Elias Zerhouni wants to promote fresh pathways to scientific success.
As the economics of science in the United States changes, so too does the scientific enterprise. There are more authors per paper, more graduate schools, more postdocs. Yet there are relatively few entry-level faculty positions. Meanwhile, in areas such as biomedical research, large-scale investment continues to increase significantly.
Nowhere is the bittersweet mix of bulging budgets and postdoc proliferation more apparent than in the life sciences, where the prospective investigator faces an often frustrating journey. A serious, system-wide post-postdoc bottleneck is changing the composition of temporary, permanent and student workers in the lab.
"The operating model of science has changed," says Elias Zerhouni, director of the National Institutes of Health (NIH). "It's not just an NIH issue. I think it's really a systemic issue." As research teams grow larger and require more complex skill sets, they need more postdocs and graduate students, who in turn take longer to complete training.
A couple of decades ago, prospective scientists could move easily from one lab to another. Zerhouni notes that scientists such as James Watson, co-discoverer of DNA's double-helix structure, and Marshall Nirenberg, who helped to decipher the genetic code, made major discoveries while still in their 20s and 30s — in today's terms, both might have received their Nobel prizes before they received their first NIH grant. "What I always used to say is you don't become a thoroughbred by longevity," Zerhouni says.
Shirley Tilghman (right) believes the balance between postdocs and the number of permanent positions in the United States needs to be redressed.
Part of the problem, he suggests, is that the system mainly recognizes those scientists who succeed in the well-trodden career pathway towards being a faculty member. Finding and promoting additional pathways may be necessary, says Zerhouni. "I look at it as probably one of the most important long-term priorities for the NIH director to focus his energies on," he says, adding that he has started to "put a process in place", although he hesitates to discuss specific solutions as yet.
Other scientists, economists and policy-makers have also offered analyses and possible remedies. Reports from the National Academy of Sciences (NAS) in 1994, 1998 and 2000 recognized the flaws in the system, particularly the plight of postdocs seeking their first faculty positions — the average first-time grantee is now over 35 years old, largely because of a longer training period. "I haven't seen anything to say that those trends have changed," says Walter Schaffer, acting director of the NIH's office of extramural programmes.
TREND SETTERS
The 1998 NAS committee attempted to offer solutions. Universities and researcher institutions, said its report, should not continue to expand enrolment in graduate programmes or develop new ones unless they are directed at a specific need (for example, in a burgeoning subdiscipline such as bioinformatics). It was this recommendation that "got the most heat", says the committee's chair, Princeton University president Shirley Tilghman, because it challenged the status quo.
Status of academic science and engineering PhDs who earned their PhD 5–7 years earlier.
The report in 1998 also advised federal agencies to fund more graduate students through training grants — which allow better scrutiny of the quality of training offered — rather than through principal investigators' research grants. The funding of graduate students and postdocs with research grants has caused the number of trainees in the lab to expand out of proportion with the number of permanent jobs they subsequently seek, says Tilghman. The trend has become more pronounced over the past five years, as successive US budgets have seen the NIH's budget double from its 1998 level. According to Schaffer, the number of students — about 17,000 — supported by training grants and fellowships has changed little during the past decade.
Also in 1998, the American Society for Cell Biology conducted a survey of over 2,400 members at various points in their careers. Later, the society, in collaboration with Harvard economist Richard Freeman, did 'life histories' of some 30 well-established labs. They obtained some discouraging results.
Older scientists reported that it is now harder for young scientists to secure a faculty position. Many postdocs complained that today's emphasis is on research productivity over training and that their work differs little from that of graduate students. "Postdocs said: 'I don't feel like I'm in training, I feel like I'm a pair of hands'," recalls co-author Frank Solomon, a biology professor at the Massachusetts Institute of Technology. He also found that prospective postgrads sought the advice of dissatisfied postdocs and graduate students, suggesting, he says, that a dissatisfied group of people might "nucleate greater dissatisfaction".
Just over a year ago, a research network of labour economists was formed to discuss analyses of the science and engineering labour markets, sponsored by the Alfred P. Sloan Foundation. Ideally, says Freeman, it will offer independent analyses of government, academic and industry institutions, and will share its information with professional societies and policy-makers. A recent report, he says, shows that, as in the United States in general, the rich have got richer and the poor poorer: although science has become more collaborative and research groups have grown, economic inequality among team members has increased.
Tuning the US engine of biomedical science will be tricky. The NIH has increased stipends, not only to raise postdocs' pay but to help stem their proliferation by increasing their cost to a principal investigator. The agency has also started a 'transition award' programme that helps cream-of-the-crop postdocs secure their first faculty position with start-up funding (see 'Making the transition', left).
Some universities, encouraged by postdoctoral groups — including the National Postdoctoral Association, which met for the first time this month — have also taken action. Stanford's medical school, which has three-quarters of the university's postdocs, has been one of the most proactive. Last September, it raised the minimum postdoc pay to $35,000, limited fellowships to five years (at which point the postdoc leaves or becomes a permanent employee), improved health benefits, offered more subsidized housing and instituted a mentoring programme to emphasize the training element of the postdoctoral experience. The agreement arose from "a really good dialogue" among faculty members, the postdocs and the administration, says William Nelson, the medical school's senior associate dean of graduate and postdoctoral education.
"Some faculty members have complained loudly," admits Nelson. Principal investigators in the middle of multiyear grants were particularly displeased by the unexpected funding changes; postdoc-heavy chemistry labs were hit hard. "My comment to that has been: 'I'm sorry but we have to do this somehow'," Nelson says.
Science and engineering PhDs hired into faculty and postdoc positions at research universities and other academic institutions, 1973–99.
But improving the postdoc situation may be the easy part. "A lot of campuses are trying to address those issues," says Paula Stephan, a labour economist at Georgia State University and a committee member for the 1998 NAS report. "What I don't see people trying to address is the harder problem, and that's what your job prospects are when you leave your postdoc."
Solomon and others suggest establishing a new class of permanent bench-scientist employees and raising the bar for prospective PhD students. "I'm a big believer in questioning the role of the postdoc position," says Solomon. Perhaps, he offers, the postdoc fellowship will be used for the select few as a transition to independent research or as a way to switch disciplines.
Where would these permanent bench scientists come from? Solomon expects that many students would forgo thoughts of a faculty position in favour of a standard working day, with no obligation to teach or deliver scientific talks. "There are a lot of people who would love to have jobs doing what they love to do, which is experiments," says Solomon.
Trends in the Early Careers of Life Scientists, NRC report 1998
http://www.nap.edu/catalog/6244.html?onpi_newsdoc091098
Addressing the Nation's Changing Needs for Biomedical and Behavioral Scientists, NRC report 2000