States have enacted or
are considering expanded health care coverage.
However, even states such as
Massachusetts
that are considered to have a large physician base suffer from an
insufficient number of providers to meet the needs of their constituents.
It is incumbent upon policymakers to increase physician supply.
The goal of physician workforce policy is getting the right number of
physicians into the right specialties and in the right locations (Grumbach,
2002). Even before one gets
into a discussion of how to accomplish this feat, one needs to estimate
future overall demand, specialty and geographic distribution, and projected
supply of physicians, a daunting task.
Maintaining current physician population ratios is inadequate because
as the population ages demand will increase, and how physicians do their
work also will change as new discoveries and technological developments
alter what physicians do (Richmond and Fein, 2005, pp. 203-204).
In the last few decades, the
U.S.
has experienced a vast change in medical organization with the rise of large
medical systems (such as HMOs and hospital owned clinics), altering the
utilization of physicians and physician extenders, while emphasizing managed
care. It is therefore perilous
to assume that the future medical organization will simply be an extension
of contemporary trends. This paper
focuses on the predicted physician shortage, examining supply and demand
factors affecting the physician workforce.
Particular attention is devoted to physician organizations’ positions
on the shortage and policy options to meet the impending problem.
Will
There Actually Be a Physician Shortage or Will We Have a Physician Glut?
For some time
physician organizations, such as the American Medical Association, have
maintained that there was an oversupply of physicians and that expansion of
medical school slots and residencies was inappropriate.
Of course, this viewpoint had a degree of self-interest in that an
increase in supply could result in reduced compensation with fewer patients
per physician and the medical organizations being able to fill their
positions without offering higher salaries or extensive signing bonuses.
The Council on Graduate Medical
Education in 1991 predicted a surplus of 80,000 physicians by 2000 and
recommended limiting the number of residencies to 10% above the number of U.S. medical graduates.
This proposed cap was aimed specifically at graduates of schools
outside of the U.S., who
had increasingly gained residency training in the U.S., and ultimately practiced in the U.S.
This added to the supply that the COGME thought was excessive.
Government, business,
and health insurers had a different take on the issue, but came to the same
conclusion that the U.S.
did not want to increase physician supply.
Their view was that health care is characterized by market failure
where workforce supply expansion was not likely to reduce reimbursements and
save money. Rather, supply
expansion would induce more demand and more physicians, especially
specialists, resulting in higher total health care costs.
Concern
about oversupply was incorporated in policy.
Medical schools, which had expanded in number in the 1960s and 1970s
as a consequence of federal policy to increase the number of physicians, now
remained relatively constant with few new medical schools.
However, even if U.S.
schools were not increasing, an increasing number of students from foreign
medical schools came to the U.S. for
training with many staying permanently.
To reduce the number of physicians trained, the federal government
included in the Balanced Budget Act of 1997 an incentive for hospitals to
reduce the number of residency slots available.
This policy was modeled on an experiment to reduce residencies that
had been implemented in New
York
State.
In the
early 2000s, opinions of major medical organizations on this issue radically
changed. The American Medical
Association and the Council on Graduate Medical Education, among other
physician groups, now recognized that a shortage, not a glut, looms large.
Each has called for a significant increase in the training of
physicians to meet the growing and aging population as a significant percent
of physicians retire. A shortage
would mean long waits for appointments and medical procedures.
Richard Cooper, former medical school dean at the Medical College of
Wisconsin and now a professor at the University of Pennsylvania,
has written extensively on this subject, issuing a clarion call for the
medical profession to recognize and meet the problem.
He and Linda Aiken of the
University
of Pennsylvania have
formed the Council on Physician and Nurse Supply, intended to assess the
problem and develop a strategic plan to meet it.
Everyone
though is not convinced that a significant shortage will in fact occur.
An editorial in the Annals of
Internal Medicine questions it (Garber and Sox, 2004).
Garber and Sox argue that Cooper relied upon a static model.
Change will alter the relationship between the demand and supply of
physicians. The editorial notes
that seniors will be healthier in the future than in the past and
technological change will modify workforce needs.
Of particular concern in Garber and Sox’s editorial is that a
significant increase in physician supply may create its own demand without
truly improving health but instead, lead to increasing costs (Graber and
Sox, 2004, p.733). An alternate
possibility is that with insurance plans emphasizing an increasing
co-payment from patients to make them price sensitive, the result may be
less demand for physician services at a time that the number of providers
has been increased. Until
firmer understanding of the future physician workforce needs is available,
Garber and Sox urge prudence—that significant expenses should not be
incurred in building new medical schools to meet a demand that may not
exist.
In the
next sections, changes in demand and supply will be examined.
Demand
Issues
Richard
Cooper and colleagues challenged the potential glut conclusion, arguing that
the U.S.
will experience a shortage of around 200,000 physicians between 2020 and
2025 (Cooper, et al, 2002). Cooper
had identified the need for expansion over a decade ago (Cooper, 1994a).
To forecast needed supply, Cooper and associates developed a model
based upon economic expansion, population growth, physician work effort, and
services provided by non-physician clinicians. They argued that increased
GDP results in increased demand for medical services.
In specific investigation of this phenomena, Cooper and associates’
study reveals a coefficient of determination of .78 between a change in
health employment five years after a change in GDP and an R2lag5=.59
for health expenditures. The
correlations though were not strong for contemporary relationships.
The lag shows that change in health employment and expenditures
follows changes in the economy (Cooper, et al., 2003).
The GDP relationship to medical expenditures has at least two
explanation elements. With a
higher GDP, individuals have more income and more health coverage and
therefore can afford more medical services.
But also with income comes a change in social expectations, driving
additional consumer demand. For
example, in the U.S. today there is a growing demand
for cosmetic surgery, a highly income-dependent service.
