S P E C I A L F E A T U R E

C o m m e n t a r y

Dissecting the Workforce and Workplace for Clinical

Endocrinology, and the Work of Endocrinologists
Early in Their Careers

Claude Desjardins, Mark A. Bach, Anne R. Cappola, Ellen W. Seely,

and Ronald G. Ehrenberg

Downloaded from https://academic.oup.com/jcem/article/96/4/923/2720836 by Anelis user on 02 November 2023

Clinical Scholars Project (C.D.), Department of Physiology and Biophysics, University of Illinois College of
Medicine, Chicago, Illinois 60612; Division of Research and Development (M.A.B.), Janssen
Pharmaceutical K.K., Chiyoda-ku, Tokyo 101-0065, Japan; Division of Endocrinology, Diabetes, and
Metabolism (A.R.C.), University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104;
Division of Endocrinology, Diabetes, and Hypertension (E.W.S.), Brigham and Women’s Hospital, Harvard
Medical School, Boston, Massachusetts 01215; and Cornell Higher Education Research Institute and the
Department of Labor Economics (R.G.E.), School of Industrial and Labor Relations, Cornell University,
Ithaca, New York 14853

Context: The United States lacks timely reliable mechanisms for assessing the professional work of
subspecialty physicians.

Objective: The aim was to use early-career members of The Endocrine Society as a model to estimate
subspecialty physician involvement in patient care, teaching, research, and administration among
clinical, academic, federal, and pharmaceutical/biotech workplaces and to assess the workforce for
research within individual workplaces.

Methods: Physicians joining The Endocrine Society from 1991–2005 and residing in North America
were invited to complete a Web-based survey. This report relies on 817 early-career endocrinol-
ogists or 29.6% of eligible respondents.

Results: Respondents from all types of workplaces engaged in patient care, teaching, research, and
administration. The time committed to the four tasks, however, differed significantly among
workplaces. Research (basic, translational, disease, patient, population, and prevention) was ac-
complished within all workplaces, but the scope and scale of investigative work was employer
dependent. Recipients of National Institutes of Health K08/23 awards succeeded in receiving fed-
eral research project grants (P ⬍ 0.001). Respondents associated research with lowered incomes,
a perception validated by an estimated drop in annual earnings of 2.8% per half-day spent on
research (P ⬍ 0.001). Women in academic settings earned less than men (P ⬍ 0.01) and were less
likely to occupy tenure-eligible positions (P ⬍ 0.01).

Conclusions: Web-based surveys offer a simple tool for estimating the work of subspecialty phy-
sicians and provide a framework for improving biomedical investigation. Several interventions
should be considered for endocrinology: recruit physicians from underrepresented demographic
groups, increase K08/23 awards, incentivize investigative careers, and improve the national infra-
structure for biomedical research. (J Clin Endocrinol Metab 96: 923–933, 2011)

ISSN Print 0021-972X ISSN Online 1945-7197

Printed in U.S.A.
Copyright © 2011 by The Endocrine Society
doi: 10.1210/jc.2010-1568 Received July 8, 2010. Accepted January 7, 2011.
First Published Online February 2, 2011
Abbreviations: AMC, Academic medical center; co-PI, coprincipal investigator; PI, principal
For editorial see page 942 investigator.

J Clin Endocrinol Metab, April 2011, 96(4):923–933 jcem.endojournals.org 923

924 Desjardins et al. Professional Activities of Endocrinologists J Clin Endocrinol Metab, April 2011, 96(4):923–933

o national mechanism is in place for an informed, United States between 1991 and 2005 (9). A total of 641 indi-
N penetrating, and systematic assessment of the phy-
sician workforce such as that achieved by the National
viduals are excluded from this survey for reasons that follow:
faulty e-mail address (n ⫽ 437), declined to participate (n ⫽ 89),
started but failed to return a questionnaire or completed less than
Science Foundation (NSF) for the periodic evaluation of
75% of survey questions (n ⫽ 43), or lacked a license to practice
the nation’s scientists and engineers (1, 2). Likewise, medicine (n ⫽ 72). A total of 817 of the 2757 prospective re-
knowledge of the workforce for clinical research is enig- spondents are included in this analysis, yielding a response of
matic and fragmentary despite the serial recommenda- 29.6%.
tions of “blue-ribbon” panels to establish a protocol for The Survey Research Institute at Cornell University distrib-
the recurrent assessment of clinical investigators early in uted materials to respondent e-mail addresses between February
their careers (3–5). Failure to adopt a national system for and April 2008. Participants submitted questionnaires anony-
mously over the Internet via a secure server. Participant instruc-

Downloaded from https://academic.oup.com/jcem/article/96/4/923/2720836 by Anelis user on 02 November 2023

