Medicare and physician autonomy.

Introduction

It would be possible to view the impact of Medicare on physicians from many perspectives: the impact on individual physicians, on a particular specialty, on academic physicians, on graduate medical education and physician specialization, on quality of care, on physician incomes, on physician autonomy, or on a variety of other aspects of medical practice. We have chosen to focus on physician autonomy, a topic that has gained prominence recently as a result of its perceived erosion.

One of the critical questions that has been raised about physician autonomy and Medicare is whether or not physicians have traded reduction of clinical autonomy or discretion for preservation of economic autonomy. Although often couched in terms of quality of care and access to care, physicians, particularly through organizations such as the American Medical Association (AMA), have in fact focused on the economic autonomy of physicians. Yet concern about loss of clinical autonomy is a major morale issue within the medical profession (Lee and Culbertson, 1990). Lewis et al. (1991) reported “growing dissatisfaction with the practice of internal medicine, primarily related to concerns over loss of clinical autonomy….”

Freidson has been identified for the last two decades as the leading theorist on professional autonomy. He has recently defined autonomy in the following manner:

Taken as an ideal type, complete autonomy is sustained by an occupational monopoly embracing several dimensions. It is first of all an economic monopoly: the profession controls recruitment, training, and credentialing so it can regulate directly the number of practitioners available to meet demand. This has obvious implications for income. Economic monopoly is viable, however, because professional autonomy also includes a political monopoly over an area of expertise; the profession is accepted as the authoritative spokesman on affairs related to its body of knowledge and skill, and so its representatives serve as expert guides for legislation and administrative rules bearing on its work. Furthermore, the profession has an administrative or supervisorial monopoly over the practical affairs connected with its work; its members fill the organizational ranks which are concerned with establishing work standards, directing and evaluating work. “Peer review” rather than hierarchical directive is the norm. Clearly as I have defined it, professional autonomy represents a privileged position of some significance (Freidson, 1994).

Freidson's emphasis in this recent definition on economic dimensions of autonomy is a departure from the thought of earlier theorists who stressed the clinical aspect of autonomy. The prominent American sociologist Talcott Parsons emphasized the superior position by virtue of technical expertise of the physician as essential to the public good. Parsons held that in order to maintain regulation of their patients, physicians must have the right as a profession to control the conditions of their clinical work and the patients they accept (Parsons, 1964).

Context

In viewing the relationship of Medicare to physician autonomy, it is useful to recall the historic opposition of the medical profession, particularly organized medicine, to any role of the Federal Government in health care financing, except for a limited role in relation to indigent care.

The historic opposition by organized medicine, particularly the AMA, to a significant Federal role in the financing of national health insurance or the more limited proposals related to the elderly have been documented in detail by others (Starr, 1982) and will not be repeated. The context and the flavor of the times were elegantly described by Ball (1995) in his article, “What Medicare's Architects Had in Mind.”

Although President Truman first proposed a program of national health insurance in 1945, it was not until after his election in 1948 that AMA leadership became alarmed about the possibility that Congress might do something. The AMA campaign was well organized and well financed and included pamphlets in physicians' offices, press attacks, public speakers, and vigorous lobbying against the proposal supported by President Truman. The attack was bitter and ultimately successful. The idea of hospital insurance for the elderly was first floated by Oscar Ewing, head of the Federal Security Administration in 1952. President Eisenhower was elected for his first term later that year, Oscar Ewing departed, and there was little support for such proposals in the political levels of the executive branch for the next 8 years (Ball, 1995). In 1957, Representative Aime Ferand (D-RI) introduced the first of a series of bills to provide hospital insurance for the elderly (Litman and Robins, 1984). In 1961, after President Kennedy's election, it was re-introduced in the House and Senate as the King-Anderson Bill.