A second
factor in Cooper’s model is growth in the
U.S.
population, which increased from 285 million in 2000 to 300 million in
October 2006. A larger
population requires more clinicians, including some combination of
physicians and non-physicians.
Most notably, the U.S. population is aging as the first baby boomers turned
60 in 2006 with a significant increase in those 85 years and older.
Between 2000 and 2039 the number of
Americans over 65 will grow by 35 million (Salsberg, 2005).
Data reveal more doctor visits for each age cohort above 5 years old
(Salsberg, 2005). In a study of Wisconsin, the projections show that while
population is expected to grow by 8.8% from 1998 to 2015, physician visits
are expected to increase by 13.5% (Wisconsin Hospital Association and
Wisconsin Medical Society, 2004, p. 11). However,
only a small percent of primary care doctors are trained in geriatrics.
With elderly individuals typically having at least one chronic
condition, often taking multiple medications, and having conditions more
prevalent among seniors, physicians trained in geriatric medicine are
needed. However, there are only
about 7,000 geriatric trained physicians practicing in the U.S.
Adding to
demand in Cooper’s model is the availability of tests and interventions that
did not exist before. Thus even
a population of the same size and age will require more practitioners
because more tests are available to diagnose ailments and more interventions
to treat them. For example, the
demand for thoracic surgeons trained in cardiac surgery multiplied when
heart by-pass surgery first became available in the late 1960s.
Similarly, more available diagnostic and screening procedures require
more physicians. For example,
at one time colon cancer was detected by a stool test for occult blood, a
rigid proctoscope that could only visualize the last part of colon, or a
barium G.I. series. These were
typically performed by primary care physicians.
Today, the colonoscopy is considered the best screening and
diagnostic procedure, using a fiber optic scope to visualize the entire
large bowel. However, this
procedure is typically performed by gastroenterologists, who have seen
demand for their services skyrocket as more insurance companies are willing
to pay for individuals over 50 to be screened every ten years (and more
often if polyps are found).
Removal of pre-cancerous growth is among the most effective ways of
preventing cancer from developing. The
combination of more available procedures and the aging of the population
significantly increases demand (Etzioni, et al., 2003).
Supply
Issues
There
has been a growth in the number of physicians over the years, reflected in
the physician to 100,000 population ratio shown in Table 1.
Projections anticipate that with expected physician retirements, the
number of physicians per 100,000 populations will level off around 2015 and
then decline (Salsberg, 2005).
Currently one-third of practicing physicians are age 55 and over.
In comparison to other developed
nations, the U.S. has fewer
physicians per 1,000 people than many developed nations.
For example, the 2004 data show that the U.S. had 2.3 physicians per 1,000 population in
comparison to France and Germany with 3.4 and Italy with 4.2.
The U.S had the same ratio of physicians to population as
Britain and slightly more than
Canada
that had a ratio of 2.1 physicians per 1,000 people.
Many nations have more physicians per 1,000 population than the
U.S.
as shown in Table 2 (OECD, 2006).
Visits to physicians also are more frequent in other nations than in
the U.S.
Thus, a significant explanation for the greater amount that the
U.S.
spends per person and the greater percent of the GDP spent on medical care
in the U.S. in contrast
to other nations has more to do with higher prices than medical care
received (Anderson, et al., 2003).
Table
1
Source: American
Medical Association (2006)
Table 2
Source:
OECD (2006)
The U.S.’s increased
ratio of physicians to population (Table 1) masks at least two elements: the
decline in workforce effort and the increased fractionalization of the
physician workforce. The
declining effort is included in Cooper’s model.
Older physicians tend to work fewer hours as they approach
retirement. Younger physicians
will also work fewer hours than their older colleagues since they are more
concerned about maintaining personal and family time than their older
colleagues were when they were young (Wisconsin Hospital Association and the
Wisconsin Medical Society, 2004, p. 13; Cooper, et al., 2002).
Many of these physicians work for
large health systems that provide more regular hours than physicians expect
when they are in solo practice or a practice with a few partners.
Some have hypothesized that the recently imposed restrictions on the
number of hours that residents are allowed to work (80 hours a week) will
carry over to a more limited work week for physicians than was the past
practice. Additionally, female
physicians, the data show, are more likely to work fewer hours, many even
part-time, after they have children.
With the growth in the percent of physicians who are women, work
effort is reduced and more physicians are needed (Cooper, et al. 2002).
The
medical workforce also has become increasingly fractionionalized as more
specialties and subspecialties emerge.
For example, a growing subspecialty in cardiology is
electrophysiology, where physicians focus on the electrical system of the
heart. Through new ablation
techniques, these electrophysiologists are able to eliminate extra calcium
channels in the heart, thus correcting an irregular heart beat.
Another recently developed subspecialty is the hospitalist, who only
see patients admitted to the hospital, thus relieving the family
practitioner or office based internist from the task of managing their
patients’ care once admitted to the hospital.
Demand for hospitalists has skyrocketed.
A similar subspecialty of internal medicine is the intensivists, who
typically provide care in the Intensive Care Unit (ICU).
These physicians are trained and experienced in the care of
critically ill patients. Even
long standing specialties are subdividing.
Orthopedists can be found focusing on the spine, hand, or adult joint
replacements, to mention a few.
With patients no longer receiving most of their care from their primary care
physician or even the traditional specialist, the ratio of physicians to
population needs to increase considerably to account for these trends.
Although
the 125 medical schools in the
U.S.
have recently increased their enrollment, it has been insufficient to meet
projected demand. Enrollment of
first year students in 2006 was about 17,200, which marked a 2.2% increase
over the previous year.