producing timely, high-quality data on the professional
tions noted that the Council of TES approved the survey and that
activities of physicians limits the application of improve- the Office for the Protection of Research Subjects, University of
ment tools for advancing clinical investigation and ulti- Illinois at Chicago, granted Institutional Review Board (IRB)
mately improving clinical practice (6). approval for an exempt protocol. Respondent instructions ex-
The present study was designed as a pilot project to test plained that participation was voluntary, that respondent con-
the feasibility of using Web-based surveys to estimate the fidentiality would be maintained, and that none of the research
administrative, clinical, didactic, and research work of conducted or published would divulge the responses of individ-
ual physicians.
subspecialty physicians employed in academic, clinical,
federal, and pharmaceutical workplaces. Physician mem-
bers of The Endocrine Society (TES) were used as surro- Design and content of survey instrument
gate prototypes of a subspecialty workforce because of We tested a preliminary version of the questionnaire by in-
their manageable number and investigative tradition. The viting a random sample of 40 physician members to respond to
results establish that Web-based surveys provide a tool to a prototype of the questionnaire. We asked pilot survey partic-
ipants to furnish written comments on any question they per-
assess the activities of a decentralized workforce employed
ceived to be ambiguous, awkward, or impertinent. Twenty-eight
in disparate workplaces and underscore the value of fo- participants (70% response rate) provided remarks to reformat
cusing on physician work within the context of particular the questionnaire. Questionnaires returned by pilot volunteers
workplaces within a subspecialty. Our report also pro- are excluded from this report.
vides a new and timely snapshot of the amount and types We asked participants to identify their employment sector:
of research performed by clinically trained endocrinolo- private practice (identified throughout as clinical care providers
gists and offers an evidenced-based framework for im- who own all or part of a solo/group practice or are employed by
a group practice, health system, or hospital), academic (em-
proving the investigative workforce in this medical
ployed by a medical school or teaching hospital), federal gov-
subspecialty.
ernment, or pharmaceutical and/or biotech industry. We re-
quested that each respondent specify the number of half-days per
week devoted to clinical service, teaching, research, and admin-
istrative work based on their supervisor’s expectations, or them-
Subjects and Methods selves in the case of solo practitioners; designate the source(s) of
salary support for time spent on research; and indicate their
Survey participants annual pretax compensation within ordered ranges. Survey in-
TES is the primary professional organization dedicated to structions advised respondents to identify one or more research
advancing the clinical practice of endocrinology in North Amer-
activities that best defined their investigative work over the past
ica. The Society agreed to collaborate on an assessment of phy-
12 months. The questionnaire allowed respondents to choose
sician members early in their careers and to provide their e-mail
from an inclusive list of research activities (see Table 3) using
addresses. Early-career members are defined as follows: they
definitions developed at a consensus conference on clinical re-
joined the TES between January 1, 1991, and December 31,
search (10) and later improved by others (11, 12).
2005; they were 49 yr of age or younger on joining TES; they
earned an M.D. or equivalent degree; they reside in Canada or We asked a series of questions to examine the indifference to
the United States; and they hold an active license to practice investigative careers. The questions relied on a five-point Likert
medicine. We adopted this profile because internists/pediatri- scale (13), collapsed for analysis to three outcomes: agree, dis-
cians in the United States complete fellowship training at about agree, neither agree/disagree. A second set of questions consid-
34 yr of age (7), obtain National Institutes of Health (NIH) ered proposals and awards for postfellowship research training
support for their first research grant at an average age of 44 yr (8), and research project grants/contracts from PIs, co-PIs, or both.
and function as principal investigators (PIs) or coprincipal in- We invited respondents to designate their age, citizenship (Can-
vestigators (co-PIs) for 5 yr or more. ada or U.S.), gender, and ethnic origin. We determined the ethnic
We sent invitations to 3398 early-career physician members backgrounds of physicians graduating from allopathic schools of
of TES, a cohort representing about 89% of internists/pediatri- medicine in the United States by estimating the mean fraction of
cians trained in endocrinology, diabetes, and metabolism in the each ethnic group earning an M.D. degree in 1991 to 2005 (7).
J Clin Endocrinol Metab, April 2011, 96(4):923–933 jcem.endojournals.org 925

Statistical methods North America. In contrast, a respective 96.6 and 99.3%

Each questionnaire received a unique computer-generated of respondents completed residency and fellowship train-
case number to allow respondent identity to remain anonymous ing in North America. Respondents earned baccalaureate
throughout the analysis of all data. Prospective estimates of sam-
degrees at 21.6 ⫾ 0.1 (mean ⫾ SEM) yr of age, graduated
ple size are based on those of Lansing and Morgan (14), and
statistical analyses used software (version 6.0) distributed by from medical school at 26.7 ⫾ 0.1 yr of age, and finished
Stata Corp. (College Station, TX). fellowship training at 33.1 ⫾ 0.1 yr of age. Over 97% of
Categorical variables are compared via Pearson’s ␹2 good- respondents are board certified or board eligible in either
ness-of-fit-test. Statistical assessments of continuous variables adult or pediatric endocrinology, an indication that the
are based on the ANOVA or a simple paired t test for between- clinical training of the respondent sample is uniform. The
group comparisons. We used a simple multiple regression model
to test the association between the receipt of grants by PIs and
ratio of internists to pediatricians in the respondent sam-

Downloaded from https://academic.oup.com/jcem/article/96/4/923/2720836 by Anelis user on 02 November 2023

co-PIs and other explanatory variables of research performance, ple was 3:1, a proportion consistent with the number of
and a Tobit regression model to determine the association be- physicians trained in endocrinology, diabetes, and metab-
tween pretax earnings and various outcome measures to accom- olism in 1991–2005 (9).
modate the censored earnings reported by respondents within We compared the self-identified ethnic backgrounds of
specified ranges. We tested dependent variables with binary re-
U.S. respondents with those of physicians graduating from
sponses using a logit regression model. Some respondents failed
to specify the number of half-days spent on research or provide allopathic schools of medicine in the United States (Table
information related to the submission of proposals for research 1). The self-identified fraction of Native American, Asian,
project grants. The absence of a response, in 5.2% of cases, is Caucasian, and Hispanic/Latino respondents approxi-
assumed to be zero or identical to the “no” responses entered by mate those of graduates from U.S. schools of medicine.
most respondents. We justified this transformation on the basis The fraction of Black or African-American endocrinolo-
that respondents who are uninvolved with research would over-
look the need to verify a zero response.
gists, in contrast, is underrepresented by an absolute dif-
ference of 3.7% when compared with the mean of Black
or African-American physicians graduating in 1991 to
Results 2005.

Respondent sample
Workplace assessment
Our results are based on a cohort of 817 physician
The results indicate that clinical care providers, for in-
members of TES from a sample of 2757 prospective re-
stance, devote 7.8 half-days/wk to clinical service. The
spondents, yielding a response rate of 29.6%. Statistical
clinical service obligations of academic, federal, and phar-
estimates of sample size indicate that a response rate of
maceutical respondents, by comparison, are less (P ⬍
21.7% is consistent with achieving a confidence level of
0.01), averaging 3.6, 2.8, and 1.0 half-days/wk, respec-
95% for a survey with 2757 prospective participants. We
tively (Table 2).
tested the cohort of actual vs. prospective respondents to
Endocrinologists from all workplaces participate in
determine whether the year in which respondents joined
teaching medical students, residents, or fellows. The di-
TES and the fractions of respondents residing in Canada
dactic commitments of academic and federal respondents,
and the United States differed in the two populations. Test
however, are greater (P ⬍ 0.05) than those from other
results established that the distributions of actual and pro-
employment sectors (Table 2).
spective respondents are similar (P ⬎ 0.25), an indication
Turning to investigative work, clinical care providers
that the respondent cohort is representative of early-career
spend 1.9 half-days/wk on research. Respondents from
endocrinologists identified for this study.
academic, federal, and pharmaceutical/biotech work-
Respondent profile places, by comparison, spend 5.1, 5.7, and 7.5 half-days/
The fractions of Canadian and U.S. respondents are 5 wk, respectively, on research (Table 2). Administrative
and 95%, respectively. The distribution of citizens, per- responsibilities including committee assignments require
manent residents, and noncitizens from the United States 1.0 half-day/wk among clinical care providers; 1.4 and 1.2
is 86, 8, and 6%, respectively; and the ratio of women to half-days/wk in the academic and federal workplaces, re-
men is 48:52. We merged the responses of endocrinolo- spectively; and 2.8 half-days/wk in the pharmaceutical/
gists from the United States and Canada, except when biotech industry (Table 2).
specified, because every measured aspect (age, gender, ed-
ucation) of their profiles is indistinguishable (P ⬎ 0.50) Workforce analysis
based on multiple statistical estimates. We examined the amount and types of investigative
A respective 11.4 and 21.9% of respondents earned work pursued within all employment sectors: clinical ser-
baccalaureate and M.D. or equivalent degrees outside of vice, academic, federal, and pharmaceutical (Table 3). An
926 Desjardins et al. Professional Activities of Endocrinologists J Clin Endocrinol Metab, April 2011, 96(4):923–933

