Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 23
CHAPTER THREE
BASIC BIOMEDICAL SCIENCES PERSONNEL
Exciting developments in the basic biomedical sciences
continue to attract talented individuals to research careers.
Although the National Institutes of Health (NIH) provide
both extramural and intramural support to advance research
and maintain a pool of skilled scientists, it is the National
Research Service Awards (NRSA) program that provides
the most promising young scientists with the funds they
need to complete their training while pursuing research top-
ics of interest to them and the nation.
Like previous NRC committees formed to address the
future direction of the NRSA program, we have considered
the appropriate level and mix of predoctoral and
postdoctoral support in the basic biomedical sciences given
advances in basic research and changing employment pat-
terns for scientists in component fields (see Appendix B for
a field taxonomy). This has been a challenging task for
three reasons. First, changes at the NIH may well favor a
shift toward more basic research in the health sciences.
However, unless resources are made available to basic bio-
medical scientists to pursue those new directions, the con-
nection between training and research will be broken. The
continued success of the NRSA program depends on its
ability to attract highly qualified and promising students to
enter training and pursue careers in research. The interest
of potential trainees in such a career and their ability to
pursue it depends, in turn, on continued federal commit-
ment to support health-related research as an important na-
tional need.
A second, possibly related, challenge we confronted in-
volves the interpretation of the current and future market
for basic biomedical scientists. We realize much has been
written recently about He difficulties some young biomedi-
cal scientists have encountered in locating positions in re-
search settings and/or securing research support. (Indeed, a
recent report of the National Research Council' s Commis
23
sion on Life Sciences [NRC, 1994] addressed He topic of
obtaining NIH support.) We believe that for some young
scientists the market has become sluggish. We cannot help
but observe, however, that the unemployment rate of basic
biomedical scientists is estimated to be about 2 percent or
less and has not changed significantly in the past two de-
cades. We attribute this finding to He fact that not all Ph.D.-
level scientists pursue careers in academic research settings;
some work in government, industry, or schools. Almost all
are essential, however, to the support of the infrastructure
of the nation' s research and training enterprise in the basic
biomedical sciences.
Our Panel on Estimation Procedures has persuaded us
that the mathematical models of supply and demand found
in previous reports on the NRSA program should be aban-
doned in favor of an analysis of the supply of scientists and
a separate look at selected indicators of market conditions.
As noted in the previous chapter (Chapter 2) mathematical
models of supply and demand have a number of deficien-
cies, among them a lag in information about He most recent
employment prospects. While the new techniques devel-
oped by the panel have not solved the problem of having
up-to-date market information, the analyses inherent in the
indicators of market conditions coupled with the use of
multistate life table analyses of changes in supply represent
at least a partial solution. These analyses do not offer a
specific assessment of future demand. Because of this, the
committee chose to develop alternative assumptions about
the growth of future demand and to examine the implica-
tions of these assumptions for the number of degrees in
biomedical sciences that would be required to meet these
assumed rates of growth. The product of that assessment
may be found in the pages that follow.
On He basis of that analysis and our subsequent delib-
erations, we conclude that He national biomedical effort
OCR for page 24
MEETING THE NATION'S NEEDS FOR BIOMEDICAL AND BEHAVIORAL SCIENTISTS
continues to benefit from the steady addition of men and
women (including minorities) to the basic biomedical sci-
ences work force. They appear to be employed produc-
tively, although not all may be working in research labora-
tories. The best predictions for economic activity and
research and development (R&D) funding in the near future
suggest, however, that demand for basic biomedical scien-
tists will grow slowly. Our primary concern at this time,
therefore, is to maintain the supply of highly skilled scien-
tists required to
· keep our nation in a lead position in all areas of basic
and applied biomedical sciences,
· respond rapidly and effectively in combating new
problems in human health and disease, and
· ensure the efficient transfer of new knowledge and
technology into developing areas of clinical promise and
industrial opportunity.
Meeting the national need for highly qualified and pro-
ductive biomedical scientists depends on attracting suffi-
cient numbers of the best and brightest high school and
college students into scientific careers at the graduate level,
which in turn depends on facilitating access of all qualified
students to this education path. The NRSA program clearly
has an important role to play in Hat effort.
In considering the future role of the NRSA program in
meeting national needs for bioscientists, we confronted our
third challenge: the increasing attractiveness of stipends for
research training from sources other than the NRSA pro-
gram. We learned from participants at the public hearing
(May 1993, Appendix C) that the NRSA program generally
remains effective in recruiting individuals into the research
training path and launching them into research careers.
However, for reasons largely related to years of stagnant
growth in stipend support, the NRSA is no longer competi-
tive with other mechanisms of training support, which have
higher stipends and more flexibility. The committee con-
sidered these issues and concluded that high priority must
be given to restoring appropriate stipend support through
the NRSA program even at He expense of overall growth in
the total number of awards in the basic biomedical sciences
over the next few years. Thus, our recommendations for
the future direction of the NRSA program reflect our deep
conviction that the NRSA program must continue to play a
significant role in the national biomedical research effort
and that this will require prompt attention to issues of sti-
pend size and flexibility. In the pages that follow we shall
address each of these issues.
ADVANCES IN RESEARCH
Innovations in basic science and in technology are inex-
tricably intertwined and inseparable. Advances in basic sci
24
once lead to development of new technologies that, on the
one hand, give rise to new therapies and industrial applica-
tions, and on the other hand, give rise to fresh advances in
science from which, in turn, emerge additional new tech-
nologies. Thus, advances of the 1960s and early 1970s in
biochemistry, microbiology, and genetics provided the in-
formation required for development of recombinant DNA
technology, which has, in turn, provided the basis for revo-
lutionary new insights in many fields of biology and medi-
cine and has given rise to novel diagnostic and therapeutic
modalities and to an entirely new biotechnology industry.
More recently, He discovery of extremely thermophilic bac-
teria that live in hot springs and deep ocean vents and study
of their biochemistry, together with other advances in mo-
lecular biology and recombinant DNA technology, has led
to a new refinement in molecular biological analyses, the
polymerase chain reaction (PCR). PCR has proven to be an
extraordinarily powerful tool in basic biomedical and clini-
cal research and has important applications in areas of medi-
cal biotechnology, such as clinical diagnostics. Indeed, it is
noteworthy that Dr. Kary Mullis shared the 1993 Nobel
Prize in Chemistry for his work in development of PCR.