Osteopathic schools, whose graduates are now accepted on par with M.D.s,
have also increased their enrollment, although their contribution to the
supply remains small—about 46,000 D.O.s in active practice. The Council on
Graduate Medical Education (COGME) now recommends an increase in enrollment
in medical schools of 30% from their 2002 level over the next 15 years
(Council on Graduate Medical Education, 2006).
Even if this is accomplished, it will not fill the deficit of 200,000
physicians predicted by Cooper.
The COGME, which argued that the U.S.
has a physician glut, now estimates a 150,000 shortage.
With the number of new physicians graduating from
U.S. medical schools now about equaling the number
retiring, the U.S.
will rely even more heavily on graduates of foreign medical schools, known
as International Medical Graduates (IMGs), who now constitute 23% of
practicing physicians in the
U.S.
Counter-trends
A
fourth factor in Cooper’s model, a counter-trend to increased supply needs,
is called the substitution effect,
the use of non-physician clinicians, including nurse practitioners,
physician assistants, and mid-wives.
Once limited in the services they could provide, most state licensing
laws now allow them to provide services that previously could only be
provided by physicians.
Although a significant number of these non-physician clinicians practice
primary care, many are now providing care in specialty areas (Weiner, 2004,
p. 54). The conventional wisdom
is that the workforce shortage is in primary care where non-physician
practitioners can have the most significant impact in filling the gap by
providing routine care.
Specialty care areas, many argue, either have only a small shortage or even
a surplus, except for a few areas.
More will be said when distribution issues are examined.
Another
counter-trend relates to technology and calls for technological developments
to simplify procedures.
Procedures such as colonoscopy, echocardiography, and others have increased
demand for specialized physicians, but technological developments could
simplify them, allowing primary care doctors to provide these services as a
routine part of care, thus reducing demand for specialists (Crossen, 2004,
p. 63).
Another
counter-trend which could reduce the number of physicians needed is one in
which specialists are able to provide services that formerly only
specialists in a different area provided.
Thus, overlapping roles could reduce the total number of physicians
needed. An example of this
overlap is in angioplasty. Once
the province of the interventionist cardiologists, this treatment to open
clogged coronary arteries is now done by some interventionist radiologists.
Similarly, interventionist radiologists are offering ablation
procedures to women with uterine fibroids (uterine cysts), reducing the need
for gynecologists to do hysterectomies.
Lastly,
outsourcing could be a way to reduce the need for
U.S.
physicians as it has for labor in other areas of the economy.
Among the first use of outsourcing has been in the field of
radiology. The introduction of
digital imaging has allowed hospitals to send their images to other
countries such as India
or Australia
to be read by their radiologists.
This has been particularly useful at night, saving cost that would
occur if local radiologists were on call to read these studies.
Developments in telemedicine, intended to provide specialty
consultations, particularly to rural communities, could open up the
possibility of outsourced diagnoses to physicians in other countries. This
would reduce needed supply of these specialists in the U.S.
Both U.S. shortages and cost
differentials could drive this trend.
One additional outsourcing movement is known as “medical tourism.”
Currently representing very small numbers but expanding, medical
tourism is when patients go to other countries where medical costs are less
for elective procedures. India and Thailand
are two countries recruiting patients from the U.S.
Distribution
Issues: Geographic and
Specialty
Beyond
aggregate supply and demand are issues of geographic and primary/specialty
distribution. These have been
of concern for many years, but predicted overall shortages heighten the need
to consider them in policies adopted.
Shortages
predominate in many rural and inner city communities (Cooper, 1994, p. 686)
Physicians, because of life style preferences, employment of spouses, and
income, prefer to locate outside of inner cities and rural area.
Thus, despite several federal policies, including the development of
the National Health Service Corps and
Community
Health
Centers, and state efforts to secure physicians
for these communities, Health Professional Shortage Areas (HPSAs) have
increased. More than 20% of the U.S. population
lives in a community designed as short of primary care physicians.
The consequences are that patients in these areas receive less
medical care, have delayed diagnoses and treatments, and have greater costs
associated with medical care because of travel cost and the need to take
more time off from work (Miller, 1989).
One goal
of increasing the number of physicians in the
U.S.
was that the increased supply would result in a diffusion of physicians from
areas with adequate supply to underserved communities.
However, the diffusion notion was stymied by physician locational
preference. Although some
occurred, especially with IMGs, the flow was small.
Illinois,
despite being among the states graduating the largest number of physicians,
still remains among the worst in terms of medical shortage areas. The
federal Health Service Corps has had only limited success because of budget
cuts. The geographic
distributional problem will be made worse as a number of physicians that
have been recruited for these areas in the 1970s retire.
For example, the Wisconsin Office of Rural Health Physician
Placement Program reports 320 openings, including 166 specialists and 154
primary care physician slots, in the state in February 2006.
A second
distributional issue is the optimal division between primary care physicians
and specialists. For many
years, the argument has been that too many medical graduates join the ranks
of specialists, leaving an inadequate number of primary care doctors.
With the income of specialists being greater than for those in
primary care practices and the heightened prestige for specialists conveyed
in medical school by professors, the specialist career choice is
understandable. Additionally, career choice is also influenced by lifestyle
and free time with specialists having more controlled hours (Richard and
Fein, 2005, p. 116). The consequence
often cited for an oversupply of specialists is that health care costs
increase because more procedures, surgery, and tests are performed and the
reimbursement rate for specialists is higher.
The often cited goal is to have 50% of physicians practicing primary
care.
Canada now comes close to this goal, but the
U.S.
remains far behind with only about one third of medical school graduates
becoming primary care physicians.
A larger number of International Medical Graduates practicing in the U.S.
do go into primary care—44% (AMA IMG Governing Council, 2006, p. 5).