TABLE 1. Self-identified ethnic backgrounds of clinically trained members of TES, early in their careers, and
graduates of allopathic schools of medicine in the United States
Self-identified backgrounds
(mean %)a Graduates of U.S.
Ethnic background Women Men Both genders medical schools (%)b
American Indian or Alaskan Native 0.3 0 0.1 0.7
Asian-American 21.8 16.2 18.9 16.9
Black or African-American 3.2 2.5 2.8 6.5
Caucasian 65.5 72.8 69.3 68.7
Hispanic or Latino-American 6.2 5.0 5.6 6.4
Multiethnic 0.8 0.7 0.8 Unreported

Downloaded from https://academic.oup.com/jcem/article/96/4/923/2720836 by Anelis user on 02 November 2023

Other: unknown or unreportedc 2.2 2.7 2.5 0.8
a
Each value is expressed as the mean percentage of responses reported by 374 women (48.1%) and 403 men (51.9%) out of a total 777
graduates of allopathic schools of medicine in the United States.
b
The ethnic backgrounds of medical school graduates were determined by estimating the mean number of individuals, within specified ethnic
groups, graduating from allopathic schools of medicine from 1991 to 2005 (7). An average of 15,713 students graduated per year between 1991
and 2005, the same 15-yr sample window adopted for the present survey of clinically trained endocrinologists early in their careers.
c
The fraction of individuals that self-identified as ⬙other⬙ may include respondents from ethnicities that were unlisted (Hawaiian Native, Pacific
Islander) in the survey instrument or individuals whose ethnicity is unknown or undisclosed.

estimated 35.9% of respondents from the clinical care sic-, translational-, and prevention-oriented research is
workplace reported research efforts, albeit at efforts that pursued by clinical care providers, but on the order of 3.2
are more modest than workplaces offering protected re- to 0.6% (Table 3).
search time (Table 3). Patient-oriented studies are pursued Respondents affiliated with the federal workforce re-
by about 21.2% of clinical care providers, an effort 4-fold ported a research commitment of 87.5%. Disease-oriented
greater than other research activities (P ⬍ 0.01). A respec- research is a priority among federal respondents (P ⬍ 0.05),
tive 5.1 and 4.5% of the clinical care providers are con- involving 28.1% of this workforce. Federal respondents
cerned with population- and disease-oriented work. Ba- report similar commitments (P ⬎ 0.15) to translation-,

TABLE 2. Assessment of the number of half-days per week devoted to clinical service, teaching, research, and
administration among all types of workplaces employing clinically trained members of TES
Time commitment in half-days/wka
Clinical care AMC or teaching Federal government Pharmaceutical
Activity providersb hospitalc (HHS/DOD/VA)d biotech/industry
Clinical service 7.8 ⫾ 0.2 (69.1) 3.6 ⫾ 0.1 (31.4) 2.8 ⫾ 0.4 (26.6) 1.0 ⫾ 0.4 (8.8)
Teachinge 0.8 ⫾ 0.1 (6.6) 1.4 ⫾ 0.1 (11.6) 1.5 ⫾ 0.1 (13.5) 0.3 ⫾ 0.1 (2.4)
Research 1.9 ⫾ 0.1 (16.9) 5.1 ⫾ 0.2 (45.2) 5.7 ⫾ 0.6 (49.6) 7.5 ⫾ 0.5 (67.2)
Administrationf 1.0 ⫾ 0.1 (7.4) 1.4 ⫾ 0.1 (11.8) 1.2 ⫾ 0.2 (10.3) 2.8 ⫾ 0.6 (21.6)
Total 11.3 ⫾ 0.2 11.4 ⫾ 0.1 11.3 ⫾ 0.5 11.7 ⫾ 0.7
Data are expressed as mean ⫾ SEM (percentage effort).
a
Each value is expressed as the mean ⫾ SEM of half-days per week reported by 760 respondents from a total of 817 survey participants. The
number of respondents within each employment sector is as follows: 288 clinical care providers, 411 AMCs, 31 federal, and 30 pharmaceutical/
biotech. Note that the ⬙total⬙ half-days per week deviate from the expected value of 10; values were not normalized to 40 h/wk to allow the data
to reflect the effort reported for each workplace. The percentage effort (shown in parentheses) allows comparisons within and among workplaces.
Respondents were asked to specify the half-days per week devoted to each of the indicated activities during the 12 months preceding the survey
based on the time they negotiated with their supervisors or themselves in the case of practitioners in a solo or group practice. The results provide
an estimate of the time/effort reported for a putative 40-h work week because respondents were advised to exclude ⬙off-the-clock⬙ commitments
for any activity that might be accomplished after normal working hours or on weekends, holidays, or vacation periods.
b
Clinical care providers refer to respondents that are private practitioners and are owners/partners of a solo or group practice, or are employed by
a group practice, health system, or hospital.
c
AMC refers to respondents employed at any medical school or teaching hospital accredited to sponsor a residency program in internal medicine
or pediatrics, or a fellowship program in adult or pediatric endocrinology, or both as approved by the Accreditation Council for Graduate Medical
Education (http://www.acgme.org).
d
Federal government includes all respondents employed by the Department of Health and Human Services (HHS), Department of Defense (DOD),
or the Department of Veterans Affairs (VA).
e
Includes the total time devoted to teaching medical students, residents, or fellows.
f
Includes the total time devoted to administrative work including committee assignments.
J Clin Endocrinol Metab, April 2011, 96(4):923–933 jcem.endojournals.org 927

TABLE 3. Dissection of employers and the investigative work pursued by clinically trained members of TES early in
their careers
Workplace and/or employer (%)a
Clinical care Federal government Pharmaceutical or
Type of investigative work providerb (HHS/DOD/VA)c biotech industry AMCsd
None 64.1 12.5 9.3 9.8
Basic 0.6 12.5 9.3 13.9
Translational 1.3 20.3 14.8 16.1
Disease-oriented 4.5 28.1 13.0 21.3
Patient-oriented 21.2 15.6 40.7 22.9
Population-oriented 5.1 4.7 11.4 9.8

Downloaded from https://academic.oup.com/jcem/article/96/4/923/2720836 by Anelis user on 02 November 2023