A second new technology win major applications in
clinical medicine and industry, as well as basic biomedical
research, arose from immunologists' need to understand the
nature of the immune response and the mechanisms con-
trolling the formation of antibodies. The so-called mono-
clonal antibody technique provides a method of exquisite
specificity and sensitivity for identifying and purifying any
molecule that is capable of eliciting an immune antibody
response. Current medical applications include, for ex-
ample, rapid and precise identification and analysis of
pathogenic microorganisms and tumor cells and purifica-
tion of specific types of immune cells for diagnostic and
therapeutic purposes.
These and other innovative technologies, in conjunction
with advances in microchemistry, instrumentation, and,
most notably, computer science, have fueled a continuing
explosion of understanding in many fields of basic biomedi-
cal sciences. These include mechanisms of regulation of
gene expression in growth, differentiation, and develop-
ment; biochemical mechanisms regulating normal and ab-
normal cell growth and multiplication; mechanisms
whereby the immune system recognizes, processes, and re-
sponds to antigenic stimuli; protein structure at atomic reso-
lution, the relationship of structure to specific protein func-
tion, and the principles of protein design; mechanisms of
nervous system development and function, including mo-
lecular bases of learning and memory; elucidation of the
human genome and its expression; and development of new
food crops to feed the world's population.
Advances over the past decade in areas of basic bio-
medical science such as Hose cited above have profound
OCR for page 25
BASIC BIOMEDICAL SCIENCES PERSON19EL
implications for understanding the genesis of major human
diseases and for future development of effective means of
prevention, therapy, and cure. Efforts to map the human
genome and identify mutant genes responsible for heritable
diseases are progressing rapidly under the aegis of the Hu-
man Genome Project and form the basis of the newly emerg-
ing field of gene therapy. It is worth noting that studies on
the human genome depend heavily on technologies which
derive from concurrent work on mouse genetics and embry-
ology, such as construction of transgenic mice. These tech-
nologies have also been important for discovery and func-
tional analysis of so-called oncogenes and tumor suppressor
genes. The protein products of those genes normally serve
as important regulators of gene expression or cellular
growth and multiplication but can, through mutation or ab-
errant expression, trigger unregulated cancerous growth of
cells. The rapidly increasing understanding of the compli-
cated biochemical and genetic means by which control of
cell growth and multiplication is achieved and lost-are
providing major new insights into the root causes of cancer
and possible strategies for prevention and therapy. During
the past decade, structural biology (specifically, determina-
tion of three-dimensional macromolecular structure at
atomic resolution) has undergone a renaissance win the ad-
vent of recombinant DNA-based methods for production of
large amounts of pure proteins and nucleic acids and con-
comitant improvements in instrumentation and computa-
tional methods. This information is crucial, not only for
understanding functional interactions of normal and abnor-
mal proteins, but also for determining how drugs interact
with their target proteins and for rational design of new
agents win improved therapeutic properties.
ASSESSMENT OF TO CURRENT MARE~T
FOR BASIC BIOMEDICAL SCIENTISTS
Employment conditions for biomedical scientists were
relatively robust throughout the 1980s (Figure 3-1~. In re
1 00000
95000
o
ILL
y
3
c
._
D 75000
E
He
90000
85000
80000
70000
65000,
60000- i 1 i ~1
1981 1983 1985 1987 1989 1991
'
her
fir
/
/
-
FIGURE 3-1 U.S. biomedical science workforce, 1981-1991.
See Appendix Table F-3.
25
spouse to expanding opportunities in heals research, the
basic biomedical work force grew dramatically, rising from
roughly 64,000 Ph.D.s in 1981 to nearly 92,000 in 1991.
This 44 percent growth is about twice that of the total sci-
ence and engineering work force and quadruple He rate of
employment growth of He total U.S. work force.)
Accompanying this dramatic work force growth were
substantial changes in its composition. Among the notable
changes were He growing prominence of females and
Asians2 and the declining prominence of native-born male
citizens. Accompanying these changes in work force char-
acteristics were changes in the nature of employment op-
portunities. Academic employment declined as a relative
share of total employment as industrial employment grew.
As Figure 3-2 suggests, about 23 percent of the basic
biomedical science work force were women in 1991, up
from 17 percent in 1981. Almost half of the women in the
20
~ 1 ;]
1981
1 99 1
FIGURE 3-2 Fraction of the U.S. biomedical science work force
who are women, 1981 and 1991. See Appendix Table F-3.
1991 work force were younger than 40, compared with
roughly 38 per cent of the men.3
The biomedical sciences work force has also become
more racially diverse over the years, but progress has been
slow. In 1991 nearly 12 percent of He employed biomedi-
cal science Ph.D.s represented individuals from a racial mi-
nority group (Table 3-1~. In 1979 these minorities repre-
sented about 8 percent of the biomedical work force. Most
of Be growth occurred for Asians. Progress in ethnic diver-
sity is less dramatic. In 1991 Hispanics represented about 2
percent of the biomedical scientists. In 1979 the compa-
rable statistic was roughly 1 percent;.
The age distribution of the work force is an important
early-warning indicator of future replacement needs. De-
spite its rapid growth over the past decade, the biomedical
work force is aging (Figure 3-34. The median age has risen
slowly from 39 to 42 years. Based on these projections, He
1989 study concluded that annual replacement needs would
increase by 28 percent between 1987 and 1995, from 5,086
to 6,543 as a result of replacement demand (NRC, 1989~.
OCR for page 26
MEETING THE NATION' S NEEDS FOR BIOMEDICAL AND BEHAVIORAL SCIENTISTS
TABLE 3-! Racial/Ethnic Composition of the Employed Biomedical Ph.D.s: 1981 and
1991
198 1a
199 1b
,
Number Percent Number Percent
Race
TOTAL58,264100.085,275100.0
White53,00591.075,83088.9
Black7221.216561.9
AsianJPacific Islander4,4387.67,5838.9
Other (InclO Native Amencan)990.22060.2
Ethnicib
TOTAL56,950100.084,803100.0
Hispanic8181.41,2631.5
Non-Hispanic56,13298.683,54098.5
aFor those who responded in 19810 Race nonresponse was 182 in 1981 and ethnic nonresponse was 1,496.
bFor those who responded in 1991. Race nonresponse was 321 in 1991 and ethnic nonresponse was 793.