The actual practice of primary care
is more than the often cited figures suggest because a number of specialists
also practice primary care (Cooper, 1994, p. 681).
For example, some pulmonologists
will see patients not only for lung related diseases but also for general
internal medicine.
Barbara
Starfield and colleagues empirically investigated the relationship between
mortality and the number of primary care doctors and specialists, using
3,075 U.S. counties from 1996 to 2000 (Starfield, et al., 2005).
They found that counties with more primary care physicians (defined
as office-based practices of those in general practice, family medicine,
general internal medicine, or pediatrics) had a lower mortality rate, but
there was no relationship between the number of specialists and mortality.
They controlled for background factors associated with higher
mortality such as percent of elderly, African-American, or incomes below
100% of the federal poverty line.
Overall, a 20% increase in primary care physicians per 100,000 was
related to a 34.6% decline in mortality per 100,000 at the state level.
However, a greater number of specialists did not have a statistically
significant impact. Thus, the
aggregate supply of physicians is less important than their distribution (Richmond
and Fein, 2005, p.204; Phillips, et al., 2005).
Although everyone acknowledges that there is a shortage of some
specialties and subspecialties (e.g. geriatricians, psychiatrists, and
pediatric subspecialties) overall based upon their ratio to the population
and their relationship to improving health (such as Starfield’s findings),
federal policy has emphasized primary care (Phillips, et al., 2005, p. 113).
The need
for more primary care doctors also has implications for geographical
distribution. Even though the U.S. has one primary care doctor for
every 1,321 Americans, for non-Metropolitan Statistical Areas the ratio is 1
to 1,821 and worse for rural communities (Phillips, et al., 2005, p. 113).
However, not everyone
agrees with the conventional wisdom regarding the primary care shortage.
Salsberg argues that the explosion of knowledge and complex treatment
options require specialists. Even Kaiser Permanente, an HMO which stresses
efficiency, has reduced its ratio of specialists per 100,000 enrollees by
44% between 1983 and 2001 (Salsberg, 2005; Weiner, 2004).
One of the most well-known articles
on this topic is Richard Coopers’ “There’s a Shortage of Specialists: Is
Anyone Listening?” (2002). Cooper
argues that the greatest need is for specialists.
Although he acknowledges that there is an oversupply in some
specialties such as ophthalmology, Cooper maintains that there is an
abundance of primary care physicians, especially with the use of nurse
practitioners and physicians assistants to see a portion of the patient load
(Cooper, 2002, p. 764; Cooper, 1994, p. 683).
In 2003, Cooper and associates surveyed medical school deans and
state medical society executives regarding their perceptions of physician
shortages. Overall, the results
showed that their experiences indicated that shortages of physicians now
exist, varying by practice area.
Although some respondents noted significant shortages in some primary
care areas, e.g., general internal medicine, in others such as general
pediatrics, family practices, and obstetrics/gynecology shortages were
either non-existent or small. However,
in several specialties, the shortage was seen as acute.
Anesthesiology and radiology led the
list (Cooper, et al., 2003). Cardiologists
have written of shortages in their field in the
American Journal of Cardiology,
anticipating a worsening as the population ages.
A proposal by cardiologists to admit new students into cardiology
fellowships after two years of general internal medicine residency is
controversial. Although it
would increase the number of cardiologists, it would also reduce the number
of general internists, given the wide disparity in income between the two
practices.
Public
Policies toward the Physician Workforce
The
1910 Flexner Report resulted in a decline in the number of U.S. physicians.
Abraham Flexner, a Carnegie Foundation for the Advancement of
Teaching researcher, undertook an evaluation of the 155 medical schools in
the U.S.
and Canada,
many of which were proprietary schools.
In his report, he classified the existing schools based upon their
quality, highlighting those schools that were clearly deficient in
curriculum, facilities, scientific orientation, and faculty.
Although reform of medical education was already underway at the time
Flexner wrote, his report had a major impact on academic medicine (Cooke, et
al., 2006).
The
consequence was that many schools closed, reducing the number of graduates.
Just before the Report in 1906, there were 162 medical schools, which
had declined to 131 at the time the Report was issued.
The drop accelerated, falling to only 76 schools by 1926, a number
that remained fairly constant for the next 20 years.
The number of graduates dropped from its peak of 5,747 in 1904 to
4,440 by 1910 to 3,047 in 1920, resulting in a significant increase in the
ratio of physicians to the population (Richmond and Fein, 2006, p. 12).
Associated with the decline in number was a decline in diversity of
physicians graduating as many schools for blacks and women closed (Starr,
1984).
The 1960s
and 1970s saw both federal and state action to increase the physician
supply. In 1963 the Health
Education Facilities Act was enacted to provide federal assistance to build
new medical schools and expand existing ones.
States also provided funds to expand the number of physicians being
graduated in their state with five new schools opening by 1963 with 10 more
planned (Cooper, 2003). To
encourage the enlargement of medical classes, the national government under
the Health Professions Education Assistance Act in 1963 began requiring
medical schools to increase their enrollment if they were to receive
construction money. By 1971 the
national government introduced the capitation grant, basing funding on the
number of students. Discouraged
that the result was that more specialists were being trained, the federal
government linked the capitation grants in 1976 to the graduation of
physicians intended to go into primary care.
The goal was not only to increase the number of physicians in the
U.S.
but also to attack geographic distribution problems.
The anticipation was that if more physicians graduated, the result
would be diffusion of physicians to medical shortage areas.
However, this did not significantly occur.
In trying to attack this problem, the National Health Service Corps
(NHSC) was enacted, providing loan forgiveness to physicians who were
willing to serve several years in medical shortage areas.
The NHSC also served as an alternative to military service, an
especially valuable option to graduating physicians during the draft and Vietnam.