Prevention-oriented 3.2 6.3 1.9 6.2
a
Each value is expressed as the mean percentage of work pursued by clinically trained endocrinologists within all types of workplaces.
Respondents consisted of 386 women (48.3%) and 414 men (51.7%) providing a total of 800 responses from 817 participants. The total number
of respondents within each employment sector is as follows: 304 clinical care providers, 433 AMCs, 32 federal, and 31 pharmaceutical/biotech.
The 800 respondents identified a total of 1215 research activities. Respondents were allowed to specify one or more of the designated activities to
reflect the type(s) of investigative work pursued in the 12 months preceding the survey. Individuals designating two lines of investigative work
were assumed to spend 50% effort on each endeavor, those designating three lines were assumed to spend 33.3% effort on each endeavor. No
respondent pursued more than three types of investigative work. Research activities were defined using terminology adopted by others (10 –12),
and appeared as follows in the survey questionnaire: basic research—laboratory-based research involving the development of new drugs,
technologies, or devices; translational research— bench to bedside or bidirectional research involving human subjects known to the investigator
关use of human specimens (cells/tissues) for laboratory studies was excluded兴; disease-oriented research—requires use of human subjects to
investigate the mechanisms or natural history of disease, or improve the detection or diagnosis of disease; patient-oriented research— clinical trials,
including phase I, II, III, IV trials of drugs, biologics, devices, and the evaluation of therapeutic interventions; population-oriented
research— outcome studies of populations, health services, and cost effectiveness research, studies of health quality including best practices and
medical errors, epidemiology and genetic studies, and community-based clinical trials; and prevention-oriented research—primary and secondary
prevention of disease in patients and health promotion via behavioral modification.
b
Clinical care providers refer to respondents that are owners/partners of a solo or group practice, or are employed by a group practice, health
system, or hospital.
c
Federal government includes all respondents employed by the Department of Health and Human Services (HHS), Department of Defense (DOD),
or the Department of Veterans Affairs (VA).
d
AMCs refers to respondents employed by any medical school or teaching hospital accredited to sponsor a residency program in internal medicine
or pediatrics, or a fellowship program in adult or pediatric endocrinology, or both as approved by the Accreditation Council for Graduate Medical
Education (http://www.acgme.org).

patient-, and basic-oriented research as indicated by re- workplace may be confounded by disproportionate num-
spective efforts of 20.3, 15.6, and 12.5%. A respective 4.7 bers of academics with clinical service commitments of
and 6.3% of federal respondents are involved in preven- 80% or more. This outcome is unlikely because respon-
tion- and population-oriented studies (Table 3). dent time commitments—for all tasks—are normally and
About 90.7% of the pharmaceutical/biotech workforce randomly distributed within and among employment sec-
is involved in research (Table 3). Patient-oriented studies tors, and estimates of the variance are consistent within
prevail among industrial respondents, as indicated by and among employment sectors as determined by the SEM
commitments of 40.7% (P ⬍ 0.01). Participation in trans- (Table 2).
lation-, disease-, population-, and basic-oriented research
is comparable (P ⬎ 0.25), as noted by respective efforts of Research funding
14.8, 13.0, 11.4, and 9.3%. Prevention research is limited We examined the sources of extra- and intramural
to 1.9% within the pharmaceutical/biotech workforce funds used to pay a portion of respondent salaries for time
(Table 3). spent on research. The estimates are independent of per-
The research activities of academic respondents differ sonal compensation but establish the fraction of respon-
from those of other workplaces. Respondent interest in dents that rely on extra- or intramural funds to pursue
patient-, disease-, translation-, and basic-oriented studies, research (Table 4).
for instance, is distributed rather uniformly, with efforts We found that 27.7% of respondents used grants/con-
ranging from 13.9 to 22.9% (Table 3). Population- and tracts awarded by federal, industrial, or philanthropic
prevention-oriented studies involve a respective 9.8 and sources to defray portions of their salaries for time spent
6.2% of academics. Only prevention research ranks below on research. The remaining 72.3% of respondents have
other research activities (P ⬍ 0.05). We considered the intramural funds to cover a portion of their salaries for
possibility that mean research time within the academic time spent on research. Intramural funding sources in-
928 Desjardins et al. Professional Activities of Endocrinologists J Clin Endocrinol Metab, April 2011, 96(4):923–933

TABLE 4. Analysis of proposals submitted and awards received by clinically trained members of TES early in their
careers for postfellowship research training and research project grants
Sources of funding available to clinically trained endocrinologists

Type of proposal or HHSa

grant award NIH Other HHSb VAa DODa Philanthropy Industry Other
Training (post fellowship)
Proposals submitted (%) 44.4 3.2 5.7 2.0 22.3 15.7 6.7
Proposals/applicantc 1.8 ⫾ 0.1 1.8 ⫾ 0.5 2.1 ⫾ 0.2 2.0 ⫾ 0.4 1.8 ⫾ 0.1 2.3 ⫾ 0.2 1.8 ⫾ 0.3
Success/applicant (%) 20.1 11.8 80.0 28.6 NA NA NA
Research as PI
Proposals submitted (%) 38.1 3.1 2.7 2.3 27.4 19.1 7.3

Downloaded from https://academic.oup.com/jcem/article/96/4/923/2720836 by Anelis user on 02 November 2023

Proposals/applicantc 2.8 ⫾ 0.2 2.1 ⫾ 0.6 1.9 ⫾ 0.3 2.2 ⫾ 0.4 2.9 ⫾ 0.2 2.7 ⫾ 0.2 2.3 ⫾ 0.3
Grants /applicantc 2.1 ⫾ 0.1 1.3 ⫾ 0.3 1.7 ⫾ 0.3 1.8 ⫾ 0.6 2.0 ⫾ 0.2 2.3 ⫾ 0.2 2.0 ⫾ 0.2
Success/applicant (%) 48.4 33.3 42.1 56.3 53.5 76.9 73.6
Research as Co-PI
Proposals submitted (%) 62.7 0.5 3.7 1.4 13.1 12.5 6.1
Proposals/applicantc 2.3 ⫾ 0.2 1.0 ⫾ 0.0 2.0 ⫾ 1.0 1.0 ⫾ 0.0 1.4 ⫾ 0.1 1.8 ⫾ 0.3 1.8 ⫾ 0.4
Grants/applicantc 1.6 ⫾ 0.1 1.0 ⫾ 0.0 1.0 ⫾ 0.0 1.0 ⫾ 0.0 1.0 ⫾ 0.0 1.8 ⫾ 0.3 1.4 ⫾ 0.3
Success/applicant (%) 51.1 100 28.6 100 50.0 96.3 58.2
NA, Data not available.
a
Federal departments: NIH, National Institutes of Health; HHS, Department of Health and Human Services; VA, Veterans Affairs; DOD, Department
of Defense.
b
Other HHS departments: Agency for Healthcare Research and Quality, Centers for Disease Control and Prevention, Food and Drug
Administration, Health Resources and Services Administration, and Substance Abuse and Mental Health Services Administration.
c
Values are expressed as mean percentage or the mean ⫾ SEM. The submission of postfellowship training proposals is based on 202 applicants out
of a total of 817 respondents (24.7%) that produced 687 proposals for consideration by federal and nonfederal agencies. A total of 166
individuals out of 817 respondents (20.4%) produced 305 applications for NIH K-series awards. Estimates for research project grants are based on
a total of 940 proposals submitted to federal and nonfederal agencies by 167 PIs out of 817 respondents (20.5%) and a total 230 proposals
produced for consideration by federal and nonfederal agencies by a cohort of 82 Co-PIs from a total sample of 817 respondents (10.1%).

