1 Federation of American Societies for Experimental Biology
, 9650 Rockville Pike, Bethesda, MD
20814, USA. (hgarrison@mail.faseb.org)
2 Immunobiology Program, Oklahoma Medical Research Foundation
, 825 NE 13th Street, Oklahoma
City, OK 73104, USA. (Kincade@omrf.ouhsc.edu
)
Are appropriate numbers of scientists being trained for research
in immunology? Available data suggest that supply is not yet outstripping
opportunities. The form of those opportunities, though, should change.
There has never been a more exciting time to conduct immunology experiments
or a more disquieting time for immunology trainees. Extraordinary advances
in technology allow rapid erosion of once intractable issues while confusion
surrounds career prospects. Immunology students who are better trained and
more talented than ever before are told that permanent jobs are scarce. Thoughtful
studies have concluded that biomedical training programs should be capped
while many laboratories complain of a shortage of good people. We examined
statistical data on education, employment and research opportunities to determine
whether proposals for change were appropriate for immunology. There has been
a slow expansion in the numbers of immunology Ph.D.s and postdoctoral fellows
during the last ten years with very little growth in faculty level positions
or numbers of National Institutes of Health (NIH) grant applications submitted.
In contrast, the amount of inflation-adjusted funding for experimentation
nearly doubled over the same interval. Changes are taking place in workforce
needs and career prospects in immunology that need to be closely monitored.
Although there is insufficient data to support recommendations to restrict
training, it may be time to restructure the opportunities in this field. Specifically,
we recommend recognizing the importance of staff scientists, expanding this
job category and introducing trainees to this career path at an earlier stage.
Emerging role of the staff scientist Recent reports1,
2,
3 have endeavored to improve the focus
of biomedical research training in an attempt to better match training efforts
with anticipated needs. Proposed remedies for the perceived problems may have
merit but we have not identified a serious crisis in relation to supply and
demand for immunologists; the data that support this position are summarized
below. Rather, the changing character of opportunities for new investigators
may require reconsideration of what a successful career entails4.
Many institutions now have a few "research-track" slots that
lack the "up-or-out" pressure of tenure-track positions and it
can be argued that the individuals in these jobs are the most highly trained
and stable part of our workforce. These scientists generally have advanced
degrees and often come to these positions via extended postdoctoral
training. Immunologists could be encouraged to deliberately make this career
choice at an earlier stage and without necessarily remaining in their training
sites. The median age of bioscience Ph.D. recipients is now 31 years5 and childbearing years can be quickly used up with extended postdoctoral
training. The security of a relatively permanent staff position along with
parental leave, childcare and other benefits would make this a particularly
attractive option for some individuals. Staff scientists have most of the
privileges enjoyed by principal investigators but more time for hands-on experimentation.
These positions should allow substantial career growth and need not preclude
options for teaching, publication and grant submission.
As research teams tend to grow larger, there is an increasing need for
supervision and training within them. Additionally, growing sophistication
in technology and dependence on core facilities demands highly skilled and
innovative scientists to run them. For all of these reasons, we recommend
that staff scientists be more actively recruited, adequately compensated and
recognized for their unique contributions to biomedical investigation. Institutions
may be forced to develop stable, long-term contracts and support for the new
life-blood of science. Exciting developments in immunology have increased
the demand for this talent and we should seek ways to make it a coveted career
objective.
Graduate training: what are the trends? How well does our current training system meet the needs of modern immunology?
Many successful immunologists have trained in fields such as medicine, biochemistry
and cell biology. Consequently, no database identifies all individuals preparing
for work in immunology. However, graduate enrollments in microbiology, immunology
and virology have been relatively stable for the 1990−1998 period, at
approximately 5,000 students (Table 1). Enrollments
rose in the early 1990s, reaching a slight peak in 1994 before declining gradually.
In 1999, graduate enrollments for these fields were nearly identical to those
of 19906,
7. A similar pattern is found for National Institute
of Allergy and Infectious Diseases (NIAID) predoctoral traineeships in immunology.
The number of these positions increased from 178 in 1990 to 231 in 1992 and
remained near that level for the rest of the decade.