NOTE: Employed biomedical Ph.D.s are those with a biomedical Ph.D., regardless calf employment field.
Estimates are subject to sampling error. Compansons between 1991 estimates and those of earlier years should
be made with caution due to changes in survey methodology. Prior to 1991, the SDR collected data by mail
methods only. In 1991, the sunrey had both a mail component and a telephone follow-up component. In this
table, 1991 estimates are based on "mail~nly. data to maintain greater comparability with earlier years.
SOURCE: NRC, Survey of Doctorate Recipients. (Biennial)
· Males
Females
44
43
42
41
cat 40
~ 39
._
38
37
36
35
34
~Total
:
1981 1983 1985 1987 1989 1991
FIGURE 3-3 Median age of U.S. biomedical science work force
gender, 1981-1991. See Appendix Table F-3.
26
· U.S. Citizen
100
95
~ 90
,~ 85
80
75 - t 1 1 ~1
1981 1983 1985 1987 1989 1991
FIGURE 3-4 Citizenship status of employed biomedical science
Ph.D.s, 1981-1991. See Appendix Table F-4.
{:3 U.S. Native-born
citizen
1
OCR for page 27
BASIC BIOMEDICAL SCIENCES PERSONNEL
Almost 95 percent of employed biomedical science Ph.D.s
were U.S. citizens in 1991 (Figure 3-4~. However, between
1981 and 1991, native-born U.S. citizens became a smaller
proportion of the total U.S. biomedical work force. That is,
in 1981 native-born U.S. citizens represented 87.5 percent
of the biomedical work force, compared with 85.4 percent
in 1991. While weak, this declining trend reflects the more
general phenomenon in science and engineering as a whole,
wherein U.S. native-born citizens represented 86.4 percent
of the Ph.D. work force in 1981 but only 82.7 percent ten
years later (NSF, 19904.
Individuals holding postdoctoral appointments are an
important component of employment in the biomedical
field. Proportionately, more women than men held
postdoctoral appointments.4 In part, the difference is be-
cause the likelihood of holding a postdoctoral position var
70
60
50
40
w
c
30 - l
20
10
O
~, ~
1981 1983 1985
Academe
X Industry
Government
HospiCiinic
1987 1989 199 1
FIGURE 3-5 Employment sector of the U.S. biomedical science
work force, 1981-1991. See Appendix Table F-5.
ies inversely with career age, and women, as more recent
participants in biomedical science, tend to be younger.
Opportunities for employment in the academic sector
have grown more slowly than have opportunities for em-
ployment in nontraditional settings (Figure 3-5~. As a re-
sult, only 54 percent of the biomedical science work force
were employed in academia in 1991 in contrast to two-thirds
in 1981. This trend reflects the changes in academic em-
ployment prospects experienced by those in many other
fields. Offsetting this trend, however, has been the dra-
matic rise in employment opportunities in industry. This
sector accounted for almost 28 percent of 1991 employ-
ment, up from almost 17 percent in 1981. The proportion
of the basic biomedical workforce employed in other sec
tors (such as government or in hospitals and clinics) re-
ma~ned about the same during that period.
Degree Production and Career Patterns
The major source of new biomedical science talent has
been our nation's university system. It is not the only source
of talent, however. Some jobs are filled by immigrants who
received their degrees in other countries. Furthermore,
some recipients of biomedical Ph.D.s are employed in other
fields, and some biomedical science jobs are filled by work-
ers with degrees in other fields.
Degree Productiorl
The most readily available source of information about
patterns of degree production is the Doctorate Records File,s
which describes degree production from U.S. universities;
the committee summarizes this information below.
A declining trend in degree production occurred between
4000 T
3750 +
, 3500;
· 4D
~ 3250 .
Z anon
2750
2500
~ 981 ~ 982 ~ 983 ~ 984 ~ 985 ~ 986 ~ 987 ~ 988 ~ 989 ~ 990 ~ 991 ~ 992
FIGURE 3-6 Biomedical science Ph.D. production, 1981-1992.
NOTE: Data limited to U.S. citizens and permanent residents.
See Appendix Table F-6.
1981 and 1985 and was followed by a relatively stronger
upward trend between 1987 and 1992. The annual number
of degrees produced in the biomedical sciences rose by 10
percent over the entire period, from about 3,400 to almost
3,800 (Figure 3-64. This rate of increase was notably slower
than the comparable rate of 31 percent for doctorates in all
fields of science and engineering (Ries and Thurgood, 1993~.
There were notable changes in the characteristics of the
degree recipients in the biomedical sciences: an increasing
fraction were female and a smaller fraction were U.S. citi-
zens. The average age of recipients increased.
Significant progress has been made in achieving gender
diversity. The number of degrees granted to women in-
creased between 1981 and 1992 by almost 60 percent (from
roughly 1,000 to about 1,600~. In 1981 women represented
29 percent of the degrees produced in biomedical sciences;
by 1992 they received 43 percent (Figure 3-7~.
Little progress has been made with respect to race and
OCR for page 28
MEETING THE NATION'S NEEDS FOR BIOMEDICAL AND BEHAVIORAL SCIENTISTS
50:
45
40
~ 35 t And
30~
25t
_/
-
,~
20 1 1 1 1 1 1 ~ i 1 1
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992
FIGURE 3-7 Fraction of biomedical science Ph.D. degrees
earned each year by women, 1981-1992. NOTE: Data limited to
U.S. citizens and permanent residents. See Appendix Table F-6.
ethnic diversity, however (Table 3-2~. When analyses are
restricted to degree recipients who are U.S. citizens or per-
manent residents, we find that whites constituted about 91
percent of 1981 degree production; in 1992 Hey represented
just over 86 percent. Roughly half of this small decline can
be accounted for by the growth in the number of degree
recipients of Asian origin. The share of degrees awarded to
Asians rose from 5.3 in 1981 to 8.4 percent in 1992.
There has also been a dramatic change in the citizenship
status of biomedical degree recipients. The percentage who
were U.S. citizens declined from 88 percent in 1981 to 69
percent in 1992 (Figure 3-8~. Similar changes are occurring
in other fields, particularly in the physical sciences and en-
gineering (NSF, 1990~. This change may ultimately be re-
flected in He citizenship characteristics of He biomedical
work force.6
loo T
95T
90
85 ~
80- .
