. The support provid- ed to the universities to sustain research efforts and to train new investigators stimulated their growth and improved their quality. By all available measures, the United States assumed world leadership in medical re- search. CHANGES IN FEDERAL SUPPORT Still, the first sign that all was not well was apparent as early as 1990. That year, a surprisingly low success Table 2. Some of the Major Advances in the Health Sciences in the Past 50 Years. - Vaccines against poliovirus, hepatitis B virus, and many Penicillin and many other antibiotic agents. Recommendations for health-promoting diet and life- style, including simple means to lower the incidence of heart disease. ciencies. other infectious agents. Replacements for many hormonal and vitamin defi- New methods of contraception. Tests to protect the blood supply from hepatitis B and C viruses and the human immunodeficiency virus. New surgical treatments, including organ transplanta- tion and implantation of pacemakers and artificial joints. cancers. Effective therapies for certain leukemias and other Drugs effective against mental diseases New treatments, such as the use of blood-cell growth factors, developed from recombinant-DNA tech- niques. Methods of in vitro fertilization. Genetic testing for many inherited diseases. Table 3. Changes in the NIH Budget from 1985 to 1995. FISCAL YW\R CHANGE IN BUOGEI (%)* 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 199s Cumulative increase - -0.7 +11.6 + 2.4 +4.4 +2.1 1-3.4 1-4.0 -0.2 +1.9 -0.6 +31.4 `Calculated on the basis of constant dollars rate for NIH grant applications triggered many news stories, congressional concern, a symposium at the Na- tional Academy of Sciences about the funding of young scientist^,^ and more political engagement by my col- leagues than I had ever seen before. In fact, this initial sign of the crisis now at hand was partly attributable to a technicality: an extension of the average term of NIH grants had increased the proportion of the agency's funds that were already committed to multiyear grants. As a result, even a slight decrease in the rate of growth had a large adverse effect on the success rate for new applications. But over the next few years, even with a return to a more acceptable average term for awards, the success rates did not improve substantially, mainly because budgets did not increase in constant dollars, as they had in earlier times (Table 3). The difficulty of obtaining grant support has had complex and sometimes divisive consequences: criti- cism of the peer-review system and scrutiny of the sys- tem for inequities; demoralized and dwindling ranks of clinical researchers, who also suffer from the loss of health care revenues; a generation of disillusioned graduate students and postdoctoral fellows, who say that the future is not what they were promised; and a younger generation of skeptical high-school and college students, who see a scientific career as dauntingly com- petitive. RECOGNIZING THE STEADY STATE These symptoms of dis-ease in the biomedical re- search community have intensified during the five years since the first signs were noted. It is now possible to make a diagnosis: the research enterprise is undergo- ing a painful transition from an era of growth to an era of steady-state activity. In the steady state, new grants can be funded only when old grants expire, new faculty can be hired only when older faculty retire, and new NIH programs can begin only when other programs have ended. Perhaps this change should not have been surpris- ing. Growth cannot go on forever, as California Insti- tute of Technology physicist David Goodstein has pointed out: To illustrate his claim that "the era of ex- ponential growth in science is already over," he points Vol. 333 No. 12 SPECIAL REPORT 813 to Derek de Solla Price's 1961 calculation that if early rates of scientific growth persisted, the number of jour- nals would reach 1 million by the year 2000 (Fig. l).5 But, in fact, the number has plateaued at a mere 40,000 or so. Goodstein goes on to say that although "it is probably still true that 90 percent of all the scientists who ever lived are alive today . . . it cannot go on be- ing true for very much longer. . . . It is a simple mathematical fact that if scientists keep multiplying faster than people, there will soon be more scientists than there are people. That seems very unlikely to hap- p en."4 Still, none of us like to acknowledge the end of growth. I continue to ask the administration, at the be- ginning of each budget cycle, for an appropriation that would restore healthier success rates for grant applica- tions. Our many constituencies go to Congress each year hoping for substantial increases in appropriations for the NIH. And many universities continue to build fa- cilities for biomedical investigators who will be expect- ed to amortize the construction costs with new grants. But during the past year, such efforts have come to seem quixotic at best. Would it not be more sensible to put our energies into optimizing scientific activities in the steady-state environment that is likely to persist for the foreseeable future? ACCOMMODATING TO THE STEADY STATE In recent months, I have come to recognize that many of the initiatives recently proposed or undertaken at the NIH - and in other domains of the scientific es- tablishment - are designed to cope with the steady state. Some of these