States
worked to expand their physician supply. From 1960 to 1980 41 new medical
schools and 8 new osteopathic schools were opened.
Consequently, the number of
graduates jumped from 7,500 in 1960 to 16,200 in 1980 (Richmond and Fein,
2006, p. 106). Beyond building new
medical schools, many states stipulated the percent of students that must
come from that state. The
assumption was that residents of the state were most likely to remain in
that state upon graduation (Richmond and Fein, 2006, p. 105).
States created offices whose
function was to help areas with physician shortages obtain physicians.
Grants and loans to students who would locate in these areas were key
tools along with assisting physicians with initial practice expenses.
Demand for
services had increased as the population increased, scientific developments
led to more effective services being available, and the passage of Medicare
and Medicaid in 1965 insured individuals who did not have funds to pay for
medical services. Even with the
increased demand, medical associations were particularly concerned that the
expansion in physician supply created a surplus that would negatively impact
practice incomes. The American
Medical Association in particular resisted efforts at any more supply
expansion. It was not until
after 2000 that the A.M.A. recognized that the
U.S.
did not have a glut of physicians, but rather a deficit.
With an increase in retirements, the A.M.A. recognized that the
projected shortage would get worse.
The AMA
and other physician interest groups are especially concerned with the
financial interests of their members.
As such, they most notably have pushed for higher reimbursements,
less government involvement, and restricted number of non-physician
professional where competition could reduce incomes (Feldstein, 1991, p.
215-222; Feldstein, 1977) In
this vain, these organizations have opposed opening of new medical schools,
the liberalization of immigration laws and expansion of residencies for
International Medical Graduates in the U.S., and the enhanced roles for
non-physicians such as nurse practitioners, psychologists, or optometrists.
Regarding the latter, for example, there is considerable opposition
to having optometrists authorized to dilate pupils or treat conditions such
as conjunctivitis. For
psychologists, physician organizations have testified against allowing them
to prescribe medication for their patients, typically requiring patients to
go to psychiatrists, who are physicians, to get psychiatric drugs with
periodic appointments to check on the efficacy and side effects of the
drugs. These patients often
also see psychologists for counseling therapy.
Because there is an insufficient number of psychiatrists, especially
child psychiatrists, patients may receive their prescription from their
primary care provider, who may know less about psychiatric medication than
psychologists, who are prohibited from prescribing in most states.
The
position of physician interest groups on the expansion in the supply of
physicians typically has been accepted by government decision-making because
they both subscribe to the argument that there was to be a glut of
physicians and fear that if demand expanded as supply did, health care costs
would increase. Thus, action of
Congress to limit federally subsidized residencies in the Balanced Budget
Act of 1997 was not only intended to limit federal Medicare costs, but also
to restrain health care costs in general by freezing the number of
physicians, especially specialists, produced
(Garber and Sox, 2004). Consumer
oriented groups, such as the AARP, also did not support the expansion in the
number of physicians because imperfections in the medical labor market meant
that supply expansion would not reduce costs for their members.
With
national and state projections of an increasing physician shortage,
physician medical organizations have changed their positions and recognized
a coming deficit, although they disagree as to its magnitude.
Although Cooper has estimated the
shortfall by 2025 to be about 200,000, physician groups have called for a
much smaller increase. For
example, the Association of American Medical Colleges (AMC) calls for an
increase of 30% and the Council of Graduate Medical Education (COGME) for a
15% increase in the number of graduates. Fifteen
percent represents only 2,800 more graduates per year. The COGME projects a
shortfall, but about 85,000 nationally, less than half of what Cooper
forecasted (Wilson, 2005, pp. 469-470; Association of American Medical
Colleges, 2006a).
However,
physician interest group’s solution rests on market adjustments with little
government involvement beyond the expansion in the number of medical schools
and medical residencies (See Council on Medical Education, Rep. 8-A-05,
2005). Unlike Canada, which now determines the
ratio of residencies for primary care versus specialties, the Association of
Medical Colleges strongly supports free choice of specialty for students and
for residency positions for teaching hospitals (Association of Medical
Colleges, 2006a, p. 7).
Policy
Options
Of
the options considered, one of the most obvious is to expand the number of
medical schools and the enrollment in the existing medical schools.
However, analysts have suggested problems in doing this.
Only one new medical school has opened in recent years, although a
few more are planned.
The
traditional objection that has been heard for decades is that if the
enrollment was expanded, students of lesser quality would have to be
admitted. This objection was
voiced in the 1960s and again today.
Douglas Wood, for example, writes, “One of the trade-offs would be to
accept applicants who are less qualified than today’s medical students.
This, however, would most likely decrease the quality of medical
care, with consequences that would be unacceptable to the American public”
(Wood, 2003, p. 98).
Wood
further argues that changes in medical education lead to smaller class sizes
rather than being able to absorb more students.
Rather than rely on lectures, instructional methods that concentrate
on small groups, especially oriented towards problem solving, have been seen
as more effective. This
approach is more faculty intensive.
As medical school curriculum expands, such as inclusion of new
findings from the Human Genome Project that will impact practice, it will be
more difficult to simultaneously expand curriculum and enrollment (Wood,
2003, pp. 98-99). With medical
faculty being asked to do more research to bring in grant money and to see
more patients, medical education takes third in priorities.
Most notably, medical
schools are expensive, which is one reason that states have been reluctant
to charter new ones in recent years.
Today’s average first year enrollment for a medical school is about
150 students. Thus, if the U.S. is to meet
the forecasted shortage, there would have to be an additional 2,500
students, or the equivalent of 16 new medical schools.
Even with increased capacity of some of the existing schools, it is
doubtful that states will find the resources to open many new schools
(Council on Medical Education Report, 8-A-05, 2005, p. 3).