' 75
70
65
60
55
50 - _
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992
FIGURE 3-8 Fraction of biomedical science Ph.D. degree earned
each year by U.S. citizens, 1981-1992. See Appendix Table F-7.
Career Patterns
Given the objective of the NRSA awards to produce
research scientists it is useful to have some notion of the
number of years over the course of a career that these scien-
tists remain engaged in R&D. The effectiveness of the pro-
gram will vary with this number. The Survey of Doctoral
Recipients a longitudinal survey that tracks doctorates in
He sciences, engineering and humanities biennially pro-
vides useful information on employment patterns, including
postdoctoral work. This survey has the potential for illumi-
nating career patterns of biomedical scientists. Thus, the
Panel on Estimation Procedures will examine more closely
the feasibility of estimating such patterns.
This section presents short-term indicators of market
conditions: unemployment and underemployment rates,
postdoctoral appoinanents, postgraduation commitments of
TABLE 3-2 Biomedical Ph.D. Production Over Time, by Race and Ethnicity
1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992
Total N 3293 3359 3243 3294 3126 3162 3119 3406 3429 3482 3684 3728
Mite %
3293 3359 3243 3294 3126 3162 3119 3406
91.5 91.1 90.4 90.3 89.9 89.5 88.9 89.5
88.7 88.5 86.3 86.3
Black 1.9 1.8 1.6 1.9 2.0 1.6 2.3 1.9 2.2 2.0 2.3 2.1
Hispanic 1.2 1.5 1.4 1.5 1.8 2.1 2.2 2.4 2.3 2.6 2.7 2.7
Asian 5.3 5.4 6.3 6.0 5.9 6.2 6.2 6.0 6.5 6.7 8.3 8.4
Native
American 0.2 0.2 0.2 0.3 0.4 0.6 0.3 0.2
0.3 0.1 0.3 0.5
NOTE: Cases with missing data are excluded. Data limited to U.S. citizens and permanent residents.
SOURCE: NRC, Survey of Earned Doctorates. (Annual)
28
OCR for page 29
BASIC BIOMEDICAL SCIENCES PERSONNEL
new doctorates, and relative salaries.7 No strong trends have
been discerned, although changes in postgraduation com-
mitments and starting salaries suggest that the demand for
basic biomedical scientists has been growing more slowly
than in earlier years.
Unemployment arid Underemployment
The most commonly used short-term indicator of labor
market conditions is the unemployment rate. In labor mar-
kets for highly skilled workers, however, the unemployment
rate is not as meaningful as an indicator of market condi-
tions. This is because such workers are able to find jobs
even in times of weak demand. Thus, the issue is not
whether the worker has a job, but whether the job is fully
utilizing the worker's skills. For this reason, the committee
has also compiled information on underemployment which
1.2
1
0.8
0.6
0.4
0.2
O
\
\ ~ \ ·~
1 1 1 1 1 1 1 1 1
1973 1975 1977 1979 1981 1983 1985 1987 1989 1991
· Biomedical Sciences ~ Physical Sciences
FIGURE 3-9 Unemployment rates of biomedical and physical
sciences Ph.D.s, 1973-1991. See Appendix Table F-8.
is defined to include workers who are working part time but
would prefer full-time jobs and workers who have jobs that
are outside of science and engineering and who indicate
they took these jobs because they could not find work in
science and engineering. Figures 3-9 and 3-10 summarize
these rates. Because concern has been expressed recently
about the weak state of demand in the physical sciences,
comparable rates for physical scientists are included so that
the reader can assess the relative status of biomedical sci-
ence labor markets as gauged by this indicator.
Several conclusions emerge. First, as noted above, un-
employment is not a serious problem. Rates of unemploy-
ment and underemployment generally hover around 1 per-
cent in each of the fields examined. The data in Figure 3-9
contrast strikingly with the rate for the entire U.S. work
force, which has ranged between 4.9 and 6.7 percent during
this period (Office of the President of the United States,
1993).
Physical Sciences
6 T
\~
0.8 ~ :~~~~~
0.6
1 .
0.4
0.2
O - i i I ~I i
1 1 1
1973 1975 1977 1979 1981 1983 1985 1987 1989 1991
· Biomedical Sciences ~ Physical Sciences
FIGURE 3-10 Underemployment rates of biomedical and physi-
cal sciences Ph.D.s, 1973-1991. See Appendix Table F-9.
Postdoctorates
The number of new Ph.D.s with postdoctoral appoint-
ments can also reflect labor market conditions.8 One of the
functions of postdoctoral appointments, for example, has
been to provide interim positions for new researchers.
Given this as one of many functions of postdoctoral ap-
pointments, this number can be expected to rise when de-
mand is weak.
Figure 3-11 summarizes information on such appoint-
ments for new biomedical researchers (i.e., those who re-
ceived their degrees 4-5 years earlier) for the period 1973-
1991. The data show that the fraction of these researchers
who are postdoctorates rose dramatically in the 1970s. This
strong trend was followed by a smaller, more erratic pattern
inthel980s.
These trends, on inspection, do not support the notion of
a weakening demand in the biomedical fields in the 1980s.
The unusual pattern observed in the 1980s suggests that
other factors may have influenced the fraction of recent bio
20
_
Hi, 1 5
10
. /
1973 1975 1977 1979 1981 1983 1985 1987 1989 1991
FIGURE 3-11 Fraction of biomedical science Ph.D.s at career
age 4-5 on postdoctoral appointments, 1973-1991. See Appendix
Table F-10.
29
OCR for page 30
MEETING THE NATION' S NEEDS FOR BIOMEDICAL AND BEHAVIORAL SCIENTISTS
medical Ph.D.s holding postdoctoral appointments in that
period. For example, the observed trends may reflect varia-
tions in the availability of funding for postdoctorates, with
increasing support in the 1970s and increasing constraint in
the 1980s.
Postgraduafion Commitments
The postgraduation plans of new doctorates may also
reflect market conditions. In particular, the percentage of
new doctorates who indicate that they have definite com-
mitments at the time they are completing their requirements
for the degree can reflect the strength of demand. When
demand is weak this percentage will fall; when demand is
strong this percentage will rise.
