Warshaw GA, Bragg EJ, Shaull RW, Lindsell CJ. Academic Geriatric Programs in US Allopathic and Osteopathic Medical Schools. JAMA. 2002;288(18):2313-2319. doi:10.1001/jama.288.18.2313
Author Affiliations: Office of Geriatric Medicine and Department of Family Medicine (Dr Warshaw) and Institute for Health Policy and Health Services Research (Drs Bragg and Lindsell and Ms Shaull), University of Cincinnati, Cincinnati, Ohio.
Context By 2030, 20% of the US population will be older than 65 years compared
with 12.4% in 2000. The development of geriatric medicine research and training
programs to prepare for this increasing number of older individuals is largely
dependent on the successful establishment of academic geriatric medicine programs
in medical schools.
Objective To assess the structure, resources, and activities of academic geriatric
medicine programs in US allopathic and osteopathic schools of medicine.
Design, Setting, and Participants Survey distributed to the academic geriatric medicine leaders of the
144 US allopathic and osteopathic medical schools in March 2001.
Main Outcome Measures Organizational structure, program information, curriculum, budgetary
issues, and characteristics of academic geriatric medicine leaders.
Results A total of 121 program directors (84%) responded. Most schools (87%)
had an identifiable academic geriatric program structure, with 67% established
after 1984. The greatest proportion of faculty and staff time (40%) was spent
in clinical practice, followed by research and scholarship (12%), residency
and fellowship education (10% each), and medical student education (7.8%).
Clinical practice accounted for the greatest portion (27%) of revenue, and
25.7% of the programs had total annual budgets of less than $250 000,
while 11% had budgets of greater than $5 million. The largest obstacles to
achieving the goals of an academic geriatric medicine program were a lack
of research faculty and fellows and poor clinical reimbursement.
Conclusions Most US medical schools have an identifiable academic geriatric medicine
program; most have been established within the last 15 years. Resources are
needed to train faculty for roles as teachers and researchers and to develop
medical school geriatric programs of the size and scope equivalent to other
By 2030, 20% of the US population will be older than 65 years, up from
12.4% in 2000.1 The aging of the US population
will have a major impact on the practice of medicine and future health care
costs for the elderly population. The principles of geriatric medicine practice
that have been developed over the past 50 years, if widely applied to the
care of older Americans, could improve cost-effective, quality care for the
well elderly and older adults with chronic illness. In addition, research
is needed to advance prevention and treatment of the diseases that result
in the greatest functional loss among older individuals. Leadership, expertise,
and commitment are required from geriatrics as well as all medical and surgical
specialties and other health care disciplines.
Geriatric medicine integrates many aspects of medicine, including internal
medicine, family practice, neurology, psychiatry, and rehabilitation. In addition,
geriatric medicine emphasizes problems that are more common in older adults,
particularly confusion and dementia, depression, falls and instability, incontinence,
chronic pain management, sensory impairment, and end-of-life care, as well
as the frequent co-occurrence of many of these problems. In addition, the
delivery of medical services to older adults occurs not only in the familiar
office and hospital settings but also in the patient's home, retirement home,
rest home/assisted living facilities, day care, nursing home, and hospice
settings. Physicians work cooperatively with practitioners representing many
health care disciplines, such as nursing, social work, and the various therapies.
Physicians without training in geriatric medicine often lack the skills to
care for patients in less-familiar settings or to practice in interdisciplinary
A 1998 Association of American Medical Colleges (AAMC) report summarizing
Liaison Committee on Medical Education curriculum data found that required
geriatric medicine courses remained rare, although 98% of medical schools
reported some form of required geriatric medicine experience.2 In
1998 and 1999, more than 40% of allopathic medical students reported that
their medical schools' geriatric medicine curriculum time was inadequate.3 In 1999, family practice and internal medicine residency
programs graduated 9780 physicians, but only 269 subsequently entered geriatric
medicine fellowships.4 Also in 1999, psychiatry
residencies graduated 1056 physicians, with only 95 enrolling in geriatric
psychiatry fellowships.4 During the same year,
an additional 14 176 physicians graduated from other residency and fellowship
programs (excluding pediatrics) whose specialties do not offer subsequent
fellowships in geriatric medicine.4 Thus, formal
geriatric medicine training for virtually all physicians ends with their residency
training. Only 25 of the more than 90 Accreditation Council for Graduate Medical
Education (ACGME) specialty program requirements (excluding pediatrics) currently
make specific references to geriatric medicine training.5
Numerous national reports and health policy leaders have advocated for
the establishment of academic geriatric programs in US medical schools,6- 14 but
the development and characteristics of these programs have not been well documented.
