Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
CHAPTER FIVE PHYSICIAN- SCIENTISTS Training in the clinical sciences is critical to maintaining our country's leadership in the translation of basic discover- ies to meaningful patient care. As the nation demands more primary care of our physicians, we must not lose sight of the tremendous advances that have been made by individu- als using basic approaches to explore interesting and sig- nif~cant clinical problems. Clinical research includes a spectrum of investigation. At one end, it is represented by the use of basic scientific approaches and tissue samples from patients or normal indi- viduals to generate fundamental insights into the disease process. At the other end, it is represented by studies in which whole patients, normal volunteers, or populations of subjects each serve as He laboratory. The clinical investigator generally has an M.D. or other health professional doctorate, although the committee rec- ognizes that basic scientists also participate in clinical in- vestigation. The committee has based its assessment of na- tional need on the fact that most government-sponsored research in He clinical sciences is performed in medical schools or academic health centers (Appendix Table F-22~. The ability of medical schools to conduct clinical research depends largely on the continuing availability of clinical faculty with strong research skills. The future availability of well-prepared clinical research faculty has come into question by a number of authors (Ahrens, 1992; Fredrickson, 1993; IOM, 1994~. Given continuing national concern over the future supply of skilled clinical investiga- tors, we have restricted our analyses in this chapter to the need for physician-scientists.2 Previous National Research Council (NRC) study com- mittees have focused on the special role that He physician- scientist has played in bringing clinical insights to bear in the laboratory and in translating new knowledge into the context of medical practice (NRC, 1981~. Almost all NRC 55 committees that have addressed research training needs in the clinical sciences have observed that there continues to be a shortage of physicians willing to prepare for research careers. Many committees have focused on He very real effects of competing-and more lucrative-opportunities available in private practice as a reason for this trend (NRC, 1978~. More recently, some committees have observed that changes occurring in the way medical schools finance their operations and structure Heir faculties simply does not pro- vide an environment conducive to preparation for a research career (NRC, 1985~. We concur and provide evidence else- where in this chapter suggesting that upcoming changes in the national support for health research and health care re- form may further erode research and research-training op- portunities in academic health centers. In addition to these contextual variables, we believe the nature and timing of National Research Service Award sup- port may directly effect the success of recruiting physicians into research careers. On the basis of information gathered by the National Institutes of Health (NIH), we believe that the Medical Scientist Training Program (MSTP) may be especially effective in launching individuals into research careers. This program was established in 1964 to permit individuals to pursue the M.D. and the Ph.D. degrees con- currently. The MSTP program has consistently had a high proportion of graduates involved in research and actively contributing to He advancement of the clinical sciences. (See also, Appendix A for a summary of available outcome studies.) Opportunities for careers in clinical research abound. Our continuing challenge is to stimulate interest of clini- cians in contributing to that effort, and He NRSA program can clearly play a role relative to that goal.3 In the sections that follow, we will review some of the more exciting ad- vances in clinical science that create He need for a continu
MEETING THE NATION' S NEEDS FOR BIOMEDICAL AND BEHAVIORAL SCIENTISTS ing research effort, summarize the current market for these scientists, and recommend specific changes in the NRSA program that may be effective in expanding the cadre of physician-scientists needed at this time. ADVANCES IN CLINICAL SCIENCE Advances in clinical science have been enormous and include, but are not limited, to the following: · Identification of the genetic defect in various genetic disorders, including cystic fibrosis. Cystic fibrosis is the most common genetic disorder in Caucasians, affecting 1 of every 2,000 children. The disease is characterized by pul- monary infections and pancreatic insufficiency and is due to a cellular defect in the development of secretions. The genetic defect associated with the disorder is found in chro- mosome 7. This discovery allows three major advances. First, it allows genetic