clude: clinical earnings, 24.7%; medical schools and/or a career success rate of 48.4% (Table 4). Applications
hospitals, 21.3%; and endowments or other sources, submitted to the Veterans Administration and Depart-
9.3%. Federal and pharmaceutical/biotech employers ment of Health and Human Services agencies (excluding
support the full salaries of 13.2 and 3.8% of respondents, NIH) are as successful as those considered by NIH (Table
respectively. 4). The success rate of proposals considered by philan-
We examined the number of proposals submitted and thropy and industry is 53.5 and 76.9%, respectively (Ta-
grants/contracts awarded to endocrinologists (Table 4). ble 4). Applications for research project grants involving
An estimated 24.7% of endocrinologists produced pro- co-PIs are as successful (P ⬎ 0.15) as those produced by PIs
posals for postfellowship research training from federal (Table 4).
and nonfederal sources. Respondents submitted an aver-
age of 1.8 proposals for NIH K08/23 awards, with a suc- Research commitments: individuals and
cess rate of 20.1% (Table 4). We tested, by multiple re- institutions
gression, whether an association exists between receiving We queried respondents about their enthusiasm for in-
a K08/23 award and explanatory variables: gender, years vestigative careers as first-year medical students, resi-
post fellowship, and the receipt of federal research project dents, or fellows. The proportion of positive responses
grants as either a PI or co-PI. A decided association is (yes) increased from 38.7% as medical students to 60.0%
evident (P ⬍ 0.001) between receiving a K08/23 award as residents and 74.3% as fellows. We asked whether re-
and the subsequent receipt of research project grants from spondents were aware of opportunities to pursue faculty-
federal and nonfederal agencies as a PI, but not as a co-PI. sponsored research projects. Positive responses (yes) in-
The association between K08/23 awards and research creased from 26.2% as medical students to 27.9% as
project grants is independent of gender but dependent on residents and 75.2% as fellows. We tested whether first-
postfellowship experience (P ⬍ 0.001). year residents and fellows are coached or encouraged to
Respondents submitted a total of 1170 proposals for participate in faculty-directed research. We found that
research project grants as PIs and co-PIs (Table 4). Pro- 27.3% of respondents reported being encouraged to par-
spective PIs produced an average of 2.8 proposals for con- ticipate in faculty research as residents, whereas 69.3%
sideration by NIH and received an average of 2.1 awards, were encouraged similarly as fellows.
J Clin Endocrinol Metab, April 2011, 96(4):923–933 jcem.endojournals.org 929

We evaluated the institutional resources available to

respondents to pursue investigative work as fellows. Ap-
proximately 53.4% of respondents accessed institutional
support to prepare IRB protocols as fellows, 60.0% used
institutional facilities for the acquisition/management of
clinical data, and 71.3% relied on institutional core facil-
ities for laboratory and clinical research. Only 51.0% of
respondents, however, had access to administrative sup-
port to produce applications for postfellowship research
training grants and/or research project grants as fellows.

Downloaded from https://academic.oup.com/jcem/article/96/4/923/2720836 by Anelis user on 02 November 2023

We determined the investment made by academic med-
ical centers (AMCs) in junior faculty by asking respon-
dents to specify the total dollar value of support they re-
ceived for personnel, equipment, supplies, and other
research expenses. The results indicate a linear relation-
ship between the number of half-days per week devoted to
research and institutional investments made to support the
nascent research programs of early-career academics (Fig.
1, top).
We asked respondents employed by AMCs whether
their institution granted tenure and whether they held a
tenure-eligible post. Tenure opportunities were limited to
50.0% of AMCs, and 40.1% of female and 59.9% of male
respondents were appointed to tenure-eligible faculty po-
sitions. Gender disparity is evident (P ⬍ 0.01) when the
regression model controlled for total time worked, half-
days per week spent on research, type of employer, and
FIG. 1. Top, Each bar designates the mean ⫾ SEM of institutional
years post fellowship. funds (in constant dollars) provided over 3 yr to 411 clinically
trained endocrinologists employed at AMCs as a function of half-
days per week devoted to research. Middle, The equation/line
Restraints on investigative careers describes the relationship between years of postfellowship
Approximately 69.7% of respondents reported that re- experience for 411 clinically trained endocrinologists (employed at
AMCs) and mean annual earnings. Mean pretax earnings increased
searchers receive less compensation than clinical care spe-
by 1.8% per year for each year of postfellowship experience based
cialists (P ⬍ 0.01). Similarly, 70.1% of respondents noted on a Tobit regression model. Bottom, The equation/line describes
that research careers are unattractive relative to those in- the relationship between half-days per week devoted to research
volving clinical care because of expectations to produce and mean annual earnings. Mean pretax earnings decrease by 2.8%
per year for each half-day per week spent on research among 411
publications and obtain extramural research support (P ⬍ early-career clinical endocrinologists employed at AMCs, based on a
0.01). About 58.8% of respondents note that job uncer- Tobit regression model.
tainty is a significant issue for researchers vs. a career in the
clinical service sector (P ⬍ 0.05). Only 50.3% of respon- experience (Fig. 1, middle). The annual compensation of
dents considered that the time and energy for pursuing a women in this cohort, however, is about 22.1% below
successful career in research exceeded that required for a that of men (P ⬍ 0.01), based on a similar regression model
successful career focused on clinical care. designating the natural logarithm of pretax annual earn-
We tested the perceived earnings differential involving ings as a function of a dichotomous variable for gender, in
investigative careers among physicians at AMCs by using addition to total time worked per week, half-days per
a regression model specifying the natural logarithm of pre- week spent on research, type of employer, and years post
tax annual earnings as a function of two key explanatory fellowship. Lastly, we used a regression model stipulating
variables (years post fellowship and half-days per week the natural logarithm of pretax annual earnings as a func-
devoted to research), as well as variables to control for tion of one explanatory variable (two or more half-days
gender, type of employment, and total work hours per per week devoted to research), as well as variables to con-
week. The results demonstrate that the pretax annual trol for gender, total time worked, type of employer, and
earnings of endocrinologists affiliated with AMCs in- years post fellowship. The results establish that annual
crease linearly, averaging 1.8% per year of postfellowship earnings drop by 2.8% for each half-day per week spent on
930 Desjardins et al. Professional Activities of Endocrinologists J Clin Endocrinol Metab, April 2011, 96(4):923–933