We report results from the first comprehensive national survey to assess the
current status of US academic geriatric medicine programs.
The study was a cross-sectional survey of academic geriatric medicine
programs in recognized allopathic and osteopathic medical schools in the United
States. Subsequent surveys are planned at 3-year intervals. Longitudinal data
from the American Medical Association (AMA) and the AAMC's National Graduate
Medical Education (GME) Census were also analyzed to track geriatric medicine
and geriatric psychiatry fellowship positions and fill rates since 1995.
This survey is part of the Association of Directors of Geriatric Academic
Programs (ADGAP) Longitudinal Study of Training and Practice in Geriatric
Medicine. The longitudinal study will describe trends in academic geriatric
medicine training and practice and relate these trends to the changing need
for expertise in geriatric medicine. The Office of Geriatric Medicine and
the Institute for Health Policy and Health Services Research at the University
of Cincinnati Medical Center are developing the database. A national oversight
panel was established to review, advise on, and approve the project work plan
A survey was mailed to the academic leaders in geriatric medicine at
the 144 allopathic and osteopathic medical schools accredited by either the
AAMC or the American Association of Colleges of Osteopathic Medicine. The
academic leader was defined as the one physician or other faculty leader at
each medical school who was recognized by the dean as providing overall leadership
for the academic geriatric medicine program. The director of the geriatric
academic program (DGAP) for each school was identified by reviewing the membership
list of ADGAP, by consulting with known geriatric academic leaders, and by
searching the Web pages of each school. Immediately before the survey, contact
was made with each school to verify the DGAP for that school.
A 12-page, 24-question survey instrument was developed with input from
members of the project's national oversight panel and was pretested by 5 academic
leaders. The survey was divided into 5 parts: organizational structure, general
program information (including budgeted staff and obstacles to achieving goals
of the academic geriatric program), curriculum, budgetary issues, and information
about the academic leader.
The survey was mailed to the 144 DGAPs in March 2001. On the same date,
the survey was made available online; it was housed on a secure server and
password protected to prevent unauthorized access. The DGAPs were requested
to complete and return the mailed survey or to complete the online survey.
Reminder e-mails were sent to all DGAPs 1 week after the initial mailing and
to nonresponders 18 days after the initial mailing. A second copy of the survey
was mailed to nonresponders at 22 days, and reminder e-mails and postcards
were sent 7 days after this mailing.
The AMA and the AAMC collaborate to annually survey the geriatric medicine
fellowship programs accredited by the ACGME.4,15- 21 We
compiled this annual longitudinal data since 1995 to evaluate trends in the
number of geriatric fellowship positions and fellows in training.
Means and medians and SDs or ranges were used to describe continuous
data. Categorical data were described using frequencies and percentages. Correlations
between variables were tested using the Spearman rank correlation coefficient
(ρ). Differences in proportions were tested using χ2 tests.
Differences in ranked or continuous data between osteopathic and allopathic
schools were tested using the Mann-Whitney U test.
Analyses were performed using SAS version 8.222 and
SPSS version 10.1.0.23P≤.05 was considered significant for all analyses.
The survey response rate was 84%, with 121 DGAPs responding (18/19 osteopathic
schools and 103/125 allopathic schools). Responders and nonresponders did
not differ by the number of enrolled medical students (U = 1241; P = .41), but most nonresponders
(n = 11 [47.8%]) were in census region 31 (Delaware, Maryland, District of
Columbia, Virginia, West Virginia, North Carolina, South Carolina, Georgia,
and Florida) (χ2 = 27.127; P = .001).
Among allopathic schools, 81 (82%) of the DGAPs held their primary appointment
in the department of internal medicine and 13 (13%) in the department of family
practice; the remaining 5 (5%) had appointments in the dean's office (1%),
department of community health (1%), or department of geriatric medicine (3%).
Among osteopathic schools, 8 (45%) of the DGAPs held their primary appointment
in the department of family practice and 6 (33%) in the department of internal
medicine; the remaining 4 (22%) held appointments in the departments of geriatric
medicine (11%) or pharmacology (5.5%) or a physician assistant program (5.5%).