counseling within families. Second, it has allowed a determination of the product of the gene. This information will provide a rational approach to devel- oping drugs to correct He defect. Finally, it will allow studies that attempt to replace the defective gene with a normal one in tissues that are affected. Indeed, such so- matic gene therapy has already begun. e Identification of the gene associated with bowel can- cer. Very recently, two separate groups of investigators demonstrated a genetic defect localized to chromosome 2, which is associated with hereditary nonpolyposis colon can- cer. The gene involved appears to control DNA repair, and a defective gene seen in patients with colon cancer leads to instability of cellular DNA. This research is a spectacular example of the different ways in which basic research can lead to clinical advances. In one laboratory the research developed from studies performed in yeast and bacteria that examined how these organisms repair DNA and the genetic defects associated with DNA instability. In another labora- tory there is a long history of studies in humans examining genetic defects associated win a variety of colon cancer syndromes. In other words, this remarkable advance in our understanding of colon cancer came from distinct pathways, one originating from basic studies of normal mechanisms in bacteria and yeast and He over from more clinically ori- ented studies looking at abnormal grown and differentia- tion of colon cells. These studies will allow Be develop- ment of reagents that can be used to screen for colon cancer. · Creation of an animal model for ankylosis spondylitis by using transgene methodology. Ankylosis spondylitis is a syndrome Hat predominantly affects joints of the spine. Approximately two decades ago it could be demonstrated that the disease was significantly associated with a specific HLA type, HLA-B27. Indeed, 90 percent of patients win ankylosis spondylitis had He HLA-B27 genotype. In an 56 attempt to demonstrate the nature of He association between the gene and the disease, investigators established a rat model in which the human HLA-B27 gene was inserted by using aansgene methodology. In some of Be animals a disease developed that mimicked human ankylosis spondylitis. These animals not only provide a model for determining just how He gene influences Be expression of the disease but also for deciding what other factors may be involved. They also provide a model for studying the effec- tiveness of various forms of therapy. The importance of clinical research to advancing our un- derstanding of clinical disorders is captured in a recent edi- torial in Science written by Editor-in-Chief Daniel E. Koshland Jr. (1993~: In the 1980s and l990s NIH researchers, intramural and extramural, performed the first trial of gene therapy in humans, proved the effectiveness of methotrexate for treating rheumatoid arthritis, developed new methods for growing skin to repair burns, showed that control of glu- cose levels slows progression of diabetes, showed effec- tiveness of cholesterol reduction in the prevention of heart disease, demonstrated an effective treatment for spinal cord injury, found a new drug for Parkinson's disease, showed that aspirin and coumadin lower the risk of stroke, developed methods of hypertension control that have reduced heart attacks and strokes by more than 50 percent, and so on for many other discoveries..... These followed many earlier discoveries, including the polio vaccine, the measles vaccine, hormone replace- ment therapy, fluoride to prevent tooth decay, to name a few. We are living longer, we are living with less pain, we are living with less cost to alleviate health deficien- cies than any previous generation because of the find ines of health researchers In the not-so-distant past, smallpox epidemics killed 25 percent of the inhabitants of towns that were invaded by the virus. Today we are storing the last traces of the virum because that dread disease has been eradicated from the Earth. Clearly, this partial list of clinically relevant discoveries supports the practical value of clinical research. lThe United States is the world leader in clinical research and we must make a renewed commitment to retain this leadership. The recommendations of this report should allow us to remain in this position of preeminence. ASSESSMENT OF TO CURRENT MARKET FOR CLINICAL SCIENTISTS Degree Production and Career Patterns Clinical scientists work in a variety of settings but pri- marily in academic heals centers. Between 1981 and 1991, the number of full-time faculty employed in clinical depart
