Why such a hurry?

Why Such a Hurry?

Payers of health care services and the academic community continue to be concerned that medical care costs are too high and are increasing too fast. Interestingly, the public does not seem to share the same concerns, as public opinion polls of the past 8-10 years show clearly that people want more medical care and are willing to pay for it.

Implementation of the prospective pricing system for inpatient care for the Medicare population was accomplished with the firm belief that it would contain costs, or at the very least reduce substantially the rate of cost inflation. It has done neither consistently.

More recently, the rapid increase of payments to physicians has led to the same kind of thinking. Somehow there must be a single new national system that will control increasing expenditures. Medicare settled on the resource-based relative value scale (RBRVS) system that will be introduced throughout the United States over the next few years. Initially, it will be aimed at services provided to the Medicare-eligible population, but undoubtedly it will be quickly adopted by Medicaid programs and by private payers as well.

Richard M. Lauve, MD, has presented |ery clearly many of the potential effects of this new physician payment mechanism on payers, providers, and consumers of care (see page 6). If his scenario is correct, the question we must ask is, "Why are we going to this system in the first place? Will it achieve what is expected of it, and will it achieve it in a way that will be better than other realistic possibilities?"

We have a long history of relative value systems. The most widely used, the California Relative Value System (CRVS), was compiled and adopted by the California Medical Association in the mid-1950s. It spread quickly and was used in many states by both public and private payers of services. By the time of its abrupt withdrawal in the mid-1970s, a large percentage of physicians were probably paid a sizable portion of their income using the original CRVS or a variant.

The proposed RBRVS is founded on research of relatively small groups of physicians whose patterns of work, resources used in providing care, and their individual opinions were melded into a unified proposal that relates the activities of physicians to those of their colleagues while incorporating due consideration for the cost of providing services. We have no real idea of the effects of this new system on providers, payers, and consumers derived from sound experimentation; we are left merely to speculate. Certainly if we desire to implement such a system it seems reasonable to test it in a number of different locales with different kinds of physicians and in different organizational and layment frameworks.

The tension facing any new payment system is one of reconciling cost control with other desirable social goals. For example, the United States wants any physician payment system to pay equitably and to provide incentives to physicians, patients, and payers to maintain or improve access, quality, accountability, continuity, and the like. But trade-offs must be made if a physician pricing system is to meet multiple and often conflicting objectives. These trade-offs center around two issues. First are the mechanics of the physician pricing system and its impacts:

* Efficiency of service production (paying the correct price so only appropriate services are provided).

* Variation in prices to encourage or discourage specific services (for example, increasing the price paid for pap smears while concurrently decreasing the price paid for tonsillectomies).

* The system's ability to respond to changes in its environment (new medical technology, new organizational and paying formats, and the like).

Second is the cost of implementing, administering, and adapting the system to continuous change. The major issue is how much, if any, emphasis to place o n control of service prices, how much on volume controls, and how much on controlling total expenditures. When regulation replaces the market, unanticipated side effects occur that lead to additional bureaucratic solutions, further response from the regulated, and so on.

From the long history in the United States and in other countries in trying to control medical expenditures, only a very few truisms are known. In order to control costs, one must control volume of services and/or total expenditures for those services. Where either or both have been successfully accomplished, costs have been controlled. Canada is the closest example.

At the same time, other countries successfully use relative value scales similar to CRVS. Canada and France are the two most notable. Canada has had substantially more success than France in controlling costs because of the manner in which it controls volume, i.e., through controlling physician revenues. France had equally good success initially but recently has had less because, while price per service is controlled, volume and total expenditures are not.

If the United States wants to control expenditures for physician services while improving access and quality and paying providers equitably, the system used may be less important than the incentives and disincentives inherent in the system. Slowing or controlling costs is only achievable when volume and expenditure controls, either direct or indirect, are instituted or when aspects of the market are allowed to operate and incentives are used to encourage or discourage patient and provider behavior.

The two-decade use by CRVS is a real alternative to RBRVS. Comparison of CRVS with its counterparts in Canada and France shows that all three are very similar in how physician activities, procedures, and diagnostic tests are evaluated. This immediately improves the likelihood that all of the systems are good mirrors of the actual practice of physicians, irrespective of nation. If this is the case, why are we moving to a new, expensive, and complicated system while ignoring successful precedents? And why are we moving so quickly to a national scheme without testing alternatives at the local level. Without such pretesting and knowledge of effects. we are likely to achieve as little as with the DRG-based hospital payment program--a one-time saving from a two-year reduction of the inflation rate. Not insignificant, but not as much as could have been achieved if other realistic alternatives had been explored.

Bernard S. Bloom, PhD, is Research Associate Professor, University of Pennsylvania, Philadelphia.