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 NINE RECOMMENDATIONS AND REMAINING CONSIDERATIONS The key feature of the National Research Service Award (NRSA) continues to be its ability to influence the quality and direction of research training within the biomedical and behavioral sciences. Its ability to promote multidisciplinary training provides a multiplier effect within graduate pro- grams. That is, the organization of the Paining experience within a program under the auspices of the NRSA can bring several disciplines to bear on the training of a single indi- vidual. The NRSA is also able to leverage the recruitment of minorities and women into research careers and influ- ence the issues that will be taken up by He research com- munity and the way in which that research will be con- ducted. In areas such as the behavioral sciences, where students have depended heavily on teaching assistantships to provide for graduate support, the NRSA reduces the ne- cessity for those commitments and can thereby facilitate the completion of doctoral studies in those fields. The committee notes, however, that there are significant weaknesses in the design of the NRSA program. For ex- ample, NRSA stipends are not competitive with other sti- pend sources, and training grant directors find themselves further frustrated by their inability to supplement those sti- pends with federal funds. These stipends are taxable, which further devalues these awards. Sensing some of these short- comings, the U.S. Congress and the National Institutes of Health (NIH) have introduced changes in the NRSA pro- gram in recent years designed to respond to emerging edu- cation and employment challenges. In 1993 the U.S. Con- gress authorized an increase in the number of NRSA trainees in response to the an increase in the number of students dropping out of doctoral degree programs during the previous decade. In the same Act (P.L. 103-43), Con- gress revised the payback provision by restricting it to postdoctoral support with the idea that the new arrangement 89 would encourage the participation and retention of physi- cian-researchers. Likewise, aware of the diminishing com- petitiveness of NRSA stipends in attracting the most able scientists to health research, NIH recently proposed to the Department of Health and Human Services that NRSA sti- pends be increased at the predoctoral level to $10,000 and at the f~rst-year postdoctoral level to $19,600.~ Each of these actions reflects a commitment on the part of the federal government to enhance the effectiveness of the NRSA program. The committee shares this commit- ment and has identified further modifications to the NRSA program to increase its effectiveness in meeting national needs for biomedical and behavioral scientists. The fea- tures of the NRSA program that merit immediate consider- ation are stipend support, multidisciplinary training, and flexibility in postdoctoral awards. STIPEND ISSUES Although interest in doing research and long-range em- ployment prospects supply compelling reasons for pursuing a research career, more immediate incentives, such as sti- pends, play an indisputable role. With that in mind, we find it disturbing to note that stipend levels for predoctoral train- ees in the NRSA program have remained unchanged since 1991 at $8,800 taxable salary per year.2 Numerous speakers at our public hearing in May 1993 voiced their concerns about the inadequacy of NRSA sti- pends. The existing structure of a $700 monthly stipend is simply not sufficient. Many state university stipends, for example, start at $11,000 and National Science Founda- tion currently pays $14,000. According to some speakers at our public hearing, most universities must work hard to supplement predoctoral stipends to raise them to $14,000
MEETING THE NATIONS NEEDS FOR BIOMEDICAL AND BEHAVIORAL SCIENTISTS because supplementation from other federal sources (e.g. research grants) is not allowed. When universities are unable to augment stipends, trainees are forced to seek other forms of part-time employment drawing them away from their academic program and extending their time to degree. Postdoctoral awarders do not fare much better, earning approximately $18,600 in their first year of training and $19,700 in their second. It becomes very difficult at this important period of training to entice a clinician, already burdened with debt, into a research career with a consider- able reduction in compensation in order to pursue prepara- tion as a scientist. Not only should stipend levels be in- creased to make them competitive, but the training budget should be sufficient to allow annual cost-of-living adjust- ments computed into each training grant's continuation base, with due consideration to differences in costs by re- gion, as suggested by public hearing participants. RECOMMENDATION: The committee recommends the NRSA stipend support at the predoctoral level be increased to $12,000 and f~rst-year postdoctoral stipends increased to $25,000 (both adjusted for inflation) by fis- cal 1997. In addition, there should be a yearly cost-of- living increase in NRSA stipends. This expansion in stipend support should be achieved through the addition of funds to the current NRSA training budget. Estimating Program Costs The committee has developed cost estimates for imple- menting these stipend increases, and balanced consideration of stipend increases against its numerical recommendations. In developing these estimates, the committee has had to make certain arbitrary decisions. First, we chose fiscal year 1993 as the base period because it is the most