PHYSICIAN-SCIENTISTS meets grew by about 38,000 to just over 59,000 (Appendix Table F-23), suggesting that the market for clinical scien- tists remained relatively robust throughout He 1980s. Degree Production The major source of new physician-scientists is the nation's medical schools. The most readily available infor- mation about patterns of enrollment and degree production is He Association of American Medical Colleges (AAMC). Data from AAMC (Appendix Table F-23) reveals Hat medi- cal school enrollments remained essentially flat between 1981 and 1991 at about 65,000 students. The number of medical degrees awarded each year also remained level at about 15,500 per year in the 1980s. Career Patterns Few data sets are available for sorting out the unique patterns of research careers among physician-scientists. The American Medical Association provides information about the number of physicians primarily engaged in research (Appendix Table F-23~- which fluctuated between 16,000 and 23,000 between 1981 and 1990. But these figures may also include trainees in graduate medical programs. Per- haps more telling is the trend in success rates of M.D.s who apply to the National Institutes of Health which peaked at about 45 percent in 1987 and has leveled off at about 37 percent (on average) thereafter (Appendix Table F-23~. Market Forces There are several influences on the availability of ca- reers in clinical research. These influences, called market forces, range from how we have traditionally trained clini- cal researchers to outside industrial and governmental spending trends. As the nation begins to develop a new system of health care delivery, these market forces will be- come increasingly important. Academic Health Centers An academic health center can be defined as a medical school working in conjunction with a teaching hospital and at least one over health professional school to achieve mu- tually agreed upon goals for education, research, and provi- sion of clinical care. Approximately 68 percent of NIH R01 support goes to these academic health centers. Aca- demic health centers therefore constitute the major sites at which heal~-related research and research training are car- ried out. Moreover, a significant amount of their support for research is derived from income obtained for the provi- sion of clinical care. This income stream is threatened by 57 changes in healthcare reform Hat place academic heals centers at a disadvantage with regard to the cost of provid- ing medical care. This presents a Treat to He market not only for training but also for support of trained investigators. Pharmaceutical and Biotechnology Industry Uncertainties in health care reform has forced industry to be exceedingly cautious win expenditures, and in some cases to lay off large numbers of employees. This posture clearly stifles innovation. One of the first areas to feel the effects of budgetary constraints is research. This soft side of the market has to be balanced by the fact that there are tremendous opportunities for the development of unique agents to Heat significant clinical disorders. Cap on Domestic Spending The federal deficit, budget reconciliation, and a cap on domestic spending indicates that support for research and training will have to compete for many other high-priority areas supported by the domestic budget. This scenario is one in which the NIH budget is likely to grow at a rate certainly not greater and probably somewhat less than the biomedical research price index. Emphasis on Increasing the Training of Generalists There clearly is an enormous pressure nationally to in- crease the proportion of generalists in medicine and de- crease the proportion of specialists. Heretofore, significant research training and research activity has occurred in asso- ciation win specialties, particularly the medical specialties. Indeed, some view the problem in the imbalance of general- ists to specialists as a result of overemphasis on research spending. This, therefore, provides a diminished en~usi- asm among some to further increase funding for research or research training. OUTLOOK FOR CLINICAL SCIENTISTS In addition to market forces, there are factors that influ- ence the demand for clinical scientists. These demand indi- cators are expenditures for clinical research and develop- ment (R&D) in medical schools; professional service income in medical schools; total revenue; budgeted vacan- cies in medical schools, both in clinical and basic science departments; and the clinical faculty/student ratio. Expenditures For Clinical Research and Development From 1985 to 1990, expenditures for clinical R&D in medical schools increased moderately. The average in