Overall, 52 (44%) of the DGAPs had been in their current position for
4 years or less (median, 5 years; range, <1 to 27 years for allopathic
schools; median, 3 years; range, <1 to 15 years for osteopathic schools).
Sixty-two DGAPs (53%) held the rank of professor. At allopathic schools,
59 DGAPs (60%) held tenured positions (vs 50% of all US allopathic medical
faculty at the professor or associate professor level who are tenured24). Among osteopathic schools, 4 DGAPs (22%) held tenured
positions. More than two thirds of the DGAPs without tenure (37 of 51) had
academic appointments that did not allow them to earn tenure in the future.
Among current DGAPs, 48 (41.7%) reported completing formal geriatric medicine
fellowship training and earning a certificate of added qualifications from
the American Board of Family Practice or the American Board of Internal Medicine.
Forty-two (36.6%) had earned a certificate of added qualifications through
the practice pathway, and the remaining 25 (21.7%) had not completed fellowship
training or earned a certificate of added qualifications.
The DGAPs at allopathic schools commonly had more than 1 reporting relationship;
26 (27%) reported to 2 or more individuals. Thirty-three allopathic DGAPs
(33%) reported directly to the dean, 67 (68%) reported to the chair of internal
medicine, and 12 (12%) reported to the chair of family medicine. At osteopathic
schools, 12 DGAPs (67%) reported directly to their dean, 2 (11%) reported
to the chair of internal medicine, and 4 (22%) reported to the chair of family
medicine. Many DGAPs were members of important medical center committees,
including curriculum (n = 30 [34%]), executive/steering (n = 27 [30%]), and
promotion/tenure (n = 21 [24%]). Ninety-seven DGAPs (83%) received institutional
support for a portion of their salary. The median support level was 50%, and
27 DGAPs (28%) received at least three quarters of their salary support directly
from their colleges.
Geriatric medicine programs are organized in a variety of structures
that include departments, divisions, sections within a division, a unit within
2 departments, or a separate program, center, or institute. Academic geriatric
units could be identified in 95 (92%) of allopathic and 10 (56%) of osteopathic
medical schools; 67% were established after 1984. Thirty-two programs reported
more than 1 structure but DGAPs were not always the leader of each of these
As of 2001, there were 6 departments of geriatric medicine, 3 in allopathic
schools (Mt Sinai School of Medicine, New York, NY; University of Arkansas,
Little Rock; and University of Oklahoma, Oklahoma City) and 3 in osteopathic
schools (Philadelphia College of Osteopathic Medicine, Philadelphia, Pa; Western
University of Health Science, Pomona, Calif; and Ohio University College of
Osteopathic Medicine, Athens). The most common academic unit was a division
or section within a division, a structure existing at 63 of the schools with
identifiable structures. Forty-two schools (38 allopathic and 4 osteopathic)
had an interdisciplinary structure consisting of a department; a unit within
2 or more departments; or a freestanding program, center, or institute.
The number of professional faculty and staff varied considerably among
the geriatric programs. For example, while the median number of physician
faculty was 5.0 full-time equivalents (FTEs), the range was 0 to 42 faculty.
Fifty-nine programs (50.9%) had fewer than 6 physician faculty, 37 (31.9%)
had 6 to less than 12, 11 (9.5%) had 12 to less than 18, and 9 (8%) had 18
or more physician faculty. The number of physician faculty was significantly
lower at the osteopathic schools (median, 2 FTEs; range, 0.5-22 FTEs) compared
with the allopathic schools (median, 6.5 FTEs; range, 0-42 FTEs) (U = 343; P<.001). Overall, only 35 schools
(30%) had 9 or more geriatric physician faculty, the number recommended by
the Institute of Medicine.Article Other faculty and
staff employed in geriatric academic programs are shown in Table 1.
Program budgets varied among the academic programs (Figure 1). Ten osteopathic programs (62.5%) reported budgets of
less than $250 000 per annum, while allopathic programs reported significantly
higher budgets; only 18 allopathic programs (19.3%) reported budgets of less
than $250 000 (U = 379; P = .001). Overall, 46 programs (42%) reported budgets greater than
$1 million per annum. Program budgets were not correlated with medical school
size, as measured by the number of enrolled medical students (ρ = 0.107; P = .27). Table 2 lists
the sources of revenue for allopathic and osteopathic schools. Most programs
had diverse revenue sources, with clinical practice representing an important
source of income; for all reporting schools, 27% of income came from clinical
practice. The Veterans Health Administration (VHA) provided 13.1% of revenue
for the allopathic academic geriatric programs. For medical schools with affiliated
geriatric research, education, and clinical centers, the VHA contribution
to total program revenue ranged from 5% to 48%. Five schools with geriatric
research, education, and clinical centers did not report the VHA as a revenue
source, so our data may slightly underestimate the overall contribution of
the VHA to academic geriatric programs. The percentage of program revenue
from educational and research grants and contracts was positively correlated
with the geriatric programs' annual budgets (ρ = 0.558; P<.001).