recent year for which reliable program data are available. Second, we developed numerical recommendations for the period 1994 through 1999 to overlap with the next assessment of the NRSA program scheduled for release in 1998. Much of the increase in program costs which result from our recommendations is concentrated in the years 1994- 1996, during which time stipend costs rise at an average annual rate of 7.8 percent per year. This results from the Committee' s recommendations to increase both the number of awards and stipend levels during that period. Annual growth in program costs from 1997 through 1999 is about 2 percent each year. This reflects the committee's recom- mendation to keep the number of awards constant and to limit stipend increases to the expected rates of inflation. Details regarding these calculations may be found in Ap- pendix H. 90 ENHANCING TEIE EFFECTIVENESS OF THE NRSA PROGRAM Flexibility in Career Training at the Postdoctoral Level In May 1993 we convened a public hearing to invite suggestions for increasing the effectiveness of the NRSA program. Most of those testifying on the role of the NRSA program in recruiting women said that the program must be more flexible in the areas of part-time training, reentry train- ing, family leave, and geographic location of training sites. Committee members have also been concerned, however, that there is a disparity between He number of women re- ceiving NRSA training and the number of recipients of NIH research grants. For example, women account for over 40 percent of the Ph.D.s produced in the life sciences, but they make up only 15 percent of the funded principal investiga- tors. The problem would seem to lie not with talented young women moving into science, but rawer with He de- velopment of Heir careers. Women appear to be leaving science between the time they receive their doctorate and the time Hat Hey fully es- tablish themselves in a research career Rack. Women are slightly less successful than men in obtaining FIRST awards, and proportionately fewer women apply for fund- ing at certain career stages (NIH, 1993a). Therefore, Here is a need to foster a supportive environment for career de- velopment. Institutional responsibilities should include mentoring, career advising, grant-writing training, and ad- vising on transitions along the career path. The NRSA pro- gram can clearly play a role in fostering the careers of these scientists. There is a need, then, to reshape NRSA awards at the postdoctoral level to encourage women to fully utilize their research talents. NRSA awards should allow retraining and career shifts to help women who have stopped out of re- search to update skills and move into emerging areas. More reentry options should be provided through NRSA pro- grams. RECOMMENDATION: The committee recommends that the NIH examine research training opportunities for women through He NRSA program and s~eng~en the role of postdoctoral support to assist women in establishing themselves in productive careers as research scientists. Monitoring Progress Toward NRSA Goals The NRSA is one of many sources of support available to individuals pursuing advanced preparation in research at the doctoral or the postdoctoral level. Designed to augment Federal support through the research assistantship, the NRSA program was designed to select qualified candidates from the pool of graduate students and postdoctoral person
RECOMMENDATIONS AND REMAINING CONSIDERATIONS net, provide them with a period of intense and advanced training, and launch them into productive research careers. Perhaps one of the most significant findings of this com- mittee is the general lack of information about the outcome of the NRSA program given almost two decades of support. Very little serious evaluation of the NRSA program has been undertaken with the support of NIH, except for a few student outcomes studies undertaken by earlier NRC com- mittees (Appendix A). We cannot underscore strongly enough the need for follow-up information to assess pro- gram outcomes. In part, this involves the organization of existing files at NIH to permit the analysis of program out- comes. In part, the analysis that is needed will require seri- ous review of data collection and analytic capabilities at NIH and the development of new strategies to assess career outcomes. Nowhere is the need for accurate information more evident than in our inability to track the participation of underrepresented minorities in the biomedical and be- havioral research effort. Data on Minority Parficipatton in Science Careers Present NIH data-collection procedures make it difficult to assess minority participation in various programs and to evaluate program effectiveness.3 From this perspective, four data concerns emerged: 1) NIH does not have a standard taxonomy for race and ethnic origin. The Public Health Service Form 398, which is used for competing research grants and for NRSA institu- tional training grants, specifies the following categories: American Indian or Alaska Native, Asian or Pacific Is- lander, black (not of Hispanic origin), Hispanic, and white (not of Hispanic origin). However, the information col- lected on the Statement of Appointment Form 2271 splits the Asian category into Asian (not a Pacific Islander) and Pacific Islander. This is a useful distinction because Pacific Islanders are generally considered to be underrepresented in biomedical research whereas other Asians are not. The dif- ferences in categories are puzzling. 