MEETING THE NATION9S NEEDS FOR BIOMEDICAL AND BEHAVIORAL SCIENTISTS crease was about 13 percent per year. An estimate of the amount of support for clinical R&D in medical schools is needed to refine the model of demand for clinical faculty. An estimate of clinical R&D expenditures in medical schools was derived by using the proportion of total NIH obligations used to support clinical research. From 1969 to 1989 this proportion increased by 60 percent (Appendix Table F-22). Since 1980 public medical schools have had higher lev- els of research expenditures than have private schools. This is partly due to the fast grown in the number of public schools. Clinical R&D in public schools grew at an annual rate of 7 percent since 1980 as compared with only about 4 percent per year in private schools. However, private schools remain more research intensive as indicated by re- search expenditures per school. Average clinical R&D ex- penditures were $14.6 million in private schools in 1990 compared win $10.2 million in public schools (Appendix Table F-24). Professional Service Income Service income generated by medical school faculty has continued to grow. From 1989 to 1990 service income gen- erated by medical school faculty grew 14 percent and from 1990 to 1991 this income grew 13 percent (figures adjusted for inflation in 1987 dollars). This has increased as medical schools have become very successful in providing clinical care. Thus, medical schools have come to depend on the clinical income to support their research and educational . . missions. Total Medical School Revenue Service income and federal research funds contributed over half of all medical school revenues in 1991. Another large portion came from state and local government sources. Tuition contributed only about 4 percent in 1991. With an average yearly tuition increase since 1985 of about 6 per- cent, medical student indebtedness, as noted by several testifiers at the public hearing, may operate as a deterrent to their pursuit of research training. The rates of total rev- enues have grown at an average yearly rate of 14 percent since 1986. Budgeted Faculty Vacancies Total budgeted medical school faculty vacancies have grown at an average yearly rate of about 6 percent since 1989. The major growth of vacancies is in He clinical sci- ence departments. There has been a steady decrease of fac- ulty vacancies in He basic science departments with a high 58 of 801 budgeted vacancies in 1985 to the 1991 low of 643 vacancies. Faculty/Student Ratio Enrollments, revenue, and clinical faculty size are the basic elements in assessing personnel needs for He clinical sciences in medical schools. The ratio of clinical faculty to enrollment is largely determined by He funds available to support faculty. Priority Fields Clinical investigation requires practitioners to stay abreast of developments in both medicine and science, each of which is in constant acceleration and often the two do not track in parallel directions (Fredrickson, 1993~. Obser- vations from the study of patients lead to the development of hypotheses, which lead, in turn, to scientific experimen- tation. Interest in the patient is always paramount, but sci- entific experimentation runs the risk today of taking the clinical investigator away from the bedside to the clinical laboratory. Ahrens (1992), in particular, has decried the reductionist direction of clinical investigation, suggesting that patient-oriented research is seriously imperiled. We concur with Ahrens view that more emphasis should be placed on the preparation of investigators familiar with the experimental paradigms associated with patient-oriented re- search. At He same time we recognize that laboratory- based clinical investigation has a significant and continuing role in He national health effort. However, from our review of the literature, and on He basis of our expert judgement, we cannot help but conclude that there is indeed a dearth of individuals adequately trained to perform patient-oriented or population-based research. With the development of new therapies and diagnostic procedures, there is an urgent need to train individuals who can carry these advances into the clinic so that Heir effec- tiveness can be measured and made available to the nation. NRSA funds, either through individual NRSA fellowships or programmatic training grants, can play an effective role in promoting the specialists that are needed. The MSTP also represents a priority field. Established in the 1960s, this program has been especially attentive to the essential training requirements for clinical investigation. A 1992 study of graduates of the Johns Hopkins University's M.D./Ph.D. program found that all of those who had completed their training were actively involved in research: 81 percent in academia, 14 percent at research institutes, and 5 percent in the biotechnology industry (McClellan and Talalay, 1992. See also Bradford et al., 1986 and Frieden and Fox, 1991~. NIH has also analyzed information about first-time ap- plicants for research grants (R01) and prior research train