How much of this
opposition to expansion of medical school is economic protectionism?
The American Medical Association, which has always played a
substantial role in the accreditation of the medical schools beginning in
the early 20th century, has a clear guild mentality, concerned
that the expansion of medical school graduates will increase competition and
reduce physicians’ income (Richmond and Fein, 2005, pp. 14-15).
The projected shortage
may encourage states to either expand enrollment at the existing medical
schools and/or build new medical schools.
Florida is the only state
in recent years to open a new medical school.
Beginning in 2008
Florida
International
University
in Miami is
scheduled to open a new school with an initial beginning class of 36
students with the goal of having 480 students in their school, 120 per
class. The University of Central
Florida has also received approval for a new
school with an intended first year class of 180.
Oregon is another state seeking to
open a new facility with a branch campus of the University of Oregon
while Texas
is expanding enrollment at its existing schools.
Currently, there are 15 new allopathic schools under consideration
(11 new schools and 4 regional campuses) plus 10 new osteopathic schools (9
new schools and one regional campus).
The AAMC projects that five new allopathic medical schools will be
open by 2015, with estimated additional enrollment from the new allopathic
schools of between 455 and 530 students (American Association of Medical
Colleges, 2006b, pp. 20 and 24).
While medical school
enrollment had been constant for many years, in 2005-2006 there had been a
2.1% increase in enrollment, 352 more students, for a total of 17,004 first
year students. A 2003 survey by
Cooper and associates found that existing medical schools said they could
expand 7.6% (Cooper, et al., 2003). In
addition to allopathic medical schools giving the M.D. degree, Osteopathic
schools, awarding the D.O. degree, have also increased enrollment with 3,889
first year students. Today,
D.O.’s practices do not differ from M.D.’s except that more osteopaths go
into primary care. Osteopathy
graduates still represent only a small percent of practitioners.
Beyond further state
and federal funding of medical schools themselves, subsidization and loan
forgiveness for students, especially targeted at those who practice in
underserved communities (and primary care if that is a goal), is seen as a
key policy approach. The cost
of medical education is high, discouraging promising undergraduates from
applying and directing specialization to fields with higher average incomes.
Although hiring bonuses and assistance by medical organizations in
paying back loans exist, college graduates considering a possible medical
career do not fully take this into account.
Although medical schools through their financial aid offices do
inform students about options and possibilities, this often occurs after
students make application and have decided on a medical career.
An upfront subsidy would increase applications from good students.
This is especially needed as the number of college graduates levels
off in upcoming decades.
Other factors
discourage students seeking a medical education.
The issue of malpractice, both the cost of insurance and perceived
threats of lawsuits, has been continually in the news.
Students will ask whether they want to go into a field with this
risk. The policy option here
would be appropriate tort reform that would reduce the cost of insurance as
well as reduce non-meritorious suits.
Further work in reducing the risk of malpractice, such as an
electronic record and computerized prescriptions, would help as well.
Along with malpractice, more physicians are being employed by large
medical systems, replacing the solo practice system.
These systems, although providing important benefits such as shared
call, also frequently impose management techniques, such as length of time
with patients that physicians believe interfere with their professional
judgment. Surveys have shown
that a significant percent of physicians say that they would choose another
career if they had to do it all over again.
This negativity filters down to potential doctors, discoursing a
choice of a medical career.
Although difficult to implement, medical organizations should adopt
recommendations encouraging these systems to treat physicians more as
professionals than as employees.
Encouraging college
students to major in science also could increase applications to medical
schools (Even if the student is not a science major, medical school
admission requires the student to have completed a list of specified science
courses). Science teaching in America has been
criticized as discouraging potential majors with science majors increasingly
composed of international students in many schools.
By reforming the science curriculum and teaching approach, a larger
student base would be available to recruit to medical schools.
With the number of U.S. college graduates going to medical schools
in other countries and ultimately practicing in the U.S., additional applicants would be
available for expanded medical school slots.
These students would have a better chance of passing the
U.S. licensing exam with the better background provided
by U.S. schools.
Under the 1997 Balance
Budget Act, the number of residencies remains capped at 98,000.
Under the cap, hospitals that choose to increase their residencies
unilaterally will not receive additional Medicare Graduate Medicare
Education (GME) dollars. But in
an attempt to produce more physicians, the Centers for Medicare and Medicaid
are working to reallocate unfilled residencies to hospitals that are more
likely to have students to fill them. The cap and reallocation represents a
health workforce planning model that the federal government has typically
avoided because of a market ideology that questions the ability to
accurately predict need and pressure from interest groups.
However, federal subsidies for residencies and attempts to reduce
federal budget deficits have lead to the residency limitation.
It should be noted that when the residency cap was approved, the
major medical interest groups were of the view that the
U.S.
had a surplus. With the 21st
century change in views, the 1997 cap may need to be revisited (Grumbach,
2002).
Several proposals
challenge the tethering of GME to Medicare as it is neither good for
Medicare financing nor for planning for physician needs.
Although proposals differ in details, they often tie planning for
specific physician needs to direct federal subsidy of training positions.
This public funding approach is a more rational way to proceed.
Given the difficulty of predicting
physician needs in each specialty, some proposals simply subsidize total
production, leaving the free market to affect deployment in different areas.
The challenge is whether the free market will meet population needs.
In the 1960s when the federal subsidies began to increase the
physician supply, more physicians were trained, but a disproportion became
specialists and a disappointed percent established practices in medically
shortage areas (Grumbach, 2002). A
1980 report by the Graduate Medication National Advisory Committee (GMENAC)
warned both of a pending oversupply of physicians and a maldistribution of
physicians; their recommendation was that policy should be more planning
oriented, directing physicians in areas of need.
