Despite dramatic increases in the National Institutes of Health (NIH) budget and a growing need for skilled scientists to carry out academic biomedical research, there has not been a corresponding increase in the number of new US doctorates or an increase in the percentage of these graduates taking postdoctoral positions.

Both the frequency and length of postdoc appointments for US biomedical scientists have continued to decline, while the number of foreign-born postdocs has been increasing. And while employment of biomedical scientists in industry has grown, there have been only modest increases in tenured or tenure-track academic jobs.

These findings about opportunities for biomedical scientists come from an analysis of recent survey data conducted by Paul W. Kincade, head of the immunobiology and cancer research program at the Oklahoma Medical Research Foundation in Oklahoma City, along with Howard A. Garrison of the Federation of American Societies for Experimental Biology (FASEB) and Susan A. Gerbi of Brown University. Their study was published in the October 2 online edition of The FASEB Journal.

"These results are surprising, since everybody thinks that there's a backlog in postdocs and that's a holding pattern for our trainees," said Kincade. "That doesn't seem to be the case anymore. In fact, a much lower percentage [of PhD graduates] than we thought are doing postdocs."

"It's been steady-state for several important parts of the pipeline," Kincade told The Scientist. "The number of students, the number of PhDs granted in biomedical sciences, and the number of postdoc fellows that we train have all been stable. Any expansion in the total number of postdocs has come from non–US-trained scientists," he said.

The implications are sobering. "If you thought the growth in research and development and in the NIH budget meant that there was a greater need for scientists, then you would have to say our domestic training programs do not match the growth," Kincade added.

From 1970 to 1995, doctorate programs were taking longer to complete, postdoc appointments were being held for longer time periods, and the proportion of scientists employed in academic careers had contracted. Opportunities for life scientists were far from robust.

But starting in the mid-1990s, funding for biomedical research increased rapidly, exemplified by the growth at NIH, whose budget doubled from $13.6 billion in fiscal year (FY) 1998 to $27.3 billion in FY 2003. However, the number of biological science graduate students at doctoral-granting institutions in 2001 was 54,099, a similar number as reported for 1995. The number of these students employed as research assistants in 2001 totaled 19,146, also close to the number in 1995.

The length of time needed to earn a PhD in the biomedical sciences in the United States increased from 6.0 years in 1980 to 6.9 years in 1993 and has remained stable since then, the report states.

Since personnel costs have historically averaged about 60% of research grant budgets, one would assume that demand for research personnel had increased with the growth in the NIH budget. But the number of graduate students has not increased. While it is true that the postdoctoral pool increased from 14,907 in 1996 to 16,913 in 2001, that growth is merely a continuation of the trend from the previous 5-year period and does not reflect any acceleration due to funding increases. Furthermore, almost all of that growth is due to increasing numbers of foreign postdocs, the report states.

Traditionally, an academic career has been the goal of most entering biomedical PhD students. But since 1997, there has been very slow growth in the number of biomedical scientists holding tenured academic positions and no increase at all in the number of tenure-track positions. In 1981, 34.3% of biomedical PhDs with 5 to 6 years experience held tenured or tenure-track positions. By 2001, only 14.4% of the PhDs were in such positions 5 to 6 years after receiving their degrees.

The paper raises other policy questions: Because an increasing number of biomedical PhDs are going into industry, does this mean doctoral training programs should be adjusted to teach students how to work effectively as part of a research team as opposed to being a good principal investigator?

And with the decline in the number of US students becoming postdocs, "Will we continue to expand the use of postdoctorals who receive their PhDs in other countries?" the authors ask. "Shall academia copy the paradigm used in industry of employing 'staff scientists' to carry out the research?"

If so, the authors conclude, salary levels need to become more commensurate with training levels in order to recognize staff scientist positions as "legitimate career outcomes." Stipends for predoctoral and postdoctoral appointments should also be increased.

Kincade said the research paper was not intended to discourage talented people from entering biomedical science. "We want to communicate honestly and with good information, so trainees know what their career options are," he said. "We don't want to put people off science. There has never been a more exciting time to be in research. But we need to give the students the best information we have about the opportunities that are available."