PHYSICIAN-SCIENTISTS ink experience (Appendix Table F-25~. They found that in 1989 nearly 60 percent of individuals holding joint M.D./ Ph.D. degrees and applying for research support had re- ceived formal research training Trough support provided by NIH; this value was 47 percent for Ph.D. applicants and 42 percent for M.D. applicants. Furthermore, among first- time NIH grant recipients in 1989, 68 percent of the M.D./ Ph.D. recipients had had previous NIH supported research training experience compared win 55 percent of the grant recipients holding Ph.D. degrees and 52 percent of those holding M.D. degrees (Appendix Table F-26~. We con- clude that continued and expanded support of the MSTP program will yield He cadre of active and successful physi- cian-scientists so sorely needed for the national research effort today. ENSURING TEW DIVERSITY OF HUMAN RESOURCES Issues remain regarding He recruitment of minorities to faculty and He retention of all M.D. investigators re~ard- less of ethnicity and gender. In addition to improved re- cruitment, there must be specific attention given to the re- tention of women as clinical investigators and faculty. Extending the tenure clock and having on-site day-care are two examples of ways to facilitate their retention. Race and Ethnicity Medical school faculty reveal a race/ethnicity mix simi- lar to the basic biomedical sciences (AAMC, 1992~. Be cause most of the U.S. population will soon be a mixture of races over than white, the market will demand a more widely representative pool of researchers. About 13 per- cent of the faculty are members of minority groups and the largest share of these workers is Asian (Table 5-1~. Table 5-1 displays the medical school faculty by rank and ethnicity: of the professors, 87.6 percent are white, 5.7 percent are Asian, and 2.4 percent declined to respond; of the associate professors, 82.5 percent are white, 7.9 percent are Asian, and 3.1 percent declined to respond; of the assis- tant professors, 77.5 percent are white and 8.6 percent are Asian, and information was missing on 4.8 percent; of the instructors, 72 percent are white and 9.4 percent are Asian, and information was missing on 7.4 percent. Although 13 percent of the entire faculty represent minorities, this mix is generally not yet reflected in higher faculty ranks. Age Figure 5-1 shows He distribution of U.S. medical school faculty by age. Out of a total of 70,187 faculty, 57.6 per- cent are ages 40-49 and 25 percent are ages 30-39. Table 5- 2 indicates that of those aged 40-49, 22.4 percent are full professors, 55.3 percent are associate professors, 39.1 per- cent are assistant professors, and 29.9 percent are insauc- tors. Of those aged 30-39, 0.3 percent are professors, 8.1 percent are associate professors, 45.8 percent are assistant professors, and 51.8 percent are instructors. Of the M.D./ Ph.D. graduates, 39.8 percent are ages 40-49 and 25 percent are ages 50-59 (Table 5-3~. Only 15.6 percent M.D./Ph.D. graduates are ages 30-39. TABLE 5-1 Distribution of U.S. Medical School Faculty by Rank and Ethnicity Professor EthnicityNumberHoNumberHoNumber SoNumber % Native American240.1130.122 0.111 0.2 Asian1,0655.71,3347.92,300 8.6590 904 Black19313191.9778 2.9300 408 Mexican American280.1410.2114 0.419 0.3 Puerto Rican960.51460.9247 0.988 104 Other Hispanic2581.42551.5461 107129 201 White16,39687.614,00882.520,838 77054,512 72 Refused4462.45313.1843 301156 2.5 Missing2151.133421,293 408465 704 Total18,72110016,98110026,896 1006,270 100 Associate Professor Assistant Professor Instructor a Declined to respond. SOURCE: Association of American Medical Colleges (1992~. 59
MEETING THE NATION'S NEEDS FOR BIOMEDICAL AND BEHAVIORAL SCIENTISTS Total = 70,187 30-39 (1,420) 2.0% (17,521) 25.0% 60-69 (8,082) 1 1 .5% Missing (1,493) 2.1% FIGURE 5-1 Dis~ibubon of U.S. medical school faculty by age. SOURCE: Association of American Medical Colleges (19921. TO NRSA PROGRAM IN TElE CLINICAL SCIENCES 50-59 (15,091) 21.5% Every NRC study committee has noted Cat recruitment of qualified clinical researchers poses special challenges. Under 30 Physicians, dentists, and veterinarians enjoy several at~ac ,(215) 0.3% live career alternatives. The vast bunk of Ph.D. trainees pursue research careers, but only one-~ird of postdoctoral M.D. trainees have followed that path, with most entering medical practice instead. Part of Me reason is that prepar- ing clinical specialists for practice in shortage fields has been an explicit purpose of the NIH funding programs. With that goal now substantially met, however, Me commit- tee doubted the need for continued subsidy to clinical train- ing for practice. TABLE 5-2 Distribution of U.S. Medical School Faculty by Degree and Age Age Professor Number % Number Associate Professor % o 801 55.3 25.1 9 1.3 1.1 100 Assistant Professor Number _ 12314 10512 2286 790 129 807 268g6 % 0.2 45.8 39.1 8.5 2.9 0.5 3 100 Number 102 3248 1875 533 168 25 319 6270 Instructor % 1.6 51.8 29.9 8.5 2.7 0.4 5.1 100 Under 30 30.39 4049 50-59 60~9 70+ Missing Total SOURCE: Association of American Medical Colleges (1992~. o 62 4185 7835 5526 1038 75 18721 o 0.3 22.4 41.9 29.5 5.5 0.4 100 o 1375 9397 4265 1S32 220 192 16981 TABLE 5-3 Distribution of U.S. Medical School Faculty by Rank and Age Age Under 30 30-39 4049 50-59 60~9 70+ Missing Total . . M.D. Ph.D./O.H.Da M.D.-Ph.D./M.D.-O.H.D.a Other Number % Number % Number % Number . . . . 60 12114 15235 8786 5203 971 456 42825 0.1 28.3 35.6 20.5 12.1 2.3 101 100 58 3715 7976 4652 1995 268 237 18901 0.3 19.7 4202 24.6 10.6 1.4 1.3 100 a O.H.D.: Other health doctorate. b Other: Faculty with non~octoral/no degree or missing degree data. SOURCE: Association of American Medical Colleges (1992~. 60 600 1529 960 586 131 31 3838 o 15.6 39.8 25 15.3 3.4 0.8 100 , . . 96 1092 1625 693 298 50 769 4623 i 2.1 23.6 35.2 15 6.4 1.! 16.6 100