With the Reagan Administration coming to power, ideas promoting
further regulation were off the agenda (Grumback, 2002).
With the difficulty of
producing more physicians, one option is to employ them more efficiently (Salsberg,
2005, p. 117-118). This has often
been the approach in managed care organization.
Key to the lower physician patient
ratio in organizations such as Kaiser Permanente is the use of teams,
including nurse practitioners and physicians assistants to treat patients (Crosson,
2004, p. 61; Sender, 2004).
Some medical schools are encouraging physicians to adopt a collaborative
care model in treating patients.
Other efficiency measures can also improve physician services and
allow them to see more patients.
Often cited is evidence-based medicine.
This refers to restricting tests and treatments to those that
empirical studies have shown to be effective.
By eliminating non-effective actions, physicians can be both more
effective and efficient.
Similarly, the electronic record, often advocated to reduce errors, is also
an efficiency measure. Tests,
including imaging studies, are readily available to the treating physicians,
eliminating wasted time in hunting for records and transporting films.
Physicians can also be
better utilized with the development of telemedicine.
Each community may not need a resident specialist, but primary care
physicians can confer and the specialist can actually examine a patient from
a distance. Although this will
not end the physician shortage, telemedicine will reduce its negative
impacts.
Conclusion
The U.S. is facing a
significant deficit of physicians.
Given the lead time needed to train physicians and time needed to
open new medical schools and expand the enrollment of existing ones, action
by the federal and state governments must occur now.
REFERENCES
AMA
IMG Governing Council. 2006. “International Medical Graduates in the U.S. Physician Workforce: A
Discussion Paper.
http://www.ama-assn.org/ama1/pub/upload/mm/18/workforce2006.pdf.
Accessed: 11/27/2006
American
Medical Association. 2006. Physician Characteristics and Distribution in the U.S.
Chicago: American Medical Association.
Anderson,
Gerard, Uwe E. Reinhardt, Peter S. Hussey, and Varduhl Petrosyan. 2003. “
It’s the Prices Stupid: Why The United States Is So Different from Other
Countries.” Health
Affairs. 22: 89-105.
May/June. No. 3.
Association
of American Medical
College. 2006a. “AMC
Statement on the Physician Workforce—June 2006.”
http://www.aamc.org/workforce/workforceposition.pdf. accessed:
11/20/2006
Association
of American Medical Colleges. 2006 (b). “Medical School Expansion Plans.”
http://www.aamc/worforce.
Accessed: 11/20/2006.
Baerlocher,
Mark. 2006. “The Importance of Foreign Trained Physicians to
Canada.”
Clinical and Investigative
Medicine. Vol. 29. June.
Boulet, John R., John J.
Norcini, Gerald P. Whelen, James A. Hallock, and Stephen S. Seeling. 2006.
“The International
Medical Graduate Pipeline: Recent Trends In Certification And Residency
Training.” Health Affairs. 25 (2)
: 469-477. March.
Ceaser,
Mike. 2005. “Sun, Sand, and an
M.D.: New For-Profit Medical Schools in the Caribbean Draw Many Americans, But are Graduates Qualified
to Practice?”
The Chronicle of Higher Education.
October 28.
http://chronicle.com/weely/v52/i10/10a05501.htm.
Accessed: 12/07/2006
Cooke,
Molly, et. al. 2006. “American Medical Education 100 Years after the Flexner
Report.
New England
Journal of Medicine. 355:1339-1343. September 28.
Cooper, Richard.
1994c. “Regulation Won’t Solve Our Workforce Problems.”
The Internist. March. Pp. 10-13
Cooper, Richard. 2002.
“There’s a Shortage of Specialists. Is Anyone Listening?”
Academic Medicine. 77: 761-766.
August.
Cooper, Richard.
1994a. “Seeking a Balanced Workforce for the 21st Century.”
Journal of the American Medical
Association. 272 (9): 680-687.
Cooper, Richard.
2004b. “Scarce Physicians Encounter Scarce Foundations: A Call for Action.”
Health Affairs.
November/December.
23:243-249.
Cooper. Richard,
Thomas E. Getzen, Heather J. McKee, and Prakash Lard. 2002. “Economic and
Demographic Trends Signal and Impending Physician Shortage.”
Health Affairs. 21:140-154.
Jan./Feb.
Cooper, Richard,
Thomas Getzen, and Prakash Lard. 2003. “Economic Expansion Is a Major
Determinant of Physician Supply and Utilization.”
28(2):675-696. April.
Cooper, Richard.
Sandra J. Stoflet, and Steven A. Wartman. 2003; “Perceptions of
Medical
School Deans and State
Medical Society Executives About Physician Supply.”
Journal of the American
Medical Association. 290 (22): 2992-2995.
December 10.
Council on Graduate
Medical Education. 1998. “Eleventh Report: International Medical Graduates,
the Physician Workforce, and GME Payment Reform.” Public Health Service. U.S. Department of Health and Human
Services. March.
Council on Graduate
Medical Education. 2004. “Meeting Minutes.”
http://www.cogme.gov/minutes 07-04.htm. Accessed 1/30/2006.
Crossen, Francis.
2004. “The Changing Shape Of The Physician Workforce In Prepaid Group
Practice.” Health Affairs. Web
Exclusive. W4-60-63.
February 4.
Council on Medical
Education. 2005. CME Report 8-A-05.
Eckhert,
N. Lynn. 2002. “The Global Pipeline: Too Narrow, Too Wide, or
Just Right.”
Medical Education. 36: 606-613.
No. 7.