Table 1. Trends in training, employment, and research support for immunology,
1990−1999
Doctorate awards in biological immunology and microbiology rose from 488
in 1990 to 629 in 19988. This represents an increase of 28.9%
(an average of 3.6% per year over the 8-year period). Most of this growth
occurred during the early 1990s and for most of the decade the number of new
Ph.D.s has been fairly stable, around 600 per year. Reflecting this stability,
membership in the American Association of Immunologists (AAI) grew only about
1% per year during the same interval.
Postdoctoral training Several recent studies have drawn attention to rapid changes occurring
in the postdoctoral phase of research training9. In biomedical
sciences as a whole, there has been a substantial expansion in numbers of
postdoctoral trainees and the average length of time spent as a postdoctoral
fellow has increased. We found that growth in the number of microbiology,
immunology and virology postdoctorates was modest but greater than that for
graduate students or new Ph.D.s (Table 1). In 1990,
there were 1,469 postdocs and by 1998 this number had increased by 43.6% to
2,1096,
7. It should be emphasized, however, that nearly half
of this increase occurred between 1990 and 1991, with very modest expansion
of this cadre since 1994. We can get more focused information on immunology
from numbers of postdoctoral trainee and fellow positions awarded by the NIAID
but these are restricted to US citizens and permanent residents. There were
284 of this type of NIAID-supported fellow in 1990. The number of positions
grew in the early 1990s and then contracted so that by 1997 there were 294
NIAID postdoctoral trainees. The program has remained at this level since
then. These indices suggest that there has not been explosive growth in numbers
of immunology postdocs and certainly not in ones that are permanent US residents.
Principal investigators in immunology Many immunology students desire an academic career and the opportunity
to compete for independent research support as principal investigators. This
goal is achieved by some with positions in nonprofit research institutes,
government laboratories and biotechnology companies but we expect that most
immunology principal investigators are affiliated with universities and medical
schools. Because NSF reports on science and engineering personnel do not show
faculty positions by discipline, data on immunologists is limited. AAMC reports
data on medical school faculty positions by department but there is no separate
breakout for immunology10. However, AAMC data are available
for microbiology departments that probably house a majority of immunologists.
In addition, the microbiology category roughly corresponds to the ones used
above for graduate students, new doctorate recipients and postdocs. In 1990
there were 1,697 microbiology faculty positions in medical colleges. This
number rose to 1,858 in 1999, an increase of 9.5% (averaging 1.1% growth per
year over the 9-year period). Numbers of grant applications referred to the
NIAID provide another measure of principal investigators in immunology. NIAID
grant applications fluctuated around 2,100 for most of the decade with a significant
rise only taking place in 1999.
The new investigator If the most prestigious career in immunology involves an independent laboratory,
the above data suggest that new opportunities of that type are not growing
rapidly. On the other hand, new investigators represent the life-blood of
science and steps are being taken to monitor and encourage their success.
In 1997, the NIH Working Group on New Investigators examined funding trends
and concluded that new investigators were not being well served by the R29
category of NIH grant11. The fixed amount of the awards was
viewed as insufficient and the designation thought to be inferior relative
to R01 grants. For those and other reasons, R29 grants were eliminated in
1998 with assurances that applications from first-time grant writers would
be clearly identified to review groups. In 1995, NIH made 1,413 R01 and R29
awards to previously unfunded applicants. The number of R01s rose to 1,623
in 1999. NIH has concluded that the elimination of R29 grants did not decrease
success rates for new investigators and at the same time made substantially
more money available for their studies12.
The same NIH Working Group on New Investigators estimated that a relatively
constant 9% of scientists leave the NIH grant system each year. In 1999, there
were a total of 19,599 researchers with R01 or R37 grants, which included
1,623 awards or 8.3% that were received by new investigators. It is important
to monitor such trends carefully to ensure that the sufficient infusion of
new immunologists maintains at least a steady state.
Growth in funds for immunology research Support for research in immunology expanded dramatically during the 1990s.
As a rough approximation, we examined funding trends for extramural research
grants awarded via the NIAID. The number of NIAID awards increased
from 567 in 1990 to 908 in 1999 (60.1%, an average growth of 6.7% per year).