It is important to recall the context of the mid-1960s when Medicare was enacted and implemented. The Civil Rights Act was passed less than 2 years before Medicare's passage in 1965, without any serious consideration of its later impact on the practice of medicine through the desegregation of hospitals, particularly in the South, and the resultant enhanced patient access to care—a laudable but unforeseen consequence. Finally, the impact of the rising costs of health care on Medicare policy was not fully appreciated in the beginning. In time, rising costs far in excess of increases in gross domestic product became the overriding force driving the Medicare policies affecting physicians.

After President Johnson's landslide victory in 1964, the likelihood that Medicare would be enacted was substantially increased, but Congress included a number of provisions to mute physician opposition. Medicare was to build on the existing system, not reform it. Claims processing and payment were to be administered by private organizations under contract as Medicare carriers to provide a buffer between physicians and government. The Blue Shield plans and commercial health insurance plans that became carriers were allowed wide discretion in interpreting Medicare policy.

Congress also adopted a payment method designed to attract physicians, permitting them to bill what they normally charged their privately insured patients, the “customary, prevailing, and reasonable” (CPR) charge. Medicare payment was based on this “reasonable” charge, defined as being the lower of the physician's actual charge, the physician's customary charge (the physician's median charge for service from the previous year), or the prevailing charge in the locality (set at the 75th percentile of the distribution of customary charges in a locality). In addition, physicians were allowed to bill their patients directly through the practice of balance billing, which allowed them to collect more than Medicare's reasonable charge.

This was the context in which Medicare was enacted and signed into law on July 30, 1965. At the time, Congress mandated in section 1801 of title XVIII that “nothing in this title shall be construed to authorize any Federal officer or employee to exercise any supervision or control over the practice of medicine or the manner in which medical services are provided …” (Blumenthal, 1988). This initial intent showed the desire of the Federal Government to avoid conflict with the perceived sphere of influence of medicine, thus guaranteeing wide protection of both clinical and economic autonomy.

Within 3 months, it became apparent that the Civil Rights Act (title VI) would have to apply to hospitals if they were to receive payment from Medicare. The U.S. Public Health Service, under the leadership of Dr. William Stewart, Surgeon General, was enlisted to assist the Social Security Administration in a broad-based, intensive effort to ensure hospital compliance with the Civil Rights Act. In sum, the hospitals that had practiced segregation agreed to desegregate—everything from separate drinking fountains to inpatient and outpatient care. This action by the Federal Government had a profound impact on physician autonomy. In the definition presented earlier, clinical autonomy was defined in part as the ability of practitioners to select or reject patients/clients from their practices. Now, physicians were no longer free to segregate their patients when they were hospitalized.

Concept of Physician Autonomy

Let us turn to a more detailed review of Medicare and physician autonomy, as the Medicare program has become the dominant force in setting physician payment policy. We will consider the concept of physician autonomy, specifically its economic and clinical dimensions, and the embodiment in public policy of these dimensions of autonomy in the Medicare program and corresponding influences upon the medical profession.

Autonomy has been cited by Freidson as the key defining characteristic in the organization of professions. Freidson (1970) suggests that “functional autonomy” is defined in medical occupations by “the degree to which work can be carried on independently of organizational or medical supervision, and the degree to which it can be sustained by attracting its own clientele independently of organization or referral by other occupations, including physicians.” The key point here is that medicine as a profession is at the pinnacle of this occupational hierarchy, and in much of the 20th century has been able to control, in cooperation with the Federal and State governments, the basic terms of medical work. Self-governance of the profession is key to a definition of autonomy. Perhaps Starr (1982) portrayed this concept most effectively when he referred to medicine as a “sovereign” profession.

Schulz and Harrison (1986) have attempted to define specific elements of autonomy based on an empirical survey of physicians. Five elements of their definition can be described as “clinical” in nature and include control over: (1) the nature and volume of tasks; (2) the acceptance of patients; (3) diagnosis and treatment; (4) the evaluation of care; and (5) other professionals. Three elements of their definition might be considered economic in character and include freedom of choice of specialty and practice location and control over earnings (Schulz and Harrison, 1986).