research (P ⬍ 0.01) relative to respondents spending 1 tors (Table 3). Note that patient-oriented research is the
half-day/wk or less on research (Fig. 1, bottom). The an- principal activity within pharmaceutical workplaces, ex-
nual earnings loss of women and men are similar if they ceeding other types of research by about 3-fold. Research
spend 2 or more half-days/wk on research. Based on our interests within the federal and academic sectors, in con-
model (Fig. 1, bottom), respondents devoting 3 half- trast to other workplaces, are distributed rather uni-
days/wk to research with pretax earnings of $135,000/yr formly, except for the minimal involvement in population
would earn about $1,134 less than counterparts spending and prevention studies.
1 half-day/wk or less on research. Respondent participation in prevention-oriented re-
search, compared with other research activities, is trivial
among all workplaces (Table 3). We did not determine
Discussion

Downloaded from https://academic.oup.com/jcem/article/96/4/923/2720836 by Anelis user on 02 November 2023

why respondents are uninvolved in prevention research,
but the meager fraction of involved endocrinologists mer-
This study relied on physician members of TES to evaluate
its the consideration of stakeholders in clinical research—
the feasibility of using a Web-based survey tool for assess-
academic, federal, industrial, philanthropic, and profes-
ing the professional work of early-career physicians
sional. The unequivocal gap in prevention studies in
within a medical subspecialty. The findings provide fresh
endocrinology provides a singular opportunity for stake-
insights about the work of early-career endocrinologists,
holders to champion interventions for improving preven-
document the clinical commitments and research accom-
tion-oriented research. Opportunities for preempting en-
plished within all types of workplaces, and extend knowl-
docrine disorders are burgeoning (15–18), and compelling
edge about the challenges limiting the pursuit of investi-
gative careers among clinically trained endocrinologists. new strategies are available to investigate risk assessment,
Our estimates of the half-days per week devoted to disease tracking, and personalized therapy (19). Enlarging
clinical service, teaching, research, and administration investments in the training and conduct of prevention-
offer a contemporary snapshot of the professional re- oriented research would facilitate the transformation of
sponsibilities of physician endocrinologists (Table 2) clinical endocrinology from a curative to preventive focus
and demonstrate that the percentage effort devoted to with an emphasis on preemptive medicine and personal
individual tasks is employer dependent (Table 3). A health care planning (19).
novel aspect of these findings emerges from considering The analysis of proposals produced by early-career re-
the extent and type of research accomplished within spondents uncovered a compelling association between
individual workplaces. receiving an K08/23 award and the subsequent receipt of
Research of all types (basic, translational, disease, pa- federal research project grants as a PI, but not as a co-PI.
tient, population, and prevention) is accomplished within We did not consider all of the variables needed to predict
each workplace, but the interest in particular pursuits is the long-term success of K-awardees as future recipients of
emblematic of individual employment sectors. Within the federal research grants (20, 21). But, the robust correla-
clinical care workplace, for instance, patient-oriented re- tion (P ⬍ 0.001) between K08/23 award recipients and
search prevails by 4- to 5-fold over other research. The research project grants suggests increasing the number of
implication embedded in this finding is that clinical care K08/23 awards to bolster interest in prevention- and pop-
providers participate in one or more types of patient-ori- ulation-oriented research (Table 3).
ented research involving practice-based research net- Our assessment of the academic workforce indicates
works, patient registries for cohort studies, and patient that endocrine-related research within AMCs is lever-
enrollment in multicenter trials of new drugs and proce- aged largely on intramural as opposed to extramural
dures or both. For reasons of practicality, the survey ques- funds. This conclusion is based on finding that 72.3%
tionnaire did not ask clinical care providers to differentiate of academic endocrinologists rely on intramural funds
among the specified investigative approaches. It seems rea- to cover portions of their salaries for time spent on
sonable to suggest, however, that clinical care providers research, whereas only 27.7% of academics reported
likely collaborate in one or more patient-oriented proto- salary support from extramural grants/contracts. Dis-
cols developed and funded through AMCs, federal, or in- section of intramural funding streams revealed that clin-
dustrial sponsors because the time available for research ical earnings accounted for almost 25% of the salary sup-
represents about 16.0% of their total professional effort port of early-career academics, or about the same level of
(Table 2). salary support derived from extramural grants/contracts.
Other evidence that workplaces shape research pur- Although AMCs are committed to supporting the research
suits is based on the profile of investigative activities ob- activities of newly appointed clinical faculty, significant
served within academic, federal, and pharmaceutical sec- portions of the investigative enterprise remain vulnerable
J Clin Endocrinol Metab, April 2011, 96(4):923–933 jcem.endojournals.org 931

to unpredictable changes in health care reimbursement. In crine-based clinical research because the ethnic back-
addition, postfellowship research training in the form of ground of physicians directing clinical protocols shapes
K08/23 awards is limited to an estimated 4.0% of respon- the ethnic profile of volunteers agreeing to participate in a
dents (Table 4). clinical trial or study (31). Participation of underrepre-
Beyond the set of constraints identified above resides an sented ethnic groups in clinical trials and studies is essen-
anachronistic, ailing, and highly fragmented federal-insti- tial (32, 33) because reliable estimates of the safety and
tutional partnership for supporting clinical research (11, efficacy of new and existing drugs or treatments are un-
22, 23). For instance, assessments of extramural research achievable unless the subjects enrolling in clinical trials
funding among individual institutes within NIH highlight mirror the demographic norm of the U.S. population at
the difficulty of supporting clinical research in general and large (31). The need to reconcile ethnic disparities in the