To address the critical shortage of geriatric faculty members in medical
schools, beginning in 1988 academic geriatric medicine centers of excellence
were designated and funded by the John A. Hartford Foundation of New York,
NY. The goal of the centers of excellence program is to enhance the training
and research productivity of selected academic geriatric programs, particularly
those with the ability to train future geriatrics faculty. The 27 responding
centers of excellence (of 28 total) had a median of 21.7 FTE physician faculty
and geriatric medicine or geriatric psychiatry fellows compared with a median
of 5.5 FTEs for other programs (U = 252; P<.001). The centers of excellence had a median of 13 FTE physician
faculty vs 3.6 FTEs at the other programs (U = 249.5; P<.001) and a median of 7 FTE fellows-in-training vs
2 FTEs for other programs (U = 300; P<.001). The centers of excellence had median annual budgets in
the category of $2 million to $5 million compared with $250 000 to $500 000
for other programs (U = 232.5; P<.001) and budget reserves of $125 000 to $250 000 vs
no reserve dollars for other programs (U = 527; P<.001). The centers of excellence received 14.1% of
their revenue from clinical practice vs 30.6% for other programs (U = 849.5; P = .02), 31.3% from research grants
and contracts vs 10.3% for other programs (U = 376; P<.001), and 7.3% from education grants/contracts vs
8.7% for other programs (U = 935.5; P = .08).
The missions of academic geriatric medicine programs, as with medical
schools in general, are diverse and complex. The allocation of geriatric medicine
faculty and staff time illustrates this diversity (Table 3). Schools devoted about 40% of their effort to clinical
practice (48% at osteopathic schools and 38% at allopathic schools). The allopathic
schools tended to devote a higher percentage of resources to research and
scholarship and residency and fellowship training than the osteopathic schools,
while osteopathic schools placed more emphasis on medical student education.
The DGAPs rated 9 potential obstacles to achieving their programs' goals
(Table 4). Lack of research faculty
and fellows and poor reimbursement for clinical care were each rated "significant"
by more than 60% of the DGAPs. Osteopathic DGAPs in particular emphasized
the lack of research faculty and also reported difficulty recruiting clinical
Future geriatrics faculty are trained at 181 allopathic geriatric medicine
and geriatric psychiatry fellowship programs and 7 osteopathic geriatric medicine
fellowship programs. Clinical certification now can be obtained with 1 year
of training in allopathic programs. Despite this shorter training period,
fill rates for geriatric medicine and psychiatry fellowship positions have
been decreasing in recent years (Table 5). In academic year 2001-2002, approximately 41% of allopathic geriatric
medicine fellows and 37% of geriatric psychiatry fellows were US medical school
graduates, compared with US medical school graduate fill rates of 60% for
cardiology, 56% for infectious disease, 89% for ophthalmology, and 82% for
general surgery.21 Academic development of
future geriatrics faculty generally requires 1 to 3 years of additional training
beyond clinical certification. In academic year 2000-2001, only 68 geriatric
medicine fellows (27%) and 3 geriatric psychiatry fellows (3%) continued training
beyond year 1.5
This survey of academic geriatric medicine programs in US medical schools
illustrates the significant progress academic geriatrics has made in the past
25 years, as well as the challenges that programs face in meeting the needs
of the aging population. The existence of a leader in geriatric medicine in
all 144 US medical schools and a distinct academic geriatric medicine program
in 105 (87%) of the 121 responding schools demonstrates remarkable progress.
