Do these guys know why they're applauding?

Do These Guys Know Why They're Applauding?

The main objective of the Physician Payment Reform (PPR) portion of the Omnibus Budget Reconciliation Act of 1989 from a federal perspective is to do for Medicare Part B providers what the prospective pricing system did for hospital inpatient activity beginning in 1983: reduce the annual compound rate of growh in Medicare outlays from the 15-20 percent range to the 5-10 percent range. The achievement of this major objective, however, was made politically possible (at least insofar as the medical profession is concerned) because of a peripheral feature of PPR: the probable reallocation of income among physician specialties by increasing payments for more "cerebral" services and decreasing payments for more procedural CPT codes.

If you were to set out to construct a system to divide a conquer the medical profession, I believe it would be difficult to devise a more effective methodology than the Resource-Based Relative Value Scale (RBRVS) that is at the heart of the PPR income reallocation scheme. The majority of U.S. physicians are reported to favor RBRVS. This is not too surprising, in that the majority of U.S. physicians are primary care givers who will initially see their fees per unit of service increase. This is an exceptionally short-sighted perspective, of course, as it is a virtual certainty that, just as was the case with prospective pricing using DRG, apparent initial generosity will rapidly give way to highly constrained or even nonexistent increases. (1) By offering an illusory carrot to the primary care, lower income specialties and exploiting their latent discomfort with the higher average incomes of the more procedure-oriented specialties, the government succeeded in diverting the medical profession's attention from the more basic economic issues in PPR.

Even if the RBRVS carrot were not illusory, however, its fundamental irrationality remains. As hard as it may be to believe, there is possibly rooted in distant reality an explanation of why warm fuzzy office visits with significant cerebral input have never commanded fees commensurate with invasive, capital-intensive, institutionally subsidized procedures that do things like restore eyesight, relieve angina pectoris, or repair traumatic damage. Notwithstanding the fact that current fees have undergone years of being subjected to usual/customary/reasonable constraints, across-the-board freezes, maximum allowable increases, participating physician premiums, etc., there remain, however faint and diluted the influence, marketplace-determined relationships from the original charge-based Medicare system implemented in 1966. The fact is, rightly or wrongly, patients genuinely value historically high-priced procedures more than historically low-priced services. Whether or not that is fair, it is true.

Consider the following analogy to manufacturing. Assume one could identify the long-run average cost of the resources consumed in the production of one unit of each of two outputs, (2) a Ford Taurus and a Jaguar. Suppose further that we can observe in the marketplace the net selling prices of the two automobiles. Our findings are summarized in the table below. Th is is obviously unfair. The markup on the Taurus is 25 percent, while that on the Jaguar is more than four times as much. Clearly this situation must be corrected, because the marketplace is discriminating against the Taurus. The solution is to regulate the prices on these automobiles to ensure that they have equal percentage arkups, i.e., to ensure that their prices are commensurate with the costs of the resources used in their production. Based on the regulator's need to reduce overall expenditures for automobiles, we shall select a 50 percent markup and reestablish equity between the two automobiles (see the table). Taurus has received a 25 percent increase in price ($3,000), while the Jaguar has suffered a nearly 31 percent decrease ($16,000). Substitute internal medicine for the Taurus and surgery for the Jaguar, and you have the RBRVS. The premise that payment should be commensurate with the cost of production (even if we knew what it was for medical services, which we don't) is wholly without economic basis, however good it may make family practice physicians feel.

To whatever extent market forces are a factor in the medical marketplace, and that extent is certainly not zeror, we are about to embark on guaranteed misallocation of scarce resources, providing a mix of services to at least Medicare beneficiaries that is different from what they might have chosen on their own. Not o nly does a differential mix of services result that is suboptimal from patients' perspectives, but regulated prices also tend to stifle investment in innovation, development of new technology, and other quality improvements. In fact, it is in precisely those specialty areas most susceptible to innovative technology that fee changes will most contrain medical advances.

Footnotes

(1) When the price of cerebral services initially increases, even if Medicare enrollees experience that increase only in the 20 percent copayment, demand can be expected to fall. Conversely, if procedure that historically have been highly valued become relatively cheaper, demand will increase. At the margin, given the likelihood that demand elasticities are much greater for the former than for the latter, the "winner" specialties may win relatively little or even lose income, while the "loser" specialties may lose less than originally predicted, provided they currently have idle capacity. And if HCFA would unexpectedly behave irrationally and allow fees for cerebral services to continue to increase in real terms, those higher fees would still not necessarily translate into higher income, for the same reasons.

(2) There is substantial doubt that Hsaio's RBRVS sork comes anywhere close to meeting this assumption for any of the 7,000 or so outputs (CPT codes) it purports to measure. For each of 18 specialties, an average of 110 physicians, who were further stratified to an average of 11 for each of 10 geographic regions, were asked to score an average of 23 services against one referent service within their specialty. They were asked their personal opinions of the relative amount of "work" involved among the 23 services. Multispecialty panels then "harmonized" the specialties relative to each other on a common numerical scale. Of the 407 services thus arrayed, only 157 were studied in depth to identify pre-, intra-, and postservice work. These results were extrapolated to the 407 ranked services, were further extrapolated to 2,000 high-volume services used by Medicare (the original services not being Medicare-specific), and will subsequently be extrapolated to the approximately 7,000 CPT codes. Within specialties and geographic regions, a percentage add-on was then applied to each procedure code's score to account for average costs of practice separately self-reported to the National Opinion Research Center. A further percentage add-on was applied to all scores for codes within a given specialty based on the amortization of opportunity costs associated with differential lengths of residency training over differential expected practice lives of such specialists. No verification of practice costs was undertaken, and the percentage add-on allocation method assumes that actual volumes of procedures performed are proportionate to their work scores and uniform across physicians. Small sample sizes, reporting and "harmonizing" biases (respondents and panel members knew the purpose of the study), extraordinary extrapolations from small bases, and arbitrary overhead and opportunity cost allocations all contribute to a lack of objective confidence in the relative values reported.

Hugh W. Long, PhD, is Associate Professor of Health Care Management, A.B. Freeman School of Business, Tulane University, New Orleans, La.