Education Commission
on Foreign Medical Graduates (ECFMG). 2006. “International Medical
Directory.” http://imed.ecfmg.org/
accessed: 10/30/2006
Etzioni, David, Jerome
Liu, Melinda Maggard, and Clifford Ko. 2003. “The Aging Population and Its
Impact on the Surgery Workforce.”
Annals of Surgery. 238
(2): 170-177. August.
Feldstein, Paul. 1977.
Health Associations and the
Demand for Legislation: The Political Economy of Health.
Mass: Ballinger Publishing Co.
Feldstein, Paul. 1991.
“Health Associations and the Legislative Process.” Pp. 209-246 in Theodor J.
Litman and Leonard S. Robbins. Health
Politics and Policy. New York:
Delmar Publishing. 2nd
edition.
Garber, Alan and
Harold Sox. 2004. “Editorial: The U.S. Physician
Workforce: Serious Questions Raised, Answers Needed.”
Annals of Internal Medicine.
141 (9): 732-734.
Grumbach, Kevin. 2002.
“Fighting Hand to Hand Over Physician Workforce Policy: The Invisible Hand
of the Market Meets the Heavy Hand of Government Planning.”
Health Affairs.
September/October.
Hagolan, Amy, Matthew
J. Thompson, Emily Kaltenbach, and L. Gary Hart. 2003. “Health Departments’
Use of International Medical Graduates in Physician Shortage Areas.”
Health Affairs.
22:242-249. Sept./Oct.
Hallock, James A.,
Stephen S. Seeling, and John J. Norcini. 2003. “The International Medical
Graduate Pipeline.” Health Affairs.
22:94-96. July/August.
Miller, Edward J.
1989. “Medically Underserved
Small Communities: The Approaches of Health Policy.”
Pp. 51-61 in Nancy S. Lind and Ann Elder.
The Small
City
and Regional Community:
Proceedings of the 1988 Conference. Stevens Point, WI:
University
of Wisconsin Foundation
Press. Vol. 8.
Mullan, Fitzhugh.
2002. “Some Thoughts on the White-Follows Green Law.” 2002.
Health Affairs. January/February.
158-159.
Mullan, Fitzhugh.
2005. “The Metrics of the Physician Brain Drain.”
New England
Journal of Medicine. 353:1810-1818.
October 27.
OECD. 2006. OECD
Health Data.
http://www.oecd.org/dataoecd/59/47/35529872.xls
June. Accessed.
9/25/2006.
Olson, Sandra. 2005
“The Physician Workforce: Recommendations for Policy Implementation
(Resolution 313, A-04, Resolution 319, A-04, Resolve 1). Report to the
Council on Medical Education.
Phillps, Robert L.,
Marty S. Dodoo, and Larry A. Green. 2005. “Adding More Specialists Is Not
Likely To Improve Population Health: Is Anyone Listening.”
Health Affairs. Web Excusive.
March 15. W5-111-114.
Ramamurthy, Rajam.
2005. “International Medical Graduates in the Physician Workforce Debate.”
PowerPoint.
http://www.texmed.org/uploadedFiles/About_TMA/IMG_Section/IMGintheworkforce.final.ppt#1
Accessed: 11/27/2006
Richmond,
Julius and Rashi Fein. 2005.
The Health Care Mess.
Cambridge:
Harvard University Press.
Salsberg, Edward and
Gaetano Forte. 2004. “Benefits And Pitfalls In Applying the Experience of
Prepaid Group Practices To The U.S.
Physician Supply.” Health Affairs.
Web Exclusive. W4-73-75.
Salsberg, Edward.
2005. “The Physician Workforce
Research Agenda: Expanding the Science, Enhancing the Impact.”
Paper delivered at the 2005 AAMC Physician Workforce Research
Conference.
Washington,
D.C. May 5.
Salsberg, Edward.
2005b. “The Need for Real Evidence in Physician Workforce Decision-Making.”
Health Affairs. Web
Exlusive. March 15.
Salsberg, Edward.
2006. “Testimony at the Legislative Hearing on H.R.4997: Physicians for
Underserved Areas Act.” U.S. House of Representatives.
Committee on the Judiciary.
Subcommittee on Immigration, Border Security, and Claims.
May 18.
Sender, Johathan.
2004. “Addressing a Physician
Shortage: UW Medical School’s Carl Getto, MD Analyzes a Looming Problem.”
http:
www.med.wisc.edu/news/item.php?id=950.
August 23. Accessed: 11/4/2006.
Simoens, Steven and
Jeremy Hurst. 2006. “The Supply
of Physician Services in OECD Countries.”
OECD Health Working Papers.
January.
Starr,
Paul. 1984. The
Social Transformation of American Medicine.
New York:
Basic Books.
Starfield, Barbara,
Leiyu Shi, Atul Grover, and James Macinko. 2005.
The Effects Of Specialist Supply On
Populations' Health: Assessing The Evidence. Health Affairs. Web Exclusive.
March 15.
The Internist.
March.
Pp. 10-13.
Weiner,
Jonathan. 2004. “Prepaid Group Practice Staffing And U.S. Physician Supply:
Lessons For Workforce Policy.”
Health Affairs.
W4-43-55. Web Exclusive. February 4.
Wilson,
Jennifer Fisher. 2005. “U.S. Nees More Physicians Soon, but How Many More is
Debatable.”
Annals of Internal Medicine.
143(6): 469-472. Sept.
20.
Wisconsin
Hospital
Association and Wisconsin Medical Society.
2004. “Who Will Care for Our Patients:
Wisconsin
Takes Action to Fight a Growing Physician Shortage.”
https://www.wha.org/physicianshortage3-04.pdf. Accessed 12/22/2006.
Wood, Douglas. 2003. “The Physician Workforce:
A Medical
School Dilemma.”
Health Affairs. 22:97-99.
July/August.