2) The format of the tear-out page in the Gaining grant and fellowship applications on which applicants identify their race and/or ethnic origin and gender appears to result in varied and often very low response rates. Applicants are not obligated to respond and, moreover, are told that the sheet will be separated from the application and that the information will be entered into the central NIH database and used for aggregate statistical analyses. 3) Once the racial and ethnic data are encoded in the NIH master file, it is not readily available to program of ricers. One cannot, for example, obtain a racial and ethnic break- down of NRSA predoctoral fellows from the program of- fice. The only information that program offices have comes from an institutional certification that a fellow is an 91 underrepresented minority but does not specify which group. 4) A fourth problem common to many programs within NIH and in other agencies is ambiguity about whether the terminology should focus on minority or underrepresented minority. Although, application materials usually specify "underrepresented minority" (in targeted programs), it is not clear that the distinction is made consistently. Applicants for NRSA institutional training grants frequency mention Asians in reporting information although they are not one of specified target groups in the Minority Recruitment Plan. The distinction is crucial. Discussion in the NIH databook on minorities in extramural grant programs5 states that mi- norities received over 8 percent of total research grants but underrepresented minorities received only 2.7 percent of such awards (NIH, 1993b). The committee concludes that many of these shortcomings could be addressed if all ques- tionnaires designed to collect information about NRSA re- cipients use the same racial and ethnic categories. The Need for Well-Designed Career Outcomes Studies Improving the effectiveness of the NRSA program will require attention to issues not new to the research commu- nity. However, with the inevitable changes Hat will occur with health care reform and budget deficit reduction, NIH may find itself in a position of justifying its support for evaluation research. The NRSA program goes hand in hand with the government's role in financing fundamental re- search. This combination of research and Paining support has considerable and continuing benefits to He health and welfare of He citizens of our country. Well-designed ca- reer outcomes studies can provide the kind of feedback that is needed to ensure that the NRSA program is both efficient and effective given constraints being placed on the federal funding effort. The Need for Studies of Institutional Impact NRSA institutional Paining grants play an important role in providing funds to universities for student support. The committee heard testimony at its public hearing in May 1993 which suggests, however, that NRSA support may have the effect of inducing academic institutions to reduce (or not expand as much as they would otherwise) their own institutional funds devoted to the support of graduate stu- dents. Research by Ehrenberg, Rees, and Brewer (1993) suggests that this type of behavior can have a small but substantial effect on patterns of support. "Compensation" of this sort is particularly likely to happen at large research universities. Future committees would benefit from more studies of He impact of NRSA support on the recipient institution' s total pattern of training support.
MEETING THE NATION' S NEEDS FOR BIOMEDICAL AND BEHAVIORAL SCIENTISTS RECOMMENDATION: The committee recommends Cat Be NIH review its data bases as management infor- mation systems and introduce changes in data collection, analysis, and dissemination to permit more effective tracking of NRSA award recipients. Emphasis should be given to He analysis of minority participation in research and Gaining. New funds should be directed to the evalu- ation of NRSA program outcomes, including studies of the impact of institutional support on graduate student support patterns at U.S. universities. NOTES 1. The results of that request were not available at the time of com- mittee discussion. 2. The committee notes for comparison that the federal government set the `'poverty level" in 1990 at $6,257 for a one-person household below age 65, according to the U.S. Census Bureau, 1992. 3. In 1989 the NRC study committee outlined in detail a strategy for evaluating the NRSA program, but NIH was unable to initiate any evalua- tion studies until 1993, when a study of the Minority Access to Research 92 Careers (MARC) program was launched. The MARC program has served and should continue to address the important goal of strengthening the training capabilities of undergraduate minority institutions as well as train- ing minority students who might choose research careers in the biomedical and behavioral sciences. 4. The information in this section was drawn from a commissioned paper by Sharon Bush. See Appendix D for a list of other contributors. 5. For a copy of this report, contact Marie Chang, Division of Re- search Grants, National Institutes of Health, 301/594-7328. REFERENCES Ehrenberg, R.G., D.I. Rees and D.J. Brewer 1993 How would universitites respond to increased federal support for graduate students? In Studies of Supply and Demand in Higher Education, ed. C.T. Clotfelter and M. Rothschild, pp.l83-206. Chicago: University of Chicago Press. National Institutes of Health (bIH) 1993a NIH Data Book: 1993. Publication No. 93-1261. September. Bethesda, MD: National Institutes of Health. 1993b Minorities in NIH Extramural Grant Programs, Fiscal Year 1982-1991. SAKS, ISB, DRG of NIH. Bethesda, MD: Na tional Institutes of Health.