Downloaded from https://academic.oup.com/jcem/article/96/4/923/2720836 by Anelis user on 02 November 2023

a lack of enthusiasm for population-oriented research future workforce for endocrinology is a pressing concern
within certain institutes (24). Furthermore, a disparity ap- requiring interventions for the delivery of clinical care to
proaching 2:1 has prevailed for funding basic vs. clinical all demographic groups (34 –36).
research across NIH (25, 26). The summed challenges This study has several strengths and limitations. From
confronting the academic workforce and those within the the standpoint of strengths, the findings indicate that
federal-institutional partnership provide the rationale for Web-based surveys offer a mechanism to deconstruct the
proposing stakeholder interventions to improve clinical professional work of subspecialty physicians with an eye
research on endocrine disorders through predictable, sus- toward the application of improvement tools for testing
tained, and prudent support of clinically oriented inves- explanatory variables and developing interventions to im-
tigative work. The thrust embedded in the proposed in- prove clinical investigation and its practitioners. Access to
tervention is to discreetly enlarge the fraction of physician timely data within a clinical subspecialty fills a void oc-
endocrinologists with postfellowship research experi- casioned by the lack of a national protocol for pinpointing
ences (K-awardees) so they can, in turn, gain extramural deficits in the scope and scale of clinical research.
support for research project grants and have access to the From the perspective of weaknesses, surveys rely on
time and funds to pursue investigative careers. self-reported estimates of individual performance and per-
Our assessment of the ambivalence surrounding inves- sonal experience. The credibility and validity of survey
tigative careers revealed that respondents perceived a sig- data are well documented, given that questionnaires ad-
nificant fiscal penalty for participating in research. We here to the general principles of survey design (37). Survey
verified this perception by showing that pretax annual effectiveness is limited by respondent participation. The
earnings of academic investigators fall by 2.8% for each 29.6% participation in this study exceeded the 95% level
half-day per week devoted to research (Fig. 1, bottom). of confidence, but a more robust response rate would ex-
Women and men in academic careers share the earnings tend confidence in the findings and the generalizations
adjustment for research time equally. The annual com- derived from them. Improvements in response rates may
pensation of female respondents in academic settings, be achieved through presurvey communications with pro-
however, is 22% below that of males, a well-documented spective respondents, and by a follow-up telephone call to
earnings disparity among women physicians (27, 28). Ap- nonresponders. The cross-sectional design adopted for
proximately two thirds of respondents identified job un- this study limits the power of this report because the design
certainty in the academic sector as a significant limitation lacks controls for establishing cause-and-effect relation-
for pursuing investigative careers. This perception is con- ships. This study was undertaken as a pilot project to in-
sistent with the persistent erosion in tenure-track appoint- form the design of future assessments and permit hypoth-
ments for clinical faculty employed by AMCs (29) and eses to be tested based on the findings presented here. Our
parallels the career uncertainties voiced by academic pul- workforce assessments, for instance, are limited by the
monary physicians early in their careers (30). absence of background variables defining the specific
The self-identified ethnic backgrounds of survey re- types of the clinical, didactic, investigative, and adminis-
spondents approximate those of physicians graduating trative activities pursued by respondents. In the same vein,
from allopathic schools of medicine in the United States additional background information about employers
between 1991–2005 (Table 1). One noteworthy disparity would offer greater insight about respondent work within
exists, however. The fraction of Black or African-Ameri- all employment sectors: academic, clinical care, federal,
can endocrinologists is about 3.7% below the mean set by and pharmaceutical/biotech. The need to capture addi-
graduates of the nation’s medical schools (Table 1). This tional background variables, however, is a challenge be-
issue merits deliberate and resolute attention of training cause surveys involving physicians are best limited to 10
programs in clinical endocrinology. It also impacts endo- min of respondent effort. Finally, our results are based on
932 Desjardins et al. Professional Activities of Endocrinologists J Clin Endocrinol Metab, April 2011, 96(4):923–933

a time horizon ranging from 1991–2005, an interval rec- tors. Washington, DC. http://www.nsf.gov/statistics/indicators/, ac-
cessed May 12, 2010
ognized for the generous support for research.
3. Kelley WN, Randolph MA 1994 Careers in clinical research: ob-
The present assessment provides a benchmark for esti- stacles and opportunities. Washington, DC: National Academy
mating the scope and setting of workplace activities pursued Press; 1–21
by early-career physician members of TES or other endo- 4. Nathan DG, Varmus HE 2000 The National Institutes of Health and
clinical research: a progress report. Nat Med 6:1201–1204
crine-related clinical groups. Significant transitions in the 5. Sung NS, Crowley Jr WF, Genel M, Salber P, Sandy L, Sherwood
workforce and workplace can be anticipated in the future as LM, Johnson SB, Catanese V, Tilson H, Getz K, Larson EL, Schei-
a result of Roadmap and Clinical and Translational Science nberg D, Reece EA, Slavkin H, Dobs A, Grebb J, Martinez RA, Korn
A, Rimoin D 2003 Central challenges facing the national clinical
Awards initiatives championed by NIH (38 – 40), the flat-
research enterprise. JAMA 289:1278 –1287
tening of the federal budget for research, the passage of health 6. Berwick DM 2008 The science of improvement. JAMA 299:1182–

Downloaded from https://academic.oup.com/jcem/article/96/4/923/2720836 by Anelis user on 02 November 2023