Figure 3-12 summarizes these plans for the period 1975
to 1992. To provide a comparative base, similar informa-
tion is provided for degree recipients in the physical sci
"°T
1n~ ~
.
or
° 100
11
a,
-
x
90
-
85
·-
I _ ~ ~:~=1_ ~ ~
~-
-Biomedical Sciences = Physical Sciences |
FIGURE 3-12 Fraction of new biomedical and physical sciences
Ph.D.s with definite commitments, 1975-1992. See Appendix
Table F-1 1.
ences, which are thought to be suffering currently from
weak demand. The data show a notable declining trend in
this percentage for each of these fields beginning in 1989,
but the trend is more pronounced in the physical sciences.
Starting Salaries
"Starting salaries" are defined as the median salaries of
doctorates, age 30-34, who currently hold full-time employ-
ment positions (excluding postdoctoral positions). Infor-
mation regarding the starting salaries of biomedical doctor-
ates relative to comparable salaries for all science and
engineering doctorates is presented in Figure 3-13. Since
1983, salaries for these scientists in fields other than the
basic biomedical sciences taken as a whole have been grow-
ing relatively faster than salaries for basic biomedical sci
30
98T
96t
94
43 92
_,
c 90
88
86
mu
. ~
84
82 - 1 1
1973 1975 1977
\ ~
1 1 1 1 1 1 1
1979 1981 1983 1985 1987 1989 1991
FIGURE 3-13 Salaries of biomedical science Ph.D.s (age 30-34)
who currently hold full-time employment positions (excluding
postdoctoral positions) as a percentage of comparable salaries for
all scientists and engineers, 1973-1991. See Appendix Table
F-12.
enlists. This suggests that relative demand has been grow-
ing more slowly in the biomedical sciences than in other
fields of science or engineering combined.9
OUTLOOK FOR BASIC
BIOMEDICAL SCIENTISTS
The labor market for biomedical scientists defines one
dimension of need. Job openings are generated by deaths,
retirements, and other types of separation from the biomedi-
cal work force. In addition, job openings are also generated
by growth in employment demand. These job openings
may be filled by recruitment from many talent pools: new
doctorate recipients, experienced doctorates from other la-
bor markets or from outside We labor market (including
doctorates from abroad), nondoctorates, etc. In this con-
text, need can be defined as filling future job openings to
achieve a particular rate of employment growth or to
achieve some alternative goal. The target rate of growth or
the alternative goal is a policy decision usually made on
normative grounds.
Given this broad context, the committee examines future
employment conditions in an effort to estimate need (ap-
proximated by job openings) and our ability to meet this
need (measured by new Ph.D.'s entering the biomedical sci-
ences workforce). Because job openings can be filled by
recruitment from a variety of talent pools, the reader is cau-
tioned that the committee's indicator of our ability to meet
this need represents a lower-bound estimate of this ability.
Table 3-3 contains estimates of the future number of job
openings to be filled under alternative scenarios about em-
ployment growth. Three scenarios are examined: zero
growth, 3.6 percent per year (the 1981-1991 compound
growth rate for the biomedical science workforce), and 1.8
percent per year (one-half the 1981-1991 compound growth
rate). The method used to generate these estimates is a
OCR for page 31
BASIC BIOMEDICAL SCIENCES PERSONNEL
TABLE 3-3 Committee Estimates of the Average Annual Number of Job Openings
Needed to Sustain Vanous Growth Rates of the Biomedical Work Forces b
Zero Growth Rate
Scenario
Half the Average
Growth Rate ScenarioC
Average Growth Rate
Scenarios
Year Numbers Needed Numbers Needed Numbers Needed
1996-1997 1291 3358 5473
1998-1999 1714 3691 6031
2000-2001 1991 3915 6506
aBiomedical work force consists of those employed or on postdoctoral appointments in a biomedical field. Data
derived from the NRC Survey of Doctorate Recipients, a sample survey.
bBased on multistate life table methods. See Appendix G for methodology.
CHalf the average referred to in footnote d or 1.8 percent.
Refers to biomedical work force's average annual compound growth rate over the past decade or 3.6 percent
(4.25 percent, uncompounded).
variant of demographic cohort-survival models. It gener-
ates flows of workers into and out of this work force, and,
on the basis of these flows, it generates estimates of changes
in the size of this work force.10 There are, of course, many
ways to do multistate life table analysis. The data presented
below should be viewed as preliminary work by the com-
mittee, which will be explored further by the Panel on Esti-
mation Procedures in the coming months.
Estimates are developed for three time periods: 1996-
1997, 1998-1999, and 2000-2001. The estimates are very
sensitive to Me growth rate assumption, varying from 1,291-
1,991 in the zero growth scenario to 5,473-6,506 in the 3.6
percent per year growth scenario. The range is substan-
tially narrower for a given growth rate scenario. The mod
est increases observed over time for a given rate of growth
partially reflect the widely expected increases in deaths and
retirements in the late l990s. Except for the zero growth
scenario, they also reflect the growth of the biomedical sci-
ence work force.1l
For comparison purposes, Table 3-4 shows the number
of new biomedical Ph.D.s entering the biomedical work
force through 1990, estimated from We longitudinal SDR.12
These numbers represent a substantial fraction of the de-
gree production that occurred in these fields, although it
does not reflect the employment outcomes of new graduates
who may have found employment in other fields or delayed
entry into the work force. An estimated 82 percent of the
biomedical Ph.D.s entered the biomedical work force
TABLE 3-4 Estimated Number of New Biomedical Science
Ph.D.s Entering the Biomedical Science Work Force in Selected
Years.
Year
Numbers
1985-1986
1987-1988
1989-1990
2985
3178
3353
aAnnual averages.
NOTE: Biomedical science work force" consists of those employed or on postdoctoral appointments in a
biomedical field. The Survey of Doctorate Recipients is a sample survey and subject to sampling error.
SOURCE: NRC, Survey of Doctorate Recipients. (Biennial)
31
OCR for page 32
MEETING THE NATION, S NEEDS FOR BIOMEDICAL AND BEHAVIORAL SCIENTISTS
(which includes postdoctorates) during the period 1985-
1990.
This level of work force entry, if maintained, could more
than meet the need for zero growth, but it will fall consider-
ably short of the number needed to maintain the annual
1981-1991 growth rate. As noted earlier, however, mainte-
nance of this growth rate may be an unrealistic objective.