initiatives address issues funda- mental to the conduct of biomedical science; others at- tempt to generate money for research by achieving sav- ings through operational efficiencies. For example, we have been working hard to improve the peer-review process at the NIH. We are experi- menting with streamlined methods - such as elec- tronic communications, a triage system for reviewing applications, and postponement of the filing of admin- istrative forms until a grant seems likely to be funded - to save money and effort and to produce more judi- cious and more understandable decisions. We have de- bated proposals for distributing our funds more equita- bly. These plans could give more young investigators a chance to develop independent research programs, pre- vent abrupt terminations of advanced research careers, and ensure the vitality of the more fragile components of our enterprise, such as patient-oriented research and basic research that lacks an obvious connection to im- proved health. Others are also grappling with the emergence of the steady state. The Committee on Science, Engineering, and Public Policy of the National Academy of Sciences recently issued a response to the frequent accusation that the country is training too many new scientists! The committee concluded - wisely, I believe - that we should not reduce the size of our Ph.D. programs but instead help them to produce scientists acquainted with and equipped for many kinds of careers - in ed- 1 1,000,000 4 100,000 g 10,000] rj IOOO] f f 1700 1800 1900 2000 Year Figure 1. Cumulative Number of Journals Founded from 1665 to the Present. The broken line shows the trend based on the years 1750 to 1950. Adapted from Goodstein, with the permission of the p~blisher.~ ucation, business, journalism, law, and other fields - not just highly specialized, laboratory-based careers. Taken to its logical extreme, planning an optimal steady-state environment for the research supported by the NIH could have profound effects on the way we do business. We could recommend how laboratory groups should be configured in the stable world. (For example, the groups might be smaller, with fewer trainees and more technicians who have advanced degrees, and these smaller groups might be obliged to work together in research consortia to achieve efficiency and techni- cal diversity.) Going further, we could try to determine the right numbers, sizes, and types of grants in our portfolio. At the bottom of all such proposals is the need to achieve a balance between two, sometimes con- flicting pressures: the short-term demand for health- promoting results of research and the long-term need to maintain the vitality of the research enterprise. Raising such issues may be a useful means to focus attention on our problems. But it is prudent to worry about a solution to our predicament that seems like a planned economy. (We should at least have learned that lesson from the failure of communism in the Cold War.) Any attempt to plan substantial changes in the way science is practiced and supported must inevitably weigh stability against competition. I believe that American science has profited greatly from competition based on expert review and that science in other coun- tries has often been penalized by premature offers of lifelong support. If we are to think concretely about the steady-state world, as I believe we must try to do, it will be important to retain a healthy level of competition. This is one way to ensure that the steady state is dy- namic and not static. Still, the competitive mode has its limits, and we may have reached them. But are we equipped to envision, let alone determine, 8 I4 THE NEW ENGLAND JOURNAL OF MEDICINE Sept. 21, 1995 - 16- Lo C 0 - a 14- .- - 2 C 0 m .- *-' .g 12- 2 a cl the correct shape of biomedical science in a steady state? Such planning is not consistent with our national traditions or with the accepted values of our field. When central planning is attempted, it can deteriorate into corporate advertising or uninspired social determinism. Evolution under enlightened guiding principles seems a more desirable outcome. To develop such principles, we require detailed knowledge of how individual investiga- tors, departments, and entire institutions have func- tioned in the recent past. It is not difficult to use our computerized data bases to determine what the NIH in- vestments have been, but a systematic survey to identify all sources of money across departments or disciplines has not, to my knowledge, been undertaken. Until such idormation is available, even on a limited scale, it may be premature to say what we want. COMBATING THE NEGATIVE STATE To this point, I have considered the state of biomed- ical science from the optimistic perspective that we can eventually accommodate to constant rather than ex- panding support from the government. But during the past few months, those of us who have been reading the recommendations of congressional budget committees have learned that many members of Congress are will- ing to consider large reductions in NIH funding in the next fiscal year. Moreover, we have heard proposals that fundmg should be frozen at these lower levels for several addltional years, while inflation continues, until the national budget is balanced in the year 2002. Two plans were initially aired. The House Budget Committee proposed to reduce the NIH budget for fis- cal year 1996 to a level 5 