|
Salsberg Web Exclusive
P E R S P E C T I V E :
P G P S T A F F I N G
W E B E X C L U S I V E
4 February 2004
Benefits And Pitfalls In Applying
The Experience Of Prepaid Group Practices
To The U.S. Physician Supply
Comparisons provide valuable insights
but should not be
the sole basis for physician workforce planning.
By Edward Salsberg and Gaetano
Forte
ABSTRACT:
The paper by Jonathan Weiner includes important improvements in the methodology
used to compare the physician workforce in prepaid group practices (PGPs) with
the U.S. physician workforce. It also provides valuable insights for policymakers
and researchers. Despite the improvements, concerns remain regarding the comparability
of the populations served and physician activities in PGPs and the country as
a whole. While PGPs appear to offer valuable lessons on how to use physicians
effectively and efficiently, it is inappropriate to use the PGP physician rates
to determine the number of physicians needed in the United States.
Jonathan Weiner is to be commended for his latest paper, which provides important
new data as well as valuable insights for policymakers. In it he presents the
results of a new analysis of physician staffing and use in several large prepaid
group practices (PGPs), primarily under the Kaiser Permanente Health Plan, and
compares the physician-to-population ratios in the PGPs with that of the United
States as a whole. He compares the results of the most recent analysis with
the results of a similar analysis of data from the early 1990s.1
As did his earlier work, Weiner’s new analysis finds that PGPs use 22–37
percent fewer physicians than the U.S. average.
Despite Weiner’s good efforts, there are reasons to believe that physician
counts from PGPs are not directly comparable to the counts for the United States.
Nevertheless, PGPs likely make better use of physician resources than does the
nation in general. The challenge is to better understand how PGPs manage with
fewer physician resources and then to apply the lessons learned to the delivery
and financing systems that dominate in the United States. However, using the
workforce needs of a model system of care used by a small segment of the U.S.
population to guide physician production for the whole country is inappropriate
and could lead to major shortages, access problems, and public dissatisfaction.
Although the paper makes important adjustments in the methodology from the earlier
work comparing U.S. physician-to-population ratios with those of PGPs, there
are still important differences between what is being measured for PGPs and
for the United States as a whole. These differences weaken the validity of the
comparison. Furthermore, while the paper suggests that the findings have implications
for workforce planners (those conducting studies and making recommendations
on how many physicians are educated and trained in this country), the findings
are of greater relevance to policymakers and payers of care.
Methodological Concerns
The study attempts to adjust data from the selected PGPs and the United States
to ensure that the comparisons between the two use the same metric. Weiner does
a noble job in trying to make sure that the analysis compares apples to apples,
but several important differences remain. The adjustments, such as for out-of-plan
use, population variations, and scope of activities, are an improvement over
the methodology used in the earlier analysis.2 However,
there is still reason to question whether the PGP physician counts are comparable
to those for the United States as a whole.
Population differences.
The analysis makes an adjustment for differences in the population served by
PGPs compared to the U.S. population; however, this adjustment may not adequately
account for the differences in the populations. Reflecting the nation’s
approach to financing care, most enrollees in PGPs are employed; this may limit
enrollment of people with chronic illness and other high-volume users in PGPs.
According to data from the Medical Expenditure Panel Survey (MEPS), people with
limitations in activities of daily living (ADLs) are less likely than others
are to be enrolled in health maintenance organizations (HMOs).3
According to the Weiner paper, Kaiser Permanente PGPs, whose patients constitute
93 percent of the PGP population covered by the study, average only 2.9 percent
of their patients in Medicaid, compared with 11 percent for the United States
as a whole. Few PGPs serve the medically indigent, who represent about 13 percent
of all Americans.
Although the proportion of the PGPs’ patients covered by Medicare is similar
to that of the United States as a whole, it may be that the healthier elderly
are served by PGPs while those with chronic illnesses and others with special
needs, who have existing relationships with providers (including those in specialized
programs and clinics), might not be receiving care from PGPs to the same extent
as in the United States as a whole.4
Scope of physicians’
activities. A
not insignificant proportion of physicians in the general physician workforce
are involved in clinical research, teaching of medical students and residents,
and caring for populations with very high need—activities that relatively
few PGP physicians are involved with, at least in their PGP positions.5
Caring for the most critically ill, often in teaching hospitals or specialized
clinics, along with caring for populations with special needs, such as the homeless,
AIDS/HIV patients, illegal immigrants, high-risk pregnant women who do not receive
prenatal care, and others, is very intensive and often limits the number of
patients a physician can see in a typical day. If PGP physicians are involved
in comparable activities, it appears that they do so to a more limited extent.
