Transition to First Independent Position
The grants system of the National Institutes of Health (NIH), built on the principles of peer review, merit-based funding, and transparency of processes, in many ways represents a paradigm for U.S. government funding, and it has served the nation well. Yet, there is room for improvement. One area in particular involves the experiences of postdoctoral researchers as they look to the future. As described in the previous chapter, postdocs do not always receive the mentoring and training required to successfully pursue a variety of careers. In addition, there are challenges to conducting independent research during postdoctoral tenures.
Over the past 20 years, scientists have been receiving their first NIH grants at increasingly older ages (see Figure 1-2). The average age at which investigators receive their first independent research support is creeping upward. The average age at which biomedical researcher receive their first faculty appointment is also increasing, but there is still a 4–7 year lag from becoming a faculty member to receiving a first R01 (Chapter 2).
The growing number of postdoctoral researchers who report themselves as risk-averse is also disturbing (Mervis, 2004a), as they adhere to restrictive definitions of “success.” Many postdocs feel that they need to author one of more papers in high-impact journals in order to have any chance at a desirable position in academia or in receiving grant funding. Both NIH and academic institutions have contributed to the growing expectations that researchers must meet. Even new investigators are held to the standards of previously funded investigators who are required to
produce preliminary results that show they can perform a particular set of experiments at a high-quality level. As a consequence, new investigators who have not had time to “establish” themselves in the specific area they wish to explore—despite perhaps brilliant postdoctoral work—are excluded from funding. Also generally excluded are scientists who may introduce new ideas and conduct important research—but do so outside of the tenure track. Talented scientists lost from the research community are a critical loss for the scientific enterprise. This chapter considers the transition from postdoctoral work to independence, whether as a tenure-track principal investigator or as a research faculty member or staff scientist, and makes recommendations for easing that transition.
CAREER TRANSITION RESEARCH GRANTS
The transition from postdoctoral researcher to independent scientist is perhaps the most difficult step in a research scientist’s career (National Postdoctoral Association, 2003). Yet, very few funding opportunities assist in crossing that bridge to independence, provided by either the federal government or private foundations (see Box 5-1 for an example in the private sector). By comparison, the NIH supports tens of thousands of postdoctoral fellows each year. This ratio is in need of adjustment, consistent with the recommendation of a previous NRC committee (NRC, 1998) and the National Postdoctoral Association (2003). The present committee reiterates the call for enhancing the number of “career-transition” grants for senior postdoctoral fellows with a well-defined program to help the most promising researchers make a transition to independent research and independent careers.
5.1 NIH should establish a program to promote the conduct of innovative research by scientists transitioning into their first independent positions. These research grants, to replace the collection of K22 awards, would provide sufficient funding and resources for promising scientists to initiate an independent research program and allow for increased risk-taking during the final phase of their mentored postdoctoral training and during the initial phase of their independent research effort. The program should make 200 grants annually of $500,000 each, payable over 5 years.
Although the current K22 serves a career-transition function, it faces significant challenges that limits its effectiveness (see Chapter 2 for an introduction to the K22 award). Listed below are characteristics of this proposed program that the committee feels would contribute to the success in fostering the independence of new investigators:
The Burroughs Wellcome Fund (BWF) offers a career transition award in biomedical sciences,a the initial structure of which was developed by the Markey Foundation. The bulk of BWF’s funding is directed to this program: to date it has given 195 awards for a $90 million investment in the careers of young scientists. The program is very competitive—of the 175 to 200 applications received per year, it funds only 8 to 10 percent.
The goal of this $500,000, 5-year award is to help postdoctoral fellows obtain faculty positions and achieve research independence. The award provides money for salary and research—up to 2 years of postdoctoral support and the balance as faculty support. An investigator may hold other concurrent awards. Recipients are allowed no-cost extensions of unused money through the tenure review process and beyond.
Individuals at BWF track the progress of the program, examining the faculty position itself, the independence of the scientist, and the institution (Pion and Ionescu-Pioggia, 2003). The tracking data allow BWF to determine if the program makes a difference in the quality of science for those individuals who were funded compared with those who were not. Eighty-four percent of awardees believe the award helped them develop an independent research program, and 69 percent believe the award allowed them to pursue risky or novel research.