Freidson contends that the clinical and economic interests of the profession have become mixed, and in the process, corrupted. Freidson makes autonomy and its preservation the foundation of the economic and consequent political strategies of the medical profession. He notes the resistance of medicine to involvement of external entities in its affairs as defined by the profession itself. He then notes the established monopoly position of the profession over the use of select scarce resources and services and suggests that “freedom to set the terms of compensation is, without some form of professional self-regulation in the public interest, obviously subject to abuse” (Freidson, 1970).

Freidson argues that the profession has made no effort at self-regulation of fee practices on the part of its members. Rather, it has left any attempt at redressing patient grievances to the courts. He suggests that in the United States, the profession has made little effort “to insure that its members do not abuse their privileged economic position by seeking more than a ‘just price’” (Freidson, 1970). He states that society in the United States has had a difficult time establishing a concept of a “just price,” but he is certain that a free market model of competition will not achieve this because physicians enjoy a regulated advantage in the division of labor as a result of preferential licensing acts.

Freidson concludes that a “flaw” exists in the autonomy of physicians, in which the economic and clinical interests of the profession become intertwined, and in which economic interests may at any point in time prevail. In this regard, his critique anticipates the professional concerns voiced most forcibly by Relman (1986) in his observations of the fiduciary responsibility of the physician and the debasement of this responsibility that he sees occurring in profit-oriented medicine. Freidson refuses to fall prey to the notion that autonomy is a purely economic device, choosing to see its development from a variety of social forces. In refuting a purely economic causal theory, he states that, “Consulting professions are not baldly self-interested unions struggling for their resources at the expense of others and of the public interest” (Freidson, 1970). Rather, it is a perception of an entitlement to a superior level of resource as a result of the insularity of the profession from the public that creates this “flaw.”

Reinhardt (1988) is particularly persuasive in calling attention to the connection of clinical issues of autonomy and economic conditions, especially as viewed from within the profession of medicine. He cites a physician colleague who summarizes this theme as “the serious damage society inflicts upon patients when limits are placed on physicians' clinical freedom to compose medical treatments as they see fit and on their economic freedom to charge whatever honoraria they deem honorable” (Reinhardt, 1988). As an economist, he is especially sensitive to the potential drift of Evans' (1984) “not only for profit” medical-economic ethic to one that is distinctly for-profit, first and foremost. He adds that the economic imperative of joint ventures in which physicians become economic partners of hospitals and investors or of direct ownership of imaging and laboratory devices to which the physician refers patients will further erode the trust basis of autonomy (Reinhardt, 1988). As Gray (1983) notes in the introduction to his study of for-profit health care, trust as a basis for professional autonomy is under attack as a myth of the profession to enhance status while at the same time preserving monopoly privilege and power in the economic sphere. This is a significant criticism, for Freidson has defined on several occasions a service orientation of trust as a social contract of the profession with society that necessitates and legitimizes autonomy for the profession (Freidson, 1970). If this contract is violated, how can autonomy for the profession legitimately be sustained?

It is Reinhardt's assertion that the absence in the United States of an overall program of budgetary control over medical expenditures, as is characteristic of the prominent European systems, results in unparalleled micro-management at the clinical level to achieve cost control unattainable on a larger scale. He writes that “…if the bureaucrats cannot somehow impose upon the healers an overall budget constraint ex ante, then they will sooner or later be driven to control their outlays on an ongoing basis, by monitoring each and every transaction for which they pay—that is, by second guessing both the providers' clinical and pricing decisions” (Reinhardt, 1988). This appropriation of the clinical dimension of autonomy would be regarded as intolerable by physicians in other medical care systems. He suggests that “European and Canadian physicians would be appalled at the numerous intrusions into clinical decisions now routinely made by these external monitors in the United States. They probably would rise up in arms over that loss in clinical autonomy” (Reinhardt, 1988).