care reform legislation, and the work environment prefer- 1184
ences of a new generation of clinically trained endocrinolo- 7. Association of American Medical Colleges 2008 Medical schools
and teaching hospitals by the numbers. AAMC Data Book. Wash-
gists. Our findings provide a starting point for initiating a ington, DC
national dialog aimed at the application of improvement 8. National Institutes of Health 2008 Average age of principal investiga-
tools to enhance the quality of clinical investigation within all tors. Office of Extramural Research, Extramural Data Handbook.
Bethesda, MD. http://search2.google.cit.nih.gov/search?q⫽cache:
workplaces. The present results, moreover, underscore the qJT97J0CyvQJ:report.nih.gov/RPG_PIs_AAMCFaculty08252007.
need to improve the coordination, collaboration, and inte- ppt⫹age&site⫽REPORT&client⫽REPORT_frontend&proxystyle
gration of the nation’s investment in clinical research among sheet⫽REPORT_frontend&output⫽xml_no_dtd&ie⫽UTF-8&
access⫽p&oe⫽ISO-8859 –1, accessed May 12, 2010
all partners in the investigative enterprise: federal, institu-
9. Brotherton SE, Etzel SI 2006 Graduate medical education, 2005–
tional, and state (11, 23). Only then will a bright future be 2006. JAMA 296:1154 –1169
assured for medicine and the growing number of patients 10. Association of American Medical Colleges 1988 Breaking the sci-
that suffer from the chronic debilitating effects of endocrine entific bottleneck: report of the Graylyn Consensus Development
Conference. Washington, DC: Association of American Medical
disorders (41). Colleges
11. Heinig SJ, Quon AS, Meyer RE, Korn D 1999 The changing land-
scape for clinical research. Acad Med 74:726 –745
12. Rosenberg LE 1999 Physician-scientists: endangered and essential.
Acknowledgments Science 283:331–332
13. Trochim W 2000 The research methods knowledge base. 2nd ed.
The authors thank each of the clinical endocrinologists who took Cincinnati, OH: Atomic Dog Publishing; 15–16
14. Lansing JB, Morgan JN 1974 Economic survey methods. Ann Ar-
the time to respond and return a survey questionnaire. Loretta bor, MI: The University of Michigan, Survey Research Center, In-
Doan, Ph.D., at The Endocrine Society assisted with all phases of stitute for Social Research; 33– 41
this project including the designation of clinical endocrinologists 15. Scheuner MT, Sieverding P, Shekelle PG 2008 Delivery of genomic
who took part in the pilot and full surveys. Thanks are also due medicine for common adult diseases: a systematic review. JAMA
the staff of the Survey Research Institute at Cornell University for 299:1320 –1334
16. Agurs-Collins T, Khoury MJ, Simon-Morton D, Olster DH, Harris
programming and administering the survey, and to Kenneth T. JR, Milner JA 2008 Public health genomics: translating obesity
Whelan, Cornell Higher Education Research Institute, Cornell genomics research into population health benefits. Obesity (Silver
University, Ithaca, New York, for managing the statistical anal- Spring) 16 (Suppl 3):S85–S94
ysis of the survey project. C.D. acknowledges the encouragement 17. Hariri S, Yoon PW, Qureshi N, Valdez R, Scheuner MT, Khoury MJ
provided by Gerald S. Moss, M.D., Dean, College of Medicine, 2006 Family history of type 2 diabetes: a population-based screening
tool for prevention. Genet Med 8:102–108
University of Illinois at Chicago, for his early endorsement of the
18. Valdez R, Yoon PW, Qureshi N, Green RF, Khoury MJ 2010 Family
clinical scholars project. history in public health practice: a genomic tool for disease preven-
tion and health promotion. Annu Rev Public Health 31:69 – 87
Address all correspondence and requests for reprints to: 19. Snyderman R, Yoediono Z 2008 Prospective health care and the role
Claude Desjardins, Clinical Scholars Project and the Department of academic medicine: lead, follow, or get out of the way. Acad Med
of Physiology and Biophysics, College of Medicine, University of 83:707–714
Illinois at Chicago, Chicago, Illinois 60612-4325. E-mail: 20. Toto RD 2008 How to launch a successful career in clinical research:
clauded@uic.edu. tips on making the most of available resources. J Investig Med 56:
843– 846
Disclosure Summary: The authors have nothing to disclose.
21. National Institutes of Health, Office of Extramural Research 2009
Mentored clinical scientist research career development award (K-08).
Bethesda,MD.http://grants2.nih.gov/grants/guide/pa-files/PA-09-042.
html, accessed May 12, 2010
References 22. Crowley Jr WF, Sherwood L, Salber P, Scheinberg D, Slavkin H,
Tilson H, Reece EA, Catanese V, Johnson SB, Dobs A, Genel M,
1. National Science Foundation, Division of Science Resources Statis- Korn A, Reame N, Bonow R, Grebb J, Rimoin D 2004 Clinical
tics 2010 Science and engineering statistics. Washington, DC. research in the United States at the crossroads. Proposal for a novel
http://www.nsf.gov/statistics/, accessed May 12, 2010 public-private partnership to establish a national clinical research
2. National Science Foundation 2010 Science and engineering indica- enterprise. JAMA 291:1120 –1126
J Clin Endocrinol Metab, April 2011, 96(4):923–933 jcem.endojournals.org 933

23. Crowley Jr WF, Gusella JF 2009 Changing models of biomedical 33. Saha S, Arbelaez JJ, Cooper LA 2003 Patient-physician relationships
research. Sci Transl Med 1:1–5 and racial disparities in the quality of health care. Am J Public Health
24. Institute of Medicine (US) 2003 Enhancing the vitality of the Na- 93:1713–1719
tional Institutes of Health: organizational change to meet new chal- 34. Cohen JJ, Gabriel BA, Terrell C 2002 The case for diversity in the
lenges. Washington, DC: National Academy Press; 1–16 health care workforce. Health Aff (Millwood) 21:90 –102
25. Nathan DG, Wilson JD 2003 Clinical research and the NIH—a 35. Hahm J, Ommaya A 2006 Opportunities to address clinical research
report card. N Engl J Med 349:1860 –1865 workforce diversity needs for 2010. Washington, DC: National
26. Moses 3rd H, Dorsey ER, Matheson DH, Thier SO 2005 Financial Academy Press; 1– 6
anatomy of biomedical research. JAMA 294:1333–1342 36. Olson S, Fagen AP 2007 Understanding interventions that encour-
27. Baker LC 1996 Differences in earnings between male and female
age minorities to pursue research careers: summary of a workshop.
physicians. N Engl J Med 334:960 –964
Washington, DC: National Academy Press; 38 –54
28. Wright AL, Ryan K, St Germain P, Schwindt L, Sager R, Reed KL
37. Fowler F 2009 Survey research methods. 4th ed. Thousand Oaks,
2007 Compensation in academic medicine: progress toward gender
CA: SAGE Publications; 63–72

Downloaded from https://academic.oup.com/jcem/article/96/4/923/2720836 by Anelis user on 02 November 2023

equity. J Gen Intern Med 22:1398 –1402
38. Zerhouni E 2003 The NIH roadmap. Science 302:63–72
29. Bunton SA, Mallon WT 2007 The continued evolution of faculty
appointment and tenure policies at U.S. medical schools. Acad Med 39. Morrison L 2008 The CTSA’s, the Congress, and the scientific
82:281–289 method. J Investig Med 56:7–10
30. Weinert CR, Billings J, Ryan R, Ingbar DH 2006 Academic and 40. Collins FS 2010 Opportunities for research and NIH. Science 327:
career development of pulmonary and critical care physician-scien- 36 –37
tists. Am J Respir Crit Care Med 173:23–31 41. Daar AS, Singer PA, Persad DL, Pramming SK, Matthews DR,
31. Getz K, Faden L 2008 Racial disparities among clinical research Beaglehole R, Bernstein A, Borysiewicz LK, Colagiuri S, Ganguly N,
investigators. Am J Ther 15:3–11 Glass RI, Finegood DT, Koplan J, Nabel EG, Sarna G, Sarrafzade-
32. Betancourt JR 2006 Eliminating racial and ethnic disparities in health gan N, Smith R, Yach D, Bell J 2007 Grand challenges in chronic
care: what is the role of academic medicine? Acad Med 81:788 –792 non-communicable diseases. Nature 450:494 – 496

Visit the Online Store

for the latest information in endocrinology!
Check out CME Self-Assessment & MOC products.
www.endo-society.org