Universities are unlikely to increase faculty size dramati-
cally in the near future, federal spending on biomedical re-
search is not likely to increase in real terms in the near
future, and private sector demand (viz., industry) is not
likely to increase rapidly in the near future.
The best predictions for economic activity and R&D
funding in the near future suggest that demand for basic
biomedical scientists will grow slowly at best. Under these
circumstances, maintenance of the current rate of entry of
Ph.D.s in the biomedical sciences should provide an ad-
equate supply for the years 1996-2001. ~3 (See Table 3-3~.
The NRSA program supports approximately 5,100
predoctoral students each year in the basic biomedical sci-
ences, although only a fraction complete doctoral degrees
in the same year as receiving NRSA support. The number
of basic biomedical degree recipients in any year having
had NRSA support is unknown but presumed to be small.~4
If current levels of predoctoral NRSA support are main-
tained and projected, demand for new Ph.D.s is estimated to
be 3,400-3,900 per year ("half the average" growth rate sce-
nario, see Table 3-4), then the NRSA program in the basic
biomedical sciences will contribute to the preparation of
doctoral scientists at a rate which future markets will likely
absorb.
Priority Fields
. .
Although market conditions suggest that the demand for
basic biomedical scientists may grow more slowly than in
the past, we believe that advances in research and continu-
ing requirements to address pressing public health concerns
will result in the demand for basic biomedical scientists
with quite specific research skills. This does not imply that
we need to step up production in all areas; rather, the NRSA
mechanism provides an opportunity to increase supply in
some areas through relatively small increases in the number
of awarders.
There is a continuing need to train young scientists who
will have skill and expertise in the well-recognized core
biomedical disciplines (e.g., biochemistry, microbiology,
and pharmacology) as well as broadly based individuals ca-
pable of effective interdisciplinary research. Scientists
trained in physical and mathematical sciences and engineer-
ing and able to apply knowledge in chemistry, physics, ma-
terials science, computational mathematics, and computer
science to problems of significance in basic and clinical
32
biomedical sciences will also be required. Much of the
current excitement and rapid progress in biomedical science
lies at the interfaces between genetics, molecular biology,
cellular biology, and developmental biology. The NRSA
programs in the basic biomedical sciences appropriately
emphasize the kinds of interdisciplinary training required to
carry out effective research at these interfaces and to apply
new findings to problems in human biology. However, it is
also necessary to ensure that there is a cadre of scientists
who are knowledgeable in fundamental areas of biomedical
science that, for whatever reason, are not at the cutting edge
of research at the time. This need, which is perhaps less
immediately obvious, is well illustrated by the periodic
emergence of new infectious diseases (e.g., Legionnaires
disease and cryptosporidiosis) and the recrudescence of dis-
eases, such as tuberculosis, caused by antibiotic- and drug-
resistant strains.
ENSURING THE DIVERSITY OF
HUMAN RESOURCES
Careers in biomedical research remain attractive to
women. At present, between 35 and 45 percent of Ph.D.s
awarded in the biomedical sciences have been awarded to
women. However, the fraction of women in full-time, inde-
pendent research positions is still disproportionately low.
Moreover, evidence suggests that women rise to the top
ranks in academia and industry in fewer numbers than men
(NRC, 1991 and 1994~. Part of the training process should
include explicit mentoring to help women achieve their full
career potential.
We need to do much more to increase the number of
black and Hispanic students entering research Gaining in
the basic biomedical sciences. Continual efforts to attract
these students into the NRSA program must be made. Spe-
cial programs to ensure progress through pre- and
postdoctoral Gaining should be encouraged. The Minority
Access to Research Careers (MARC) program shows prom-
ise as a reliable source of NRSA trainees.
THE NRSA PROGRAM IN THE BASIC
BIOMEDICAL SCIENCES
Earlier committees' assessment of the need for basic bio-
medical scientists and the level of Raining Hat should be
provided by the federal government under the NRSA pro-
grams depended heavily on its analysis of the academic la-
bor market, because that was the dominant sector both in
terms of the number of bioscientists employed and the
amount of federally sponsored research performed. The
number of individuals receiving Ph.D. degrees in the bio-
medical sciences and the number holding postdoctoral ap-
pointments were taken as indicators of supply. Demand
OCR for page 33
BASIC BIOMEDICAL SCIENCES PERSONNEL
indicators were undergraduate and graduate enrollment and
the availability of funds for R&D, both of which were per-
ceived to drive the demand for faculty in these fields. Those
data, combined win conservative estimates of future Rends,
were used to make recommendations about Me number of
trainees needed.
When the committee first convened in 1975, it quickly
discerned an oversupply of researchers in Me biomedical
sciences based on trends in academic employment. Al-
though demand for Ph.D. faculty had experienced rapid
grown during the 1960s, by the 1970s the number of stu-
dents was leveling off, federal funding increases were mod-
erating, and Me relatively young faculty members hired dur-
ing the period of peak expansion were suspected to undergo
very little attrition in Me near future. In its second report,
the committee recommended cutbacks amounting to 30 per-
cent in the number of predoctoral fellows supported annu-
ally between fiscal year 1975 and fiscal year 1979, from
6,000 to 4,250, and a level of support of 3,200 postdoctoral
fellows. These recommendations were based on evidence
of reduced growth in the overall demand for biomedical
scientists and continue to affirm the vital role played by the
training grant and fellowship programs in training high-
quality researchers.
Subsequent reports in 1978, 1979, and 1981 reiterated
these recommended levels and suggested that time was
needed to evaluate Me effects of these cutbacks and further
developments in Me labor market before new recommenda-
tions could be made. By 1981, the committee discerned
signs of improvement in the overall job market for biomedi-
cal researchers. Academic employment was expanding
slightly, largely because of rising enrollments in Me bio-
medical sciences. R&D funding was also beginning to rise,
and most promising was the rapid increase of employment
in the new biotechnology industry. The committee foresaw
continued, strong demand from both small start-up compa-
nies and large established corporations that were entering
this business. It also saw good employment prospects in the
high-priority fields of biostatistics, toxicology, and epide-
miology. Nonetheless, it expressed concern about Me con-
tinued grown of the postdoctoral pool and recommended
that the numbers of pre- and postdoctoral awards remain
steady at 4,250 and 3,200, respectively.