It seems problematic that physicians in the United States would willingly and knowingly sacrifice the clinical element of autonomy that Freidson considered to be the more consequential element of his two-part definition of autonomy. Clinical autonomy, after all, constitutes the primacy of the physician in the health care division of labor and is the basis on which arguments for political and economic autonomy are formed.

Reinhardt's answer to this seeming paradox is that physicians in the United States have traded off clinical autonomy “in their tenacious fight to preserve the individual physician's right to price his or her services as they see fit” (Reinhardt, 1988). This observation has been distilled into a formula referred to as Reinhardt's “Law” or “Irony.” Reinhardt has summarized his law as follows: “In modern health care systems, the preservation of the healers' economic freedom appears to come at the price of their clinical freedom” (Reinhardt, 1988). The application of Reinhardt's Law to the late-20th-century United States scene would appear to indicate a priority on the part of physicians to pursue economic betterment at the expense of clinical autonomy. If so, this would be critical in reformulating a definition of autonomy for the future, for this observation implies the willingness of physicians to sacrifice control of the division of labor. This strategy may also ultimately undermine the ability of physicians to continue their dominance of the political economy of health services.

Medicare's Impact on Physician Autonomy

At the time of the establishment of Medicare, the Federal Government deferred to the medical profession's definition of autonomy in both clinical and economic realms by accepting the principle of usual, customary, and reasonable fees. This was based on the convention that it was the physician's prerogative to establish prices for services (Starr, 1982). Physicians were to be left alone by public policy design to structure their clinical work and exercise relative freedom in the economic arena.

As Starr has observed, however, the tension “between a medical care system geared toward expansion and a society and state requiring some means of control over medical expenditures” led to modifications in Medicare, which were first observed in the area of economic autonomy and subsequently in the clinical dimension (Starr, 1982). Medicare expenditures for physician services grew rapidly from the outset of the program, and both the price and volume of services rose rapidly. Part B of Medicare (primarily physician visits) grew from 18.1 million visits in 1967 to 43.8 million in 1970 and 155 million in 1980. Expenditures rose from $900 million in 1967 to $10.1 billion in 1980 (Health Care Financing Administration, 1996).

Initially, the impact of these program modifications was observed in the economic realm. However, as Reinhardt predicted, the perceived reduction in economic benefit to the profession has also resulted in programmatic compromises that have limited clinical autonomy. These latter changes have been more subtle than the economic changes but are nonetheless real elements of the historical development of the Medicare program. These alterations in the program are summarized in Table 1.

Table 1

Effects of Medicare Policy Revisions on Physician Autonomy

Revision	Economic Autonomy	Clinical Autonomy
Title XVIII (1965)	Reinforced and legitimized economic autonomy in public policy	Federal Government reinforced clinical autonomy—cannot interfere in practice of medicine
Wage and Price Controls (1971)	Direct effect; fee freeze	No direct effect
Professional Standards Review Organizations (1972)	Indirect effect	Negligible direct impact on physicians; sanctions expressed through hospital denials
Medicare Economic Index (1975)	Direct effect	None
Prospective Payment System for Hospitals (Diagnosis-Related Groups) (1983)	Indirect effect on most physicians, but direct effect on hospital-based specialists	Indirect effect through hospitals
Deficit Reduction Act of 1984	Direct effect; incentives created to limit full fee recovery through voluntary Physician Participation Program	Indirect effect
Omnibus Budget Reduction Act of 1986	Direct limitation through establishment of price maximums through “maximum allowable actual charge” limits on non-participants; reduction in prevailing charge for overvalued procedures	Indirect effect limited to specific specialties
Omnibus Budget Reduction Act of 1987	Reduction in prevailing charge for overvalued procedures	Indirect effect limited to specific specialities
Consolidated Omnibus Budget Reconciliation Act of 1985 (Physician Payment Review Commission Established)	Congress mandated direct limitation through schedule for service based on relative values for clinical work of physicians	Indirect effect; slight clinical intervention through volume monitoring
Omnibus Budget Reconciliation Act of 1989	Direct limitation through 5-year implementation of fee schedule; balance billing limits, volume performance standard	Indirect effect at practitioner level, but possible impact on physicians
The Future: Managed Care Through At-Risk Prospective Payment	May restore some economic autonomy to physicians through “delegation” of resource control	Unclear at present: Utilization controls that limit clinical autonomy may or may not result from new delivery structures

SOURCE: Culbertson, R., Indiana University, and Lee, P.R., U.S. Department of Health and Human Services, 1996.