Of the incumbent awardees eligible for tenure-track positions, 98 percent currently have them, with many at the top NIH-funded institutions. The mean amount of time from the last doctoral degree to the faculty position is 6 years, and the average age at the first faculty appointment is 35. The awardees receive start-up funding at levels at or above national averages. The average age for receiving an initial R01 is 36. Of 33 awardees in the two early BWF classes (1996 and 1997), 12 are now associate professors, 1 is a full professor, and 10 are assistant professors. Approximately 60 percent of career awardees receive their degrees from institutions ranked in the top 25 institutions based on NIH funding. On average, awardees are 33 years old (slightly older for MDs) and have completed 41 months (slightly less for MDs) of postdoctoral work at the time of award.
In addition to formal studies of recipients (Pion and Ionescu-Pioggia, 2003), BWF conducts an annual survey and encourages feedback from the awardees. This feedback has helped change the structure of the award over time.
Provide postdoctoral training support for a maximum of 2 years for the awardee to develop an independent research program and 3 years of support once a fully independent research position has been obtained. Individuals would apply for these career transition research grants during years 1 through 3 of their postdoctoral appointment and would, of course, adhere to the maximum 5-year tenure in postdoctoral positions (see Recommendation 4.1). Individuals could apply without a sponsoring institution while in a “mentored” research position. Those awardees not receiving an independent research position at the end of the postdoctoral period could reserve the independent support for 1 additional year until an independent research position is obtained.
The award would support awardees to continue in a mentored position for approximately 1 year following receipt of the award, although no minimum mentored postdoctoral period is required. As the purpose of the proposed award is to facilitate transitions to independent awards, individuals already accepted into independent positions would be expected to apply for the new investigator R01s described in Chapter 6.
The award would support the transition to an independent position that is either tenure-track or considered career-path (e.g., research faculty or staff scientist), serving as the first research grant in these positions.
The award would include sufficient resources to provide at least partial salary support (at competitive salary levels) and funds for research and career development activities (e.g., participation in scientific conferences and in career development workshops on laboratory management or similar topics).
Flexibility should be incorporated into the award conditions regarding time spent on research in the first independent position to comply with a hiring institution’s policies for new faculty or staff (e.g., a minimum of 50 percent time spent in research to allow time for teaching or clinical responsibilities rather than the greater research time commitment now required).
The award should have uniform requirements and conditions across all NIH institutes. These grants would replace the current collection of K22 awards, which differ from institute to institute. They should neither limit the award to NIH intramural candidates nor require that the postdoctoral training phase be carried out at an NIH intramural laboratory.
The award would allow individuals from other disciplines to engage in mentored research in the biomedical sciences before setting off on independent careers. For example, it would be appropriate for those with a background in chemistry, physics, engineering, computer science, mathematics, psychology, or other disciplines to apply for these career transi-
tion research grants with the mentored period conducted in a biomedical research group.
The receipt of such a grant should not prohibit the investigator from applying for and receiving additional R01s—on an equal basis with everyone else.
The funds should be drawn from a separate account to eliminate competition between new and previously funded investigators. As a research award, it should allow for full indirect cost recovery during the independent second phase.
NIH is encouraged to provide staff to guide applicants through the application and award process. The staff should encourage applicants to submit plans for higher-risk research and provide study sections with explicit guidance for these projects during scientific review.
Although some NIH institutes use the current K22 mechanism in ways consistent with the above proposed guidelines, treatment varies greatly.1 For example, of the 12 institutes that had active K22 awards in 2003, only four did not require that the applicant spend the early years of the award (anywhere from 18 months to 3 years) as an intramural postdoctoral researcher. The requirement for intramural postdoctoral training may be unattractive to many senior postdoctoral candidates who need more flexibility in their career path. For some institutes, a tenure-track position itself is required rather than the possibility of an equivalent position, a condition that does not match well with the current statistics on academic employment patterns. One NIH institute restricted eligibility to its own NRSA trainees and fellows (or other NRSA recipients working in relevant areas) and two others allow individuals to apply for only the faculty portion of the award.
Others have suggested this kind of award, including the National Research Council (1998) and National Postdoctoral Association (2003); it draws upon the strengths of existing award programs including the existing NIH K-series awards, the Burroughs Wellcome Fund Career Awards in the Biomedical Sciences (see Box 5-1), the discontinued Markey Scholar Awards, and the National Multiple Sclerosis Society Career Transition Fellowship. Although the spirit of the career transition award is similar to previous recommendations (especially NRC, 1998), the specificity of implementation seeks to address the possible challenges with the K22 programs as currently configured.