Total funding for NIAID awards rose even faster. Adjusting for inflation,
it increased from US$112,596,000 in 1990 to US$213,834,000 in 1999 (an increase
of 89.9% with an average increase of 10.0% per year). Some of the increase
in research support may have been absorbed by more expensive methodologies.
Even so, funds for research in immunology have clearly expanded more rapidly
than the number of newly trained scientists. This imbalance may account for
anecdotal information about a personnel shortage, frequently stated as: "there
just aren't enough good students and postdoctoral trainees". In reality
there may be a rising demand for experimentalists at the bench.
Limitations of the data Recently there have been several major studies of the USA's capacity for
educating scientists2,
3,
13. Using the same survey databases,
these studies have drawn similar conclusions about what has happened to the
scientific workforce. There is far less agreement, however, about what should
be done to change the number and characteristics of the new scientists being
trained. Extrapolations of past trends are notoriously fallible as predictors
of the future and such efforts are unable to anticipate major events that
have a huge impact on the supply and demand for scientists, for example, the
launching of Sputnik, the end of the Cold War, the Tianamen Square massacre,
the emergence of the Internet and the increasingly successful effort to double
the NIH budget in five years.
The time required to collect and process data from national surveys further
diminishes the value of these data sets as guides to the future. The evaluation
of the NIH training and fellowship programs, for example, which was published
in September of 2000, had to rely on data from a 1997 survey3.
Even the most comprehensive surveys (those conducted by and for the National
Science Foundation) often provide only a partial perspective on the USA's
science resources. In the biomedical sciences, for example, there are no comprehensive
data for physician-scientists.
Additional obstacles arise from the use of different classifications of
fields and different levels of aggregation. For example, the NSF survey of
doctorate awards breaks out data for biological immunology, whereas the NSF
surveys of graduate students and postdocs report microbiology, immunology
and virology combined. Reports that rely on the NSF-sponsored Survey of Doctorate
Recipients are constrained to current findings in aggregate categories such
as "biological scientists" because the cost and response burdens
associated with this detailed, longitudinal survey of career outcomes dictate
the use of a sample of the population of scientists. There are insufficient
numbers of respondents in each detailed subfield to permit accurate estimates
at the specialized-field level.
Detailed surveys conducted at a single point in time lack comparative perspectives.
Some researchers have attempted to artificially create a longitudinal perspective
by contrasting views of older and younger scientists. But this blurs the distinctions
that arise from a number of different sources: age (career development), time
period (historical circumstances) and attrition (survival)4.
Surveys conducted by professional associations14,
15 or research
institutions10,
16 are valuable but do not cover individuals
as they move out of the sponsoring organization and into other realms of research.
Studies of specific fields or specific disciplines are hard to link with other
databases. Scientists may move from one subfield to another with relative
ease and the classification of their activities may vary even when the work
they are doing remains the same.
The NIH, NSF and National Research Council have taken significant steps
in recent years to standardize and coordinate the use of typologies for scientific
fields but more needs to be done. While preserving important time-series data
and allowing special purpose studies the discretion needed to obtain relevant
information, more emphasis needs to be put on the collection and reporting
of data in comparable categories. It would be very helpful to have a typology
of biomedical fields that captures some of the major research areas without
resorting to categories that are too small to be analytically useful.
It would also be helpful if the NIH could make more data available for
policy studies. We hope that the full implementation of the new IMPAC-II computer
system will enable researchers to analyze funding and training patterns. Adequate
funding for these activities must be forthcoming to ensure the responsible
stewardship of valuable research resources.
Conclusions While carefully focused information would allow more incisive recommendations
about careers in immunology, some trends seem obvious. At first glance it
seems that the prospects for careers as independent investigators are tightening
because the growth in full-time academic positions is even less than the modestly
increased numbers of immunology Ph.D.s and postdocs. However, of special note
is the increased availability of funds for immunology researchers and the
possibility that a workforce shortage exists for some categories of research
personnel. This would certainly be the case if there were increased opportunities
for immunologists in biotechnology and pharmaceutical companies. At the aggregate
level, we know that there has been an expansion of industrial employment of
biomedical scientists. Although statistical data on immunologists in industry
are limited, one report that uses special tabulations from the Survey of Doctorate
Recipients showed that the number of immunologists employed in industry rose
rapidly in the early 1990s and doubled between 1991 and 199517.