By 1985 Me job market showed clear improvement. For
the first time since Me reports began, the committee noted a
slowing of postdoctoral buildup. The number of Ph.D.s
awarded each year also slowed, and although university fac-
ulty employment still remained stable, demand in the bio-
technology and genetic engineering industries was growing
sharply, at more Man 9 percent a year. Although Me com-
mittee did not expect substantial increases in the number of
academic positions in the foreseeable future, it did expect
retirements to increase markedly by the mid to late 1990s.
33
In 1989 the study committee projected that an increasing
demand for biomedical scientists would exceed Me supply
through the year 2000. The committee recommended Mat
the level of NRSA predoctoral support be increased to
5,200. The committee also recommended that postdoctoral
support be increased gradually as degree production in-
creased.
RECOMMENDATIONS
Total support for the training of basic biomedical scien-
tists increased from just over 9,000 awards in fiscal 1991 to
an estimated 9,630 in fiscal 1993 (Table 3-5~. This in-
cludes about 630 awards for the undergraduate preparation
of minority scholars. Predoctoral support is offered prima-
rily through institutional training grants (traineeships), al-
though a limited number of individual fellowships are avail-
able. Postdoctoral support in the form of portable
fellowships allows eligible applicants to seek advanced
preparation in a wide variety of areas. Postdoctoral training
grants have emphasized preparation in such areas as genet-
ics, clinical pharmacology, trauma and burn research, and
anesthesiology. In making its recommendations in this area,
Me committee has assumed that Me current ratio of
predoctoral and postdoctoral support would remain essen-
tially constant, with Me majority of support available at the
predoctoral level (primarily in the form of aaineeships).
Predoctoral Training
On the basis of its review of available information de-
scribing current and anticipated market conditions and in
consideration of pressing national research needs, the com-
mittee strongly endorses the continuation of predoctoral
NRSA training programs in the basic biomedical sciences.
Although evaluative data remain incomplete, Me evidence
indicates Mat these predoctoral training programs remain
highly effective in fostering the development and sustaining
the health of interdisciplinary graduate programs of bench-
mark quality, and in catalyzing the entry of highly qualified
students into graduate Paining.
However, Me committee is concerned that the current
low level of stipend support, $8,800 per year, will erode the
impact of these programs and their ability to attract the most
talented students. We recommend Mat stipends be increased
incrementally over a 2-3 year period to $12,000 for all
predoctoral awarders. We consider the recommended in-
crease in stipend to be of higher priority than any possible
increases in number of trainee slots, and therefore recom-
mend that He number of predoctoral awards remain at FY
1993 levels during this period. The committee recognizes
that these recommendations are being made in an era of
OCR for page 34
MEETING THE NATION'S NEEDS FOR BIOMEDICAL AND BEHAVIORAL SCIENTISTS
TABLE 3-5 Aggregated Numbers of NRSA Supported Trainees and Fellows in Basic
Biomedical Sciences for FY 1991 through FY 1993
Fiscal
Year Level of Training
Type of Support
TOTAL Traineeship Fellowship
1991 Number of awards 9,021 7,199 1,822
Predoctoral 4,593 4,313 280
Postdoctoral 3,861 2,319 1,542
MARC Undergraduate 567 567
1992 Number of awards 9,317 7,477 1,840
Predoctoral 4,777 4,487 290
Postdoctoral 3,910 2,360 1,550
MARC Undergraduate 630 630
1993 Number of awards 9,633 7,740 1,~893
Predoctoral 5,171 4,811 360a
Postdoctoral 3,836 2,303 1,533
MARC Undergraduate 626 626
a Includes minority scholars supported through the National Minority Fellowship Program. See
Appendix E.
NOTE: Based on estimates provided by the National Institutes of Health. See Summary Table 1.
fiscal constraint. Should additional funds become available
for research training in the basic biomedical sciences, the
NIH might wish to consider expanding NRSA support in
this area.
RECOMMENDATION: The committee recommends
that the number of Predoctoral trainees and fellows sup-
ported annually in the basic biomedical sciences be main-
tained at 1993 levels or approximately 5,175 each year
(Table 3-6~.
Postdoctoral Training
Postdoctoral research training sharpens the technical and
intellectual skills of the doctoral-level scientist and pro-
vides important (and frequently used) opportunities for
cross-disciplinary training as preparation for undertaking a
career as an independent investigator.
The committee is concemed, however, that persistent
low-level stipends may discourage qualified applicants from
seeking postdoctoral training through NRSA support. Thus,
to permit NIH to introduce further and more realistic
changes in stipend levels at the postdoctoral level, the com-
mittee recommends that the number of postdoctoral awards
be maintained at fiscal 1993 levels (Table 3-6~. Should,
however, additional program funds become available for
postdoctoral training in the basic biomedical sciences, the NIH
may also wish to expand support for postdoctoral training.
34
RECOMMENDATION: The committee recommends
that the number of postdoctoral trainees and fellows sup-
ported annually in the basic biomedical sciences be main-
tained at 1993 levels or 3,835 each year.
Minority Access to Research Careers
Current federal efforts to attract minority group mem-
bers to careers in the basic biomedical sciences include un-
dergraduate support through He MARC program. The core
of this program is the Honors Undergraduate Program
launched in fiscal 1977 to support college juniors and se-
niors. In fiscal 1993 approximately 630 individuals received
undergraduate support (see Table 3-54.
As noted in Chapter 9 of this report, NIH recently initi-
ated an 18-month study of the career outcomes of the
MARC program. The committee endorses continuation of
funding for this program at current levels to support the
training of individuals in the basic biomedical sciences until
the NIH assessment is complete and information is made
available to subsequent NRC study committees.
RECOMMENDATION: The committee recommends
that He number of NRSA awards made available through
the MARC undergraduate program for research training
in He basic biomedical sciences be maintained at about
630 awards each year.