Anecdotal and limited empirical data, including discussion at the workshop and the BWF outcome study (Pion and Ionescu-Pioggia, 2003), suggest that these career transition programs may assist in the development of an independent research program by:
providing the ability to develop independent research while still in postdoctoral training;
facilitating movement into tenure-track or independent research career paths (e.g., prestige of award and funds);
providing stable resources and protected time to establish a laboratory;
enhancing the ability to pursue novel directions and more risky research avenues; and
obtaining preliminary data and testing the feasibility of research study designs and/or plans for future grant applications.
The importance of the proposed new mechanism is its potential to increase the possibility of achieving independence and promoting risk-taking. For example, it might help recipients to obtain a first faculty position—even causing the creation of positions. The additional funds necessary for this program should be modest if NIH reconfigures the current K22 programs in accordance with the above recommendation.
The committee recognizes that this proposal faces numerous challenges. First, even with extramural support from NIH, sufficient resources are still required by the hiring institutions and mentors to enact this program (e.g., laboratory space, equipment). In addition, questions remain as to whether the indirect cost rate typically assigned to career development awards is appropriate here. As these awards support the first years of an independent position, at least those portions of the awards should include recovery of full indirect costs, consistent with R01 and other research awards. Finally, an expansion in NIH career transition awards may persuade private foundations to reduce their contribution in this area.
The committee recommends that NIH award 200 5-year grants annually of $500,000 each. These grants would only be one pathway to independence and would only affect a relatively small number of eligible individuals in postdoctoral positions. But the 200 awardees each year could represent a significant percentage of those entering faculty positions; for instance, Federation of American Societies for Experimental Biology (FASEB) President Paul Kincade suggested that there are about 800 tenure-track jobs filled in the biomedical sciences each year.2 Thus, the pro-
gram would support about one-quarter of those entering biomedical faculty positions each year. Other chapters in this report address a number of other approaches to career transitions; no single program provides a global solution to the problem. This award amount and tenure is modeled on that of the BWF Career Award.
Ongoing evaluation and assessment are critical for calibrating the number of awards. Systematic assessment of previous K22 recipients, for example, can identify whether the intended outcomes have occurred. In particular, a sample of K22 awardees, their faculty mentors, and institutional administrators should be interviewed to garner feedback that could be used to improve the program. This assessment should also involve some reasonably simple data collection (e.g., tracking of grantees as to the type of independent position) and use of IMPAC II3 data (how many applied for and received subsequent research project grant funding). Ideally, this effort should be carried out in collaboration with other agencies and private foundations that have similar programs in order to obtain comparable data on a core set of outcomes. Additional data collection would be necessary to assess other desired outcomes (e.g., “innovation” or other external research support). It also would be useful to correlate outcomes with the different award conditions (e.g., intramural versus extramural postdoctoral training) to examine the relationship between outcomes and award characteristics.
More rigorous assessment of expected outcomes should be conducted for recipients of the proposed career transition awards, including the use of appropriate comparison groups if possible. Such planning “up front” on data needs and useful types of comparison groups significantly increases the likelihood of obtaining meaningful information on outcomes and on the important variables related to these outcomes.
5.2 NIH needs to develop enhanced data collection systems on staff scientist and other non-tenure-track researchers to include all NIH-supported researchers, regardless of specific funding mechanism. This will allow NIH to track the effectiveness of its programs and thereby make more informed programmatic decisions.
It is essential that such data collection be appropriately disaggregated to detect any differences between demographic and other groups.
It is important to collect data on individuals in non-tenure track staff scientist research positions at academic institutions. Because of its experience in research on scientific and engineering personnel and the overall research enterprise, such data collection efforts may be best coordinated by the National Science Foundation (NSF) and funded by NSF, NIH, and other federal agencies that sponsor scientific research. Important questions to answer include:
Who is in “other academic” category from the Survey of Doctorate Recipients?
How long do appointments last?
To what extent are these positions “stepping stones” to faculty and other independent employment positions?
What are the current models of laboratory space assignment for non-tenure-track individuals?
Are individuals in these positions able to apply for independent NIH grants? To what extent do individuals in these positions apply for NIH grants? Of those who apply, what are their success rates?
How do schools handle start-up costs for research assistant professors/staff scientists?
To what extent are scientists on “soft money” able or allowed to train graduate students?
In sum, we need a greater understanding of the factors that influence a transition to a successful career, either as a tenured faculty investigator or as a staff scientist involved in large-scale collaborative efforts or independent exploration. This need will become greater if the career transition program recommended in this chapter is implemented.
The next chapter focuses on mechanisms to facilitate the establishment of stable research programs.