PHYSICIAN -S CIENTISTS In its first report in 1975, the committee found the avail- able data on clinical researchers to be wholly inadequate to its needs, and it declined to recommend any change in ex- isting levels of funding: 140 predoctoral and 3,340 postdoctoral clinical sciences trainees. Beginning with the 1976 report, the committee began to grapple in earnest with the scope of its task. On the basis of unique value and the special demands of clinical research and the fact that pro- fessional schools do not ordinarily prepare students for ca- reers as researchers, the committee concluded that postdoctoral clinical trainees should generally receive their support in the form of training grants made to professional schools, which permit these institutions to build in a short time the critical mass of students and web of resources nec- essary for high-quality programs. Evidence available in the early 1970s suggested that, un- like the burgeoning supply of Ph.D. researchers, the pool of M.D. investigators was shrinking while demand was grow- ing steadily. Despite 6 percent annual grown in medical school faculties, the American Medical Association figures showed a significant drop since 1968 in the number of nhv- sicians engaged primarily in research. Therefore, the com- mittee recommended funding a total of 2,800 postdoctoral traineeships and fellowships, up 10 percent from the num- ber funded in 1975. It also praised the MSTP, initiated in fiscal year 1964, which supported students in combined 6- year M.D./Ph.D. courses. The committee recommended funding 600 MSTP traineeships, up 19 from the 581 funded in 1975. Win the next several reports beginning in 1977, the com- mittee began exploring why the number of physicians-sci- entists was dropping. Although both enrollment and R&D funding were rising rapidly at medical schools, many estab- lished clinical faculty members were spending relatively little of their time conducting research. In response to this fin ding, the committee detailed a number of factors that it believed might discourage physicians from undertaking re- search careers: · the risk of failing at an untried field after demonstrat- ing the ability to succeed in medical practice, · the loss of income as compared to practice, · a growing perception among students that patient care has greater value than research, · social pressure on students to enter primary care fields, and · an image that paperwork and red tape inhibit research- ers more than in the past. In addition, the committee noted a discrepancy between the medical training calendar and the NIH grant cycle. Physi- cians who were planning for residencies to begin in July had to do so as early as the preceding October, many months before NIH announced its training awards. 61 In view of these circumstances, He committee continued through the 1970s to recommend 2,800 postdoctoral traineeships and fellowships in clinical sciences. It also continued to praise He MSTP awards as a highly effective method of producing clinical researchers, recommending incremental increases in He program. By 1979 the committee's warnings appeared to have had an effect. Presidents of four leading societies discussed He threat of clinical investigator shortage in major addresses, as did a conference at He University of Chicago. Various agencies had already begun trying to counter the shortage. NIH, for example, had expended and modified its grant mechanism to ease the transition from medical school to research training and then to independent research. A 1978 amendment to the NRSA Act encouraged students to do short-term research under 3-month grants not subject to payback. Several other developments that pointed toward a brighter outlook were an increase in the number of physi- cians reporting research as a major activity, an increase in the number of clinical science traineeships and fellowships, and survey results showing that medical students were growing more interested in research. The committee main- tained its recommendation of 2,800 postdoctoral awards. During the early 1980s the committee continued to rec- ommend holding the number of awards stable at 2,800. Market opportunities for clinical investigators continued to be favorable, win medical school faculties still growing and