Geriatric program structures vary, but it is too early to declare a
"best" approach. Although it has been stated that academic geriatric medicine
programs will develop most effectively in an environment that encourages cooperation
among medical disciplines, other health disciplines, and social scientists,25 the optimal organizational structure for geriatric
programs remains unclear. Medical schools are organized around discipline-specific
departments, with department leaders holding most of the power and control
for negotiating resources from the dean. In this context, academic geriatric
medicine programs have struggled to take root in many US medical schools despite
generous support from the federal government, especially the National Institute
on Aging and the VHA; some state governments; and the private sector, especially
the John A. Hartford Foundation.12 Even though
there are currently only 6 departments of geriatric medicine, one third of
the academic leaders report directly to their dean. This typically creates
access to new resources and continued interdepartmental influence. However,
many of the academic leaders are division chiefs within a single department,
and it remains to be seen if they can provide institutional leadership.
The relatively low number of fellowship-trained DGAPs (42% of the leaders
surveyed) reflects the youth of the discipline. Leadership has been drawn
from senior faculty who completed their formal training before the availability
of fellowship programs. Since geriatric fellowships became more common after
1980, it is likely that most of these senior faculty are nearing the end of
their careers, suggesting the demand for new leaders will be high in the coming
A 1993 report argued that US medical schools and residency programs
had a significant shortage of geriatric medicine faculty.26 An
Institute of Medicine national advisory panel recommended that each medical
school have 9 geriatric physician faculty to sustain their programs,12 but our survey found that 81 (70%) of the respondents
had fewer than 9. Many programs lack the financial resources to recruit additional
faculty, particularly clinician-educators. While mechanisms exist through
the National Institute on Aging, the VHA, and the private sector to support
the career development of clinician-researchers, similar career support mechanisms
currently are rare for clinician-educators. Funding from the Donald W. Reynolds
Foundation (Las Vegas, Nev) Geriatric Education Program,27 the
Hartford Foundation through the AAMC medical student curriculum initiative,28 and the Bureau of Health Professions Geriatric Academic
Career Award has been established to help support new clinician-educator faculty.
The variance in medical school spending on geriatric medicine reflects
the diversity of the size and scope of the various programs. The centers of
excellence strategy to yield maximum training of the next generation of academic
leaders appears effective for these schools, but many medical schools with
smaller geriatric medicine programs apparently do not have the resources to
recruit new faculty from these centers of excellence. New strategies to nurture
these emerging programs are required.
The DGAPs reported that the primary obstacle to the development of their
academic programs was the lack of research faculty and fellows. The recruitment
of high-quality fellows into geriatric medicine and geriatric psychiatry programs
remains a challenge for the discipline.26,29 The
source of applicants for geriatric medicine training is primary care residency
programs, and geriatric medicine shares many of the current challenges faced
by primary care, including the growth of specialization, the growing income
gap between specialists and generalists, managed care, and protechnology biases
in fee-for-service payment,30 as well as economic
dependence on the complex Medicare program. In response to the concern that
the original 2-year geriatric medicine fellowship deterred graduates of family
practice and internal medicine residency programs, in 1998, geriatric clinical
training was reduced from 2 years to 1 year for geriatric medicine board eligibility.
This change may attract applicants with a stronger interest in clinical geriatric
practice,31 but recruiting fellows and retaining
future research faculty trainees remain a significant challenge. In a recent
survey of geriatric medicine fellowship graduates, half of the survey respondents
reported being influenced by a role model or mentor. Faculty role models that
embody geriatric career pathways as clinician-educators and research scientists
will be essential to the growth of academic geriatrics.32
The DGAPs also reported concern about the impact of poor reimbursement
for clinical activity. Medicare is the primary payer for most clinical services
provided by geriatricians. In addition to relatively low clinical reimbursement
levels, Medicare's teaching and supervision guidelines further burden teaching
physicians, and these documentation rules are particularly challenging in
the home and nursing home setting.
Our study has some limitations. While our response rate of 84% is very
good, 23 medical schools are not represented in this report. Furthermore,
the accurate gathering of faculty and staff numbers and budget detail for
complex, interdepartmental programs is challenging and may add to the variability
of our data. Academic programs are always changing, and our report describes
the status of geriatric medicine programs in the spring of 2001; follow-up
surveys will provide a longitudinal view of academic geriatric medicine.
In summary, after substantial public and private investment over the
past 25 years, many medical schools now have credible academic geriatric programs
with the faculty and resources to implement clinical, educational, and research
activities. However, many other academic centers lag far behind the program
development required to ensure the adequate training of all future physicians.
The need continues for investment in training faculty as teachers and researchers
and development of medical school geriatric programs that are the size and
scope of other academic disciplines.