The challenge of Medicare physician spending

The dilemma for Medicare is to devise a physician payment strategy that promotes the delivery of the appropriate volume and mix of services, provides good access to physician services for seniors, and moderates or controls physician spending.

With the new emphasis in Medicare and elsewhere in health care on rewarding “good performers”—that is, those who provide high-quality, patient-centered, and efficient care—the challenge has become even more daunting.

Current Law

Under current law, the physician fee schedule is based on the (relative) resources required to provide each of the 7,000+ services listed under the CPT code. The resource-based fee schedule was implemented in 1992, based on data from the mid-1980s, and contains values reflecting work effort, practice expenses, and malpractice liability involved in providing a particular service. Unlike other parts of Medicare, annual changes in the fee schedule are modified by the difference between actual spending and the “targeted” spending. The update is higher than it would otherwise be if spending is below the target and lower if spending is above the target.

Although the physician community did not seem particularly concerned when the “sustainable growth rate,” or SGR, concept was passed into law as part of the Balanced Budget Act in 1997, which tied the physician spending target to the growth in the economy, it has become a matter of great concern over the past few years. Because the economy was growing at a robust level in the late 1990s, updates to the fee schedule initially were larger for physicians than for most other providers in Medicare. Since 2002, however, the scheduled updates have involved reductions in fees, reflecting the slower growth in the economy and the faster growth in the volume of services provided by physicians. Nonetheless, in every year but 2002, Congress spared physicians these payment reductions, refraining from writing these negative updates into law. An unfortunate result is that this problem has become more expensive to fix.

Physician Payments Versus Other

Medicare Payments

Most services provided under Medicare are paid using “bundled” payments. With bundled payments, a single payment is made for all of these services associated with a particular event—like a hospital stay. Bundled payments—which are used to pay hospitals, home care agencies, and nursing homes—produce incentives to provide the services contained within the “bundle” as cheaply and efficiently as possible. In contrast to the strategy used for physicians, Medicare updates bundled payments using a “bottom-up” strategy, which includes measures for inflation and adjustments for productivity. The strategy for physicians is a “top-down” one based on spending relative to a spending target.

Physician Payment History

Looking back since 1965, the history of payments to physicians under Medicare falls into three different periods. During the first period, from 1965 to 1984, physician fees were based on historical charges. With few constraints in place, it is not surprising that charges and volume increased substantially during this period. During the second period, from 1984 to 1991, fee growth was limited by the Medicare Economic Index. This period also experienced a rapid growth in spending.

The primary lesson learned over these first two periods was that controlling fees under a disaggregated fee schedule is not a very effective way to control or moderate spending on physician services. During these 25 years, spending for physician services grew more than 2.5 percentage points faster than spending for all services.

Legislation passed late in 1989 that introduced the concept of a resource-based relative value system, brought a series of changes to physician payment, leading to the third historical period, from 1991 to the present. These changes included:

  • Abandoning a charge-based system
  • Limiting the liability that beneficiaries
  • could face (that is, limiting balance billing by physicians)
  • Redistributing physician payments
  • Introducing explicit controls on volume

The first volume-control mechanism was the volume performance standard, or VPS. It tied annual updates to physician spending relative to a target by impacting the conversion factor, which translated the relative values of the RBRVS to dollars. Over the next several years, however, problems primarily relating to unstable updates became increasingly apparent although the rates of increase in spending definitely declined. From 1992 to 1998, for example, the MEI varied from 2.0 percent to 3.2 percent, while the annual update varied from 0.6 percent to 7.5 percent.

The VPS was replaced by the SGR approach in 1997. It tied growth in the physician fee schedule to the growth in “real” (that is, inflation-adjusted) growth in the economy, per capita. The update now adjusts the MEI by the cumulative spending on physician services relative to the target. The problem is that since 2002 and for the next four years of the budget period, the update has been and is predicted to be a negative 4 percent to 5 percent.

The concern with spending targets is not that the targets don’t work to control spending. The VPS and SGR have shown that if used, tying updates to the difference between desired spending and actual spending will indeed control future spending growth. The problem is that the resulting reductions in fees are widely viewed as unfair, because they apply uniformly to all physicians, without regard for each physician’s actual behavior; they could result in reductions to access; they reinforce the compartmentalized nature of spending on Medicare; and they are completely contrary to the spirit of results-based payments, or pay for performance.

The concern with spending targets is not that the targets don’t work to control spending. The VPS and SGR have shown that if used, tying updates to the difference between desired spending and actual spending will indeed control future spending growth. The problem is that the resulting reductions in fees are widely viewed as unfair, because they apply uniformly to all physicians, without regard for each physician’s actual behavior; they could result in reductions to access; they reinforce the compartmentalized nature of spending on Medicare; and they are completely contrary to the spirit of results-based payments, or pay for performance.

The situation poses a challenge for Congress if payments that are as disaggregated as those under the fee schedule will always result in rapid spending increases without the use of a target. If so, Congress will need to consider whether the movement toward a somewhat less disaggregated fee schedule, with or without a gradual shift to a pay-for-performance strategy, could improve the financial environment for physicians while avoiding subjecting the Medicare program to rapid rates of increased spending.

None of these payment changes will be easy. All will require several years of transition and some experimentation. Given the pressure to protect physicians from repeated fee reductions, the next session of Congress is none too soon for the nation’s lawmakers to start considering appropriate reforms.


Gail R. Wilensky, PhD, is a senior fellow at Project HOPE and a commissioner on the World Health Organization’s Commission on the Social Determinants of Health. She was the HCFA administrator from 1990 to 1992 ([email protected]).

Gail Wilensky, Ph.D.

Categories