OCR for page 35
BASIC BIOMEDICAL SCIENCES PERSONNEL
TABLE 3-6 Committee Recommendations for Relative Distribution of Predoctoral
and Postdoctoral Tra~neeship and Fellowship Awards for Basic Biomedical Sciences
for FY 1994 through FY 1999
Fiscal
1994 Recommended number of awards
P red oc to rat
Postdoctoral
MARC Undergraduate
Level of Training
TOTAL
9,640
5,175
3,835
630
Type of Support
Traineeship Fellowship
7,745
4,815
2,300
630
1995 Recommended number of awards
Predoctoral
Postdoctoral
MARC Undergraduate
1996 Recommended number of awards
Predoctoral
Postdoctoral
MARC Undergraduate
1997 Recomrr ended number of awards
Predoctoral
Postdoctoral
MARC Undergraduate
1998 Recommended number of awards
Predoctoral
Postdoctoral
MARC Undergraduate
1999 Recommended number of awards
Predoctoral
Postdoctoral
MARC Undergraduate
9,640
5,175
3,835
630
9,640
5,175
3,835
630
9,640
5,175
3,835
630
9,640
5,175
3,835
630
9,640
5,175
3,835
630
1,895
360
1,535
7,745
4,815
2,300
630
7,745
4,815
2,300
630
7,745
4,815
2,300
630
7,745
4,815
2,300
630
7,745
4,815
2,300
630
1,895
360
1,535
1,895
360
1,535
-
1,895
360
1,535
-
1,895
360
1,535
1,895
360
1,535
NOTES
1. The slowdown in the rate of growth between 1989 and 1991 was
accompanied by an absolute decline in academia. In part, this may reflect
methodological changes that occurred in the Survey at that time. Howev-
er, it may also reflect a weakening in demand, particularly in the academic
sector.
2. Includes both citizens and noncitizens, where citizens includes both
native-born and naturalized citizens.
3. Special run, Survey of Doctoral Recipients (SDR). SDR is a bien-
nial survey of a sample of scientists and engineers conducted by the NRC
behalf of the federal government.
4. Special run, SDR. A number of authors have also observed that
women are generally underrepresented in tenure-track faculty positions
relative to the numbers among doctoral recipients (NRC, 1981; Zuckerman
et al., 1992).
5. The Doctorate Records File is a compilation of responses to the
Survey of Earned Doctorates, which has been conducted each year since
1958 by the NRC's Office of Scientific and Engineering Personnel and its
predecessor organizations. Questionnaires, distributed with the coopera-
tion of the graduate deans of U.S. universities, are filled in by graduates as
they complete requirements for their doctoral degrees. The doctorates are
reported by academic year and include research and applied-research doc-
torates in all fields. See Ries and Thurgood, 1993.
6. Much more work is needed to document fully the impact of foreign
participation on the U.S. science and technology work force. Owing to the
nature of many of our data sources, we are unable to determine at this time
how many non-U.S. citizens who earn doctoral degrees in this country
remain in the U.S. and we know very little about the careers and research
contributions of non-U.S. citizens to the U.S. research effort regardless of
the origin of their doctoral degrees.
35
OCR for page 36
MEETING THE NATION' S NEEDS FOR BIOMEDICAL AND BEHAVIORAL SCIENTISTS
7. The Panel on Estimation Procedures considered numerous "short
run" indicators such as wage adjustments for young workers relative to
older workers, relative tenure-earning ratios, and job openings. However,
owing to limitations of time and resources, the Panel and Committee re-
stricted these analyses to more readily available information.
8. The postdoctoral appointment has become an essential component
of advanced training in most subfields of the basic biomedical sciences.
Past studies by the National Research Council (Garrison and Brown, 1984,
for example) have suggested that individuals with postdoctoral training
enter more productive research careers than those individuals without post-
doctoral training. See Appendix A of this report for a brief summary of
the career outcomes studies of NRSA postdoctoral appointees. Nonethe-
less, the expansion of postdoctoral appointments in the basic biomedical
sciences has been identified by some researchers as an indicator of job
shortages in some component fields (Coggeshall, et al., 1978; NRC, 1981).
9. An alternative interpretation of this finding is that the relative sup-
ply of biomedical scientists increased faster than that of other scientists
and engineers. The decline in starting wages would thus result from an
increase in relative demand.
10. The flows are generated from multistate life tables. These tables
are based on matrices of age-specific transition rates estimated from the
Survey of Doctoral Recipients historical data. These rates are assumed to
remain constant over time. For a more detailed description of the method-
ology, see Appendix G. This analysis will be reviewed closely by the
Panel on Estimation Procedures along with other approaches to the estima-
tion of national needs relative to human resource training and policies.
11. Recall that in developing these estimates, it is assumed that age-
specific separation rates remain stable. There is, however, evidence of a
strong positive relationship between these rates and age ARC, 1989). Giv-
en this relationship, the upward trend in the numbers may also be reflect-
ing the expected aging of this population.
12. SDR. See note 3.
13. The estimate is presented as a minimum value because these job
openings could also be filled by recruiting workers with degrees and train-
ing in closely related fields or workers from abroad.
14. On the basis of information gathered from the National Science
Foundation the committee estimates that less than 15 percent of graduate
students in the life sciences received NRSA support in FY 1990.
36
REFERENCES
Coggeshall, P., J. C. Norvell, L. Bogorad, and R. M. Bock
1978 Changing postdoctoral career patterns for biomedical scien-
tists. Science 202:487-493.
Matyas, M. and L. S. Dix (eds)
1992 Science and Engineering Programs: On Target for Women?
Washington, D.C.: National Academy Press.
National Science Foundation (NSF)
1990 Immigration of Scientists and Engineers to the United States:
A Literature Review. Science Resources Studies Division.
Mimeographed. March. Washington, D.C.
National Research Council (NRC)
1994 The Funding of Young Investigators in the Biological and
Biomedical Sciences. Washington, D.C.: National Academy
Press.
1981 Postdoctoral Appointments and Disappointments. Washing
ton, D.C.: National Academy Press.
1989 Biomedical and Behavioral Research Scientists: Their Train
ing and Supply, Volume I: Findings. Washington, D.C.: Na
tional Academy Press.
1991 Women in Science and Engineering. Increasing Their Num
bers in the 1990s. Washington, D.C.: National Academy
Press.
1994 Women Scientists and Engineers Employed in Industry: Why
So Few? Washington, D.C.: National Academy Press.
Office of the President of the United States
1993 Economic Report of the President. Washington, D.C.: U.S.
Government Printing Office.
Ries, P. and D. H. Thurgood
1993 Summary Report 1992: Doctorate Recipients from United
States Universities. Washington, D.C.: National Academy
Press.
Zuckerman, H., J.R. Cole, and J.T. Bruer
1991 The Outer Circle: Women in the Scientific Community. New
York: W. W. Norton and Company.
Representative terms from entire chapter:
biomedical sciences