percent below that in fiscal year 1995 (or 9 percent below the 4 percent increase that the Clinton administration has proposed for 1996 to maintain the steady state). The Senate Budget Com- mittee intended to reduce the fiscal year 1996 budget to a level 10 percent below the level in 1995 (or 14 per- cent below the President's request). Events took a more favorable turn when one of the strongest champions of biomedical research, Senator Mark Hatfield of Oregon, proposed an amendment that would restore most of the money that the Senate Budget Committee had pro- posed to cut below the 1995 level. With the help of thousands of scientists and patient advocates who peti- tioned their congressional representatives, this amend- ment was approved by the resounding vote of 85 to 14 on May 24. The final House-Senate budget resolution, however, would still leave the NIH budget 1 percent be- low the 1995 level in 1996 and 3 percent below the cur- rent level until the year 2002, when a balanced budget would be achieved. To appreciate the cumulative effects of this proposal, the numbers have to be viewed in the context of the an- nual inflationary rate for biomedical research, which is 4.3 percent (Fig. 2). With this rate of inflation, we would lose nearly a third of our purchasing power by the year 2002 if the joint budget resolution were followed. (In contrast, the plan recently announced by President Clinton for balancing the budget by the year 2005 would hew quite closely to the inflationary projection.) Of course, the budget resolution is an early phase 0 1995 Appropriation (base line) e House-Senate proposal 0 1995 BRDPI (4.3%) 101, I I I I I I I 1995 1996 1997 1998 1999 2000 2001 2002 Fiscal Year Figure 2. NIH Budget Projections from 1995 to 2002, Based on the Recommendations of the House-Senate Budget Confer- ence, the Biomedical Research and Development Price Index (BRDPI), and Constant Appropriations Equal to the Fiscal Year 1995 Ease Line. of the process required to allocate public money to agencies such as the NIH. The next phase occurs when the Appropriation Subcommittees in the House and the Senate meet to allocate their assigned funds to the many programs under their jurisdiction. Thus far, the NIH has fared well, perhaps even surprisingly well, in this crucial part of the process. Under the strong leadership of Rep. John Porter of Illinois, the House La- bor, Health and Human Services, and Education Ap- propriations Subcommittee designated $1 1.92 billion for the NIH, a proposal that has been endorsed by the full Appropriations Committee and the entire House of Representatives. This amount represents a 5.7 percent increase (more than $600 million) over the 1995 budget for the NIH, or somewhat more than the President pro- posed to keep pace with inflation. Now a parallel appropriations process must take place in the Senate, and any differences between the two legislative bodies must be resolved before the ap- propriations bill goes to the White House for the Pres- ident's signature or veto. So there will be many more events that merit close attention before the NIH can know whether it will be operating above or below the steady-state funding level in fiscal year 1996. Further- more, we will be facing similar budgetary battles for several more years, until a balanced budget is within sight and the national debt has been markedly re- duced. SOME ARGUMENTS FOR NIH FUNDING For these reasons, I believe it is prudent to marshal the arguments for the NIH - and for other programs that we endorse - to ensure that the best things that the government does for our society survive the frenzy of the budget cutters. Four such arguments have seemed especially effective to me in recent months. First, it is inappropriate to curtail a government pro- gram simply because it must "share the pain" of budget reduction. The Congress and the administration have an obligation to decide which programs they value most and to nurture them, even in rough times. It is a mistake to assume that every agency can absorb a 5 or 10 percent decrease simply by resetting internal priori- ties. This is especially true of agencies, such as the Vol. 333 No. 12 SPECIAL REPORT 815 NIH, that are setting priorities every year by choosing among programs and selecting only the most highly qualified applicants to receive support. Second, budget makers must distinguish between the services and the investments provided by government agencies. At an embassy dinner this spring, an influen- tial Republican senator told me that cutting the NIH budget made him just as uncomfortable as cutting funds for Amtrak or National Public Radio. I am a contented rider on the Metroliner and a devoted listener of Morn- ing Edition, but I know there is an important distinction between reducing the schedules of these services and re- ducing expenditures by the NIH. When federal funds are invested, not simply spent, even short-term reduc- tions have long-term effects. When we cannot afford to support an excellent grant application, the effect is greater than an inconvenient train schedule. We waste a resource in which the country has already made an appreciable investment, we forfeit knowledge on which others could build, we deprive ourselves and our chil- dren of better health, and we send a signal to our gifted youth that biomedical science lacks a future. Third, we must recognize and advertise the many benefits of NIH-supported