These findings are not supportive of a cap on immunology training. In addition,
it should be noted that an exact match between students enrolled and expected
research positions is a misguided goal. There will be some students who, as
a result of their experiences in graduate school, realize that they do not
wish to pursue a research career. Others will be found to lack the necessary
aptitudes and abilities. Then there will be those students who have plans
for careers outside the laboratory who determine that a doctorate is desirable
for job advancement, personal satisfaction or other reasons. They should not
be prevented from entering or completing advanced training in immunology.
Additionally, it is not possible to predict how much the next decade will
resemble the past. Although we can probably expect continued expansion of
research funding, the amount is not known. The need to replace the 9% of immunologists
that retire or leave research each year is certain. Additional talented scientists
are needed to harvest the opportunities created by advanced technology, increased
research resources and the need to protect against emerging infections. It
is also clear that the applications for immunology in cancer therapy, transplantation
and other important areas will expand as we learn more about how the immune
system can be manipulated.
This would seem to be an ideal time to expand research faculties at medical
schools and private research institutes, even if teaching needs have remained
constant. Increased utilization of well compensated and fully appreciated
staff scientists would provide one means to capitalize on emerging opportunities
in immunology. Another goal must be to attract even more talented students
to a field that remains important, exciting and fresh.
National Research Council. Reshaping the Graduate Education of Scientists and Engineers (National Academy Press, Washington DC, 1995).
National Research Council. Trends in the Early Careers of Life Scientists (National Academy Press, Washington DC, 1998).
National Research Council. Addressing the Nation's Changing Needs for Biomedical and Behavioral Scientists (National Academy Press, Washington DC, 2000).
Marincola, E. & Solomon, F. Mol. Biol. Cell9, 3003−3006 (1998). | PubMed | ISI | ChemPort |
Sanderson, A., Dugoni, B., Hoffer, T. & Selfa, L. Doctorate Recipients from United States Universities: Summary Report 1998 , 102 (National Opinion Research Center, Chicago, 1999)
National Science Foundation, Division of Science Resources Studies. Graduate Students and Postdoctorates in Science and Engineering: Fall 1997 NSF 99−325 (National Science Foundation, Arlington, 1999)
National Science Foundation, Division of Science Resources Studies. Graduate Students and Postdoctorates in Science and Engineering: Fall 1998 NSF 00−322 (National Science Foundation, Arlington, 2000).
Hill, S. T. Science and Engineering Doctorate Awards: 1998 NSF 00−304 (National Science Foundation, Arlington, 2000).
National Research Council, Committee on Science, Engineering and Public Policy. Enhancing the Postdoctoral Experience for Scientists and Engineers: A guide for postdoctoral scholars, advisors, institutions, funding organizations, and disciplinary societies (National Academy Press, Washington DC, 2000).
Association of American Medical Colleges. AAMC DATA Book: Statistical information related to medical schools and teaching hospitals (AAMC, Washington DC, 2000).
National Institutes of Health. Report of the NIH Working Group on New Investigators (NIH, Bethesda, 1997).
Seto, B. FASEB News33, 5−6 (2000).
Federation of American Societies for Experimental Biology. Graduate Education: Consensus Conference Report. (FASEB, Bethesda, 1997).
American Chemical Society, Department of Career Services. Current Trends: Chemists' employment 5/2000 (ACS, Washington DC, 2000).
American Institute of Physics, Education and Employment Statistics Division. 1998 Initial Employment Report: Follow-up of 1997 physics and astronomy degree recipients (AIP, College Park, 1999).
Ammons, S. W. & Kelly, D. E. Acad. Med.72, 820−830 (1997). | PubMed | ISI |
National Academy of Sciences, Committee on Science, Engineering and Public Policy. Experiments in International Benchmarking of US Research Fields , 249−306 (National Academy Press, Washington DC, 2000).
Biomedical Research and Development Price Index (http://www4.od.nih.gov/ofm/brdpi/1979-96.stm)
Acknowledgments We thank T. Zemlo and M. Hernandez for their comments and suggestions and
K. Kang for assistance in obtaining NSF survey data. Opinions expressed in
this article do not necessarily reflect the views of the institutions and
organizations with whom the authors are affiliated.