providing places for young scientists interested in re- search careers. The immediate problem was the recruit- ment of physicians to undertake research Raining. The com- mittee was concerned about a looming physician surplus, which would probably slow the growth of medical school enrollments and faculty and in turn reduce the positions available to new clinical researchers. Even win fewer open- ings, however, Be committee believed that clinical investi- gator posts would remain hard to fill. In 1985 He committee recommended a rise in He num- ber of NRSA awards. It believed Hat demand would grow faster than expected, in part because of increasing attrition from an aging faculty pool. The 1985 report also highlighted some important changes in medical school financing that the committee feared might further weaken clinical departments' commitment to re- search. As revenue from patient care steadily climbed, the committee believed that clinical faculty might find these demands competing for the time needed to prepare propos- als, collect data, write grants, and so forth. In addition, as faculties grow less rapidly, medical school might favor hir- ing clinician-teachers over physician-scientists. The committee also examined factors affecting young dentists' decision to pursue careers in clinical research. Al- though, unlike physicians, dentists have ample opportuni
MEETING THE NATION' S NEEDS FOR BIOMEDICAL AND BEHAVIORAL SCIENTISTS TABLE 5-4 Aggregated Numbers of NRSA Supported Trainees in Me Medical Scientist Training Program (MSTP) for FY 1991 through FY 1993. Fiscal Year 1991 1992 1993 Number of Predoctoral Trainees 783 806 822 NOTE: Based on estimates provided by the National Institutes of Health See Summary Table 1. ties for research during postdoctoral specialty training, only a few trainees receive salaries and some must even pay tu- ition. The committee recommended special consideration to providing adequate support for training dentist-researchers. In 1989 the committee noted that the number of NIH traineeships and fellowships for clinical investigators (whom it chose to call physician/scientists) had not in- creased as fast as health-related R&D expenditures. The percentage of M.D.s who were principal investigators on NIH research grants had fallen, although the number of M.D./Ph.D. principal investigators had remained constant for the past decade. The committee speculated that He demand for physi- cian-scientists would increase in the future as health-related R&D increased. However, given the lack of compelling data about supply and demand and questions about the effec- tiveness of physician research training, the committee rec- ommended that the number of training positions remain the same until current training programs could be evaluated. RECOMMENDATIONS The following recommendations are made to enhance our excellence in physician-based research. The Medical Scientist Training Program In 1963, NIH granted funds to three institutions to sup- port just under 20 individuals who pursued the M.D. and Ph.D. degrees concurrently. Early NRC study committees reported findings from studies that consistently showed that a substantial fraction of MSTP awarders remain produc- tively engaged in research, often with greater success in securing research support than M.D.s who pursue post-M.D. research training not leading to a doctorate. Current support for M.D./Ph.D. Raining provides for about 820 awards (Table 5-4~. Given the success of this program in contributing workers to the national research effort and the continuing and increasingly difficult problem of attracting M.D.s without Ph.D. training to research ca- reers, we believe this program should be expanded signifi- cantly in the coming years (Table 5-5~. RECOMMENDATION: To meet He nation's continu- ing need for clinical investigators, the committee recom- mends that the number of NRSA trainees supported through the MSTP program be increased from 822 in 1993 to 1,020 trainees each year by the year 1996. Individual Fellowships Because of the urgent need for clinical scientists familiar with patient-based research techniques, we urge NIH to in- crease the number of postdoctoral NRSA fellowship awards to permit the preparation of patient-based investigators. RECOMMENDATION: The committee recommends Hat NIH increase support for individuals to train in pa- tient-based research by increasing He number of TABLE 5-5 Committee Recommendations for Predoctoral Tnuneeships In the Medical Scientist Training Program for FY 1994 through FY 1999. Fiscal Year Number of Predoctoral Traillees 1994 1995 1996 1997 1998 1999 890 955 1,020 1,020 1,020 1,020 ~. 62