Despite the shortcomings and the great difficulty of assuring the comparability
of the “physicians” being compared in the PGPs to those in the United
States as a whole, if these adjustments could be made, it is likely that PGPs
would still be found to be more efficient than the fee-for-service system is
in using physicians. The challenge, then, is to determine which elements of
the PGP system contribute to greater efficiencies and effectiveness and how
to achieve these benefits nationally.
Implications for U.S. Physician Workforce Planning
Weiner and others have suggested that PGPs’ lower physician-to-population
ratios have implications for how many physicians should be trained and produced
in the United States. Some suggest that since PGPs can provide high-quality
care with fewer physicians, then the nation really “needs” fewer physicians.
However, the experience with managed care efforts in the 1990s to control access
and limit the use of physicians indicates the difficulty of translating the
managed care experience to the broader U.S. health care system. It also demonstrates
the inappropriateness of applying the PGP staffing model’s parameters to
the nation as a whole.
Thus, while the PGP experience may help identify how many physicians the United
States would need if it changed its delivery and financing system, the experience
provides no guidance on how to change the system. In the absence of such a change,
reducing the supply of physicians is likely to add to access problems and sizable
gaps between supply and demand, not improved efficiencies or effectiveness in
the use of physicians.
While the PGP experience does not have direct implications for how many physicians
to produce in America, there are lessons to be learned from better understanding
why and how PGPs deliver effective services with fewer physician resources.
The paper notes that it is not just the use of nonphysician providers (NPPs),
since the national ratio per physician is now higher than the ratio for PGPs.
Perhaps PGPs use NPPs differently and more effectively. Perhaps the lack of
financial incentives for PGPs to provide more services, incentives to provide
preventive services, incentives to use group sessions for the chronically ill,
or a dozen other differences in how PGPs use physicians affect efficiency and
effectiveness.
Thus, while Weiner’s paper helps demonstrate that we can improve the U.S.
health care system, it does not provide guidance for physician workforce planners.
The United States should not limit its production of physicians because theoretically
it could get by with fewer physicians if its health care system changed. That
would be analogous, with similar results, to determining the nation’s future
oil supply based on the assumption that all cars in 2015 will get seventy miles
per gallon because we know that some cars can do so today. The likely result
would be massive shortages, accompanied by sharp rises in prices and public
dissatisfaction. Attempting to force system delivery changes through the use
of controls on supply is backward and likely to be ineffective in the long run.
Despite our concerns with the methodological shortcomings and the interpretation
of the findings, Weiner’s paper is an important step forward. In addition
to the improvements in methodology, it provides valuable insights for policymakers
and analysts. Further studies of the use of physicians and NPPs under different
systems of delivering and financing care—including the impact on efficiency,
effectiveness, and quality—would greatly benefit the nation.
NOTES
1. J. Weiner, “Forecasting the Effects of Health Reform
on U.S. Physician Workforce Requirement: Evidence from HMO Staffing Patterns,”
Journal of the American Medical Association 272, no. 3 (1994): 222–230.
2. Ibid.
3. Authors’ analysis of data in J.S. Banthin and A.K. Taylor,
HMO Enrollment in the United States: Estimates Based on Household Reports,
1996, MEPS Research Findings no. 15, Pub. no. 01-0014 (Rockville, Md.: Agency
for Healthcare Research and Quality, 2001).
4. See, for example, S. Nicholson et al., “The Magnitude
and Nature of Risk Selection in Employer-Sponsored Plans,” NBER Working
Paper no. 9937 (Cambridge, Mass.: National Bureau of Economic Research, July
2003).
5. See, for example, B.E. Vath, R. Schneeweiss, and C.S. Scott,
“Volunteer Physician Faculty and the Changing Face of Medicine,” Western
Journal of Medicine 174, no. 4 (2001): 225.
Edward Salsberg directs the Center for Health Workforce Studies, School of Public
Health, University at Albany, State Univerisity of New York, Gaetano Forte (gjf01{at}health.state.ny.us)
is a program research specialist there.
Please click on the author's
names to read related papers by Jonathan
Weiner, Francis
Crosson, W.
Bruce Fye, David
C. Goodman,
Fitzhugh Mullan, and Stephen
C. Schoenbaum.
DOI: 10.1377/hlthaff.W4.73
©2004 Project HOPE–The People-to-People Health Foundation, Inc.
|
