From Physician Glut to Physician Shortage (NYT)

August 31, 2012, 6:00 am

By UWE E. REINHARDT

Uwe E. Reinhardt is an economics professor at Princeton. He has some
financial interests in the health care field.

In my most recent post, I made light of the argument that the
Affordable Care Act would lead to a major shortage of physicians in
this country. I was unpersuaded in part because the newly insured are
likely to present only a marginal added demand for physician services.
More important, I am not sure what we mean by “physician shortage.”

Forecasters looking at the health work force have never reached a
consensus on the ideal physician-population ratio for this country.

Indeed, widespread worries over a looming physician shortage are a
relatively new phenomenon. They come at the time when experts are also
lamenting an “epidemic of overtreatment” of patients, said to cost
America $210 billion a year.

Throughout the 1980s, however, and until the late 1990s, the dominant
narrative among experts on the American health work force was that,
with the exception of primary care physicians, the United States faced
a large overall future physician surplus. There were only a few
demurrals from that dominant narrative.

The problem is that forecasting the future supply of and demand for
any type of health professional is a highly complex and nuanced
enterprise with wide margins of error (see, for example, Figures 9-1
and 9-6 in my 1991 paper).

Crucial in such forecasts is the assumption one makes about the
average annual physician productivity in future years. That variable
depends chiefly on two factors: the number of hours per year that
physicians typically devote to patient care, and the degree to which
physicians delegate to others tasks for which an M.D. degree is not
required – for example, administrative tasks to clerks or business
managers and certain medical tasks to physician assistants or nurse
practitioners trained to perform tasks now performed by physicians.

How crucial that assumption is can be inferred from a once highly
influential paper written in 1994 by Jonathan Weiner, a Johns Hopkins
University health services researcher. In that study, Professor Weiner
sought to estimate the impact of the then-impending Clinton health
reform on the country’s future work force situation.

Professor Weiner noted that, in 1992, well-managed, clinically
integrated, staff- or group-model health maintenance organizations
that were compensated by prepaid capitation (an annual lump-sum fee
per patient) required an average of only about 120 or so physicians
per 100,000 enrollees, while the overall ratio of patient-care
physician per 100,000 population in the United States was as high as
180 (about 220 in 2011; see Table 2 in this publication).

It appeared that the H.M.O. had pushed task delegation to nonphysician
personnel further than had the rest of the health system. Furthermore,
H.M.O.’s freed clinicians substantially from many administrative
chores that physicians elsewhere must perform. Such H.M.O.’s,
incidentally, would be the ideal form of the accountable care
organizations called for in the Affordable Care Act of 2010.

Assuming, when he made the forecast in 1994, that as a result of the
Clinton health reform some 40 to 60 percent of the United States
population would be enrolled in such H.M.O.’s by 2000, Professor
Weiner projected that the demand for and supply of primary care
physicians would be more or less in balance in 2000, but that the
supply of specialists would exceed the demand for them by more than 60
percent (a projected surplus of 165,000 physicians).

This prospect – widely accepted at the time — subsequently led the
prestigious Council of Graduate Medical Education to recommend in its
report of 1996 that “that the number of physicians entering residency
be reduced from 140 percent to 110 percent of the number of graduates
of allopathic and osteopathic medical schools in the United States in
1993.”

And why were health policy makers and work force specialists so
worried at the time about an impending physician surplus? Did not
standard economic theory predict that an imminent surplus would
drastically drive down physician fees – particularly specialists’ fees
— and thus make health care more affordable and accessible?

The problem is that this theory has found little empirical support in
the data, in part because third-party payment intervenes. Furthermore,
there has always been a strong belief, especially among policy makers,
that modern medical practice, when coupled with third-party payment,
is subject to an analogue of Parkinson’s Law. It is named after the
British historian Cyril Northcote Parkinson (1909-93), who promulgated
the law more or less in jest in 1955, then with regard to the British
civil service.

According to Parkinson’s Law, “work will expand to fill the time
available for its completion.” In medicine, its manifestation is
feared to be the overtreatment of patients – sometimes harmful – even
though individual physicians may sincerely believe that more care
implies superior quality of treatment.

As the late economist Eli Ginzberg, an early pioneer in work force
studies, noted as early as 1966: “Physicians are in a position to
create their own demand.” He added that the effective use of physician
manpower “depends in the first instance on a taut supply of
physicians.”

Academic economists since that time have tied themselves into analytic
knots over whether or not Ginzberg was right, in exercises reminiscent
of medieval scholasticism. At the theoretical level, their models are
mathematically elegant but lack predictive power. The data available
at the empirical level does not allow economists to distinguish
between health care actively demanded by patients and health care
passively accepted on the doctor’s recommendation, nor between
services prescribed by doctors in good conscience and those rendered
mainly to shore up doctors’ incomes.

As on so many other areas of the real world, the views of economists
on this matter cancel one another out.

Policy makers in the real world, however, seem to have no doubt that
Parkinson’s Law applies to medical practice, as well. Consequently,
they prefer paying physicians by annual capitation or bundled payments
instead of “inflationary” fee-for-service, and they often seek to
impose global budgets on physicians.

If Eli Ginzberg was right – and often he was – the suspected physician
shortage now imputed by critics of the Affordable Care Act may
actually drive our health system into more efficient medical practice.
Step No. 1 in that direction, of course, would be to lighten the
enormous administrative load now heaped by our health insurance system
onto physicians devoted to rendering patient care.