PHYSICIAN - S CIENTI S TS TABLE 5-6 Aggregate Numbers of NRSA Supported Trainees and Fellows in Clinical Sciences for FY 1991 through FY 1993 Fiscal Type of Support Year Level of Trading TOTAL Traineeship Fellowship 1991 Number of awards 2,894 2,814 80 Predoctoral 755 736 19 Postdoctoral 2,139 2,078 61 1992 Number of awards 2,970 2,887 83 Predoctoral 819 800 19 Postdoctoral 2,151 2,087 64 1993 Number of awards 2,974 2,877 97 Predoctoral 855 826 29 Postdoctoral 2,119 2,051 68 _ NOTE: Based on estimates provided by the National Institutes of Health. See Summary Table 1. postdoctoral fellowships in the clinical sciences from 68 in fiscal 1993 to 160 by fiscal 1996. Institutional Training Grants in the Clinical Sciences To permit the expansion of the pool of MSTP trainees and postdoctoral clinical science fellows, we believe mod- est reductions should be made in the number of postdoctoral awards made through institutional training grants. NIH re- ports Cat 2,087 awardees were supported in fiscal 1992 through this mechanism (Table 5-64. We believe a gradual decrease should occur in the number of awards (Table 5-7~. This would be done to permit the expansion of He MSTP program (described above). RECOMMENDATION: The committee recommends that He number of postdoctoral institutional aa~neeships supported Trough the NRSA program in He clinical sci- ences be slightly decreased from 2,051 to 1,965 between 1993 and 1996. NOTES 1. Several studies, it must be added, have identified a lack of rigor- ously trained individuals who know how to perform patient-based research (e.g., Ahrens, 1992) as a special need at this time. 2. The clinical sciences are understood to include individuals holding degrees in a variety of health professions including: medicine, veterinary sciences, dentistry, nursing, clinical psychology, and social work. The research training needs of clinical psychologists have been addressed in chapter 4 of this report ("Behavioral Sciences"), dentistry needs are sepa- rately addressed in chapter 6 ("Oral Health Research"), and nursing ad- dressed in chapter 7 ("Nursing Research"). The committee did not address research training needs in the veterinary sciences or social work, but rec- ognizes that these fields contribute to the national research effort and merit support through the NRSA program. 3. A recent report of the Institute of Medicine, Careers in Research: 63 Obstacles and Opportunities (1994) investigates ways to improve the qual- ity of training for clinical investigators and delineates pathways for indi- viduals pursuing careers in clinical investigation in nursing, dentistry, med- icine and other health professions engaged in human research. REFERENCES Ahrens, E. H., Jr. 1992 The Crisis in Clinical Research: Overcoming Institutional Obstacles. New York: Oxford University Press. Association of American Medical Colleges (AAMC) 1992 U.S. Medical School Faculty: 1992. Washington, D.C.: Asso- ciation of American Medical Colleges. Bradford, W.D., S. Pizzo, and A.C. Christakos 1986 Careers and professional activities of graduates of a Medical Scientist Training Program. Journal of Medical Education. 61:915-918. Fredrickson, D.S. 1993 Clinical Investigation. Paper prepared for the Committee on National Needs for Biomedical and Research Personnel. Frieden, C. and B.J. Fox 1991 Career choices of graduates from Washington University's Medical Scientist Training Program. Academic Medicine. 66:162-164. Institute of Medicine (IOM) 1994 Careers in Clinical Research: Obstacles and Opportunities. Washington, D.C.: National Academy Press. McClellan, D.A. and P. Talalay 1992 M.D.-Ph.D. training at the Johns Hopkins University School of Medicine, 1962-1991. Academic Medicine. 67(1):36-41. National Research Council (NRC) 1975 Personnel Needs and Training for Biomedical and Behav ioral Research. Washington, D.C.: National Academy Press. 1978 Personnel Needs and Training for Biomedical and Behav ioral Research. Washington, D.C.: National Academy Press. 1985 Personnel Needs and Training for Biomedical and Behav ioral Research. Washington, D.C.: National Academy Press. 1989 Biomedical and Behavioral Research Scientists: Their Train ing and Supply, Volume I: Findings. Washington, D.C.: Na tional Academy Press.
MEETING THE NATION' S NEEDS FOR BIOMEDICAL AND BEHAVIORAL SCIENTISTS TABLE 5-7 Committee Recommendations for Relative Distribution of Predoctoral and Postdoctoral Tra~neeship and Fellowship Awards in Clinical Sciences for FY 1994 through FY 1999 Fiscal Type of Support Year Level of Training TOTAL Trameeship Fellowship 1994 Recommended number of awards 2,975 2,875 100 Predoctoral 895 875 20 Postdoctoral 2,080 2,000 80 1995 Recommended number of awards 2,910 2,780 130 Predoctoral 895 875 20 Postdoctoral 2,015 1,905 110 1996 Recommended number of awards 2,860 2,080 180 Predoctoral 895 875 20 Postdoctoral 1,965 1,805 160 1997 Recommended number of awards 2,860 2,680 180 Predoctoral 895 875 20 Postdoctoral 1,965 1,805 160 195~d Recommended number of awards 2,860 2,ou ~1bU Predoctoral 895 875 20 Postdoctoral 1,965 1,805 160 In Recommended number or awards 2,860 ~,osu Lou Predoctoral 895 875 20 Postdoctoral 1 965 1,805 160 64