research. Laura D'Andrea Tyson, the President's chief economic advisor, has maintained that federally supported research has an average economic return of 150 percent, which is even better than the return on private investment in re- search. The benefits of biomedical research are espe- cially great. In addition to the economic stimulus pro- vided to the industries that depend on our discoveries, the benefits include reduced expenditures for medical conditions that we have learned to prevent and hard- to-estimate values that the public places on improved health and extended life. The public needs to be re- peatedly reminded of the many vaccines against bac- terial and viral infections, the drugs that control psy- chiatric illnesses, the fluoridation of drinking water to prevent dental caries, the preventive strategies that have halved the rates of mortality from coronary artery disease and stroke, and effective treatments of certain cancers. These are just a few of the NIH-sponsored ad- vances whose multiple benefits demand continued em- phasis. The possibility of impeding new achievements - gene therapy for inherited diseases, methods to alle- viate chronic neurologic diseases, or new treatments for common cancers - should be a powerful disincentive to reduce the NIH budget. Finally, the public must be reminded that it highly values many things the federal government does with its discretionary money. By all reputable polls, our citi- zens are enthusiastic supporters of biomedical research and would even consider raising taxes to pay for it. I se- riously doubt that the public would have endorsed the cuts that the House and Senate Budget Committees proposed for the NIH; the public simply did not know what was happening. In his brilliant essay, "I Love Big Government," in the Washington Post Magmine,' Bob Garfield attempts to counter the proposal to return "power and autonomy to the states," by reminding us of the importance of cer- tain federal functions. Without them, Garfield imagines, [we] might have, say, a Vermont Division of Antitrust, but it would scarcely put the fear of God in a person. Think of tak- ing medicine regulated by the Utah Food and Drug Adminis- tration. Imagine pinning your cancer-cure hopes on the South Dakota Institutes of Health. Imagine the US. Virgin Islands Centers for Disease Control. The New Hampshire Aviation Administration. The Wyoming Secret Service. The Idaho Aeronautics and Space Administration. The Massachusetts Intelligence Agency. The Arkansas Reserve. A MODEST PROPOSAL FOR THE NIH In my role as a federal administrator, I have suggest- ed that the biomedical research enterprise has been un- dergoing a difficult but manageable change. But it is now also threatened with the more perilous prospect of declining fortunes. As a bench scientist, physician, and citizen, I must ask, Why should this be? Why should the NIH -with its popular goals, productive record, eco- nomic benefits, central role in sustaining our universi- ties and training new scientists, and prospects for im- proving health - be valued any less by the federal government than Social Security and other mandatory entitlement programs? The vicissitudes of the past few years and months have convinced me that the government and the public should fmd the means to secure a stable fiscal base for the NIH. Under such a plan, the NIH would be guar- anteed a budget not less than that of the previous year, with an inflationary increment determined on the basis of the rate of inflation in biomedical research and de- velopment. This plan would not create entitlements for individual scientists or scientific programs. Trainees would continue to compete for jobs in the academic, government, and industrial sectors. Investigators would still compete for support in both the extramural and in- tramural research communities. Institutes and centers would compete for allocations within the fixed NIH budget, and initiatives within each component of NIH would compete for support. Congress and the admin- istration would be able to supplement funding to meet specific threats to health or to increase the total size of the enterprise, if the economy permitted such sup- plements and they were justified by the science. With- in this secure framework, NIH-supported investiga- tors would be encouraged to think sensibly about the state of the enterprise and to develop the principles to guide its evolution. This would be a legacy at least as great as a balanced budget to present to future gener- ations. REFERENCES 1. 2. 3. 4. 5. 6. 7. Bishop JM, Kirschner M, Varmus H. Science and the new administration. Sci- ence 1993;259:444-5. Varmus H. Basic science and the NIH. Mol Biol Cell 1994;5:267-72. National Academy of Sciences, Committee on the Funding ofYoung Investi- gators in the Biological and Biomedical Sciences. The funding of young in- vestigators in the biological and biomedical sciences. Washington, D.C.: Na- tional Academy Press, 1994. Goodstein DL. Scientific elites and scientific illiterates. Engineering & Sci- ence. Spring 1993:23-31. de Solla Price DJ. Science since Babylon. New Haven, Conn.: Yale University Press, 1961. National Academy of Sciences, Committee on Science, Engineering, and Public Policy. Reshaping the graduate education of scientists and engineers. Washington, D.C: National Academy Press, 1995. Garfield B. I love big government Washington Post Magazine. February 5, 19951-8.25-6.