Wage and Price Controls (1971-74)

The first intrusion into the economic autonomy of physicians occurred in 1971, with the introduction by the Nixon administration of wage and price controls. Although this program was part of a general approach to deal with inflation throughout the economy, the health industry was singled out for specific attention. Fee increases were limited according to stringent Federal price guidelines, constituting a direct attack on the premise of economic autonomy. This program remained in effect through 1974 for the health sector, the last segment of the economy to be relieved of such controls (Litman and Robins, 1984).

Professional Standards Review Organizations (1972-Present)

In 1972, the first foray into clinical autonomy through economic sanctions was instituted in the passage of Public Law 92-603. This program, established in the face of significant but unsuccessful opposition by organized medicine, established a review program to ascertain the appropriateness and quality of care delivered in hospitals to beneficiaries of Federal programs. Certainly in retrospect, it may be argued that this program was a benign one with respect to its impact on clinical autonomy. It functioned on the basis of peer review committees within the structure of the hospital organization, which were in turn comprised primarily of physician members. It may be argued that this approach was not in conflict with the key characteristic of professional autonomy identified by Freidson of judgment of practice by one's own professional colleagues.

Furthermore, the economic impact upon physicians of the Professional Standards Review programs was quite muted as well. Sanctions, when applied, were limited to reduction of hospital payment for inappropriate stays or lengths of stay and were applied concurrently or retrospectively (Gray, 1991). It may be argued that a pattern of indirection in matters that might impact upon clinical autonomy was deliberately built into the Professional Standards Review Organizations and was to be a continuing feature of Medicare policy throughout the next 15 years.

Medicare Economic Index

In 1975, the Medicare Economic Index (MEI) was established to address concerns regarding medical price inflation following the discontinuation of price controls. Under this program, the MEI was used to adjust prevailing charges. The significance of this program in relation to the economic autonomy of physicians was the break in the linkage of actual charges to Medicare payment rates. Following the enactment of the MEI, physicians might raise their rates for fee-for-service patients but observed significantly lesser increases in Medicare-allowed payments for comparable services through the generally lower allowable percentage adjustments of the MEI.

Diagnosis-Related Groups (1983)

A revolutionary change in the payment of hospitals under Part A of the Medicare program occurred in 1983 with the enactment of a system of prospective payment for hospitals. This system dramatically restructured financial incentives by defining specific diagnosis-related groups (DRGs) to represent conditions for which patients are hospitalized and setting specific payment amounts for each group.

This program placed hospital organizations at risk for formula-based payments under Medicare, whereas previously, payment of “costs” to the hospital had been assured. The DRG system was geared to equate levels of care with resources necessary to produce that care and to penalize “inefficient” hospitals.

Hospital-based physicians, such as radiologists, anesthesiologists, and pathologists, were brought directly into these discussions of economic issues and their consequences for hospitals. Practice arrangements of these physicians in many cases were restructured into contractual or private-practice arrangements to remove these expenditures from overall hospital costs. Although attending physicians were not placed directly at risk for hospital performance under this program, policy changes in hospital payment clearly affected physician behavior. Shorter lengths of stay and fewer hospital admissions attained through this program led to a change in practice and movement of physician services away from the inpatient setting to ambulatory environments.

The enactment of the DRG program, although not directly infringing upon the clinical autonomy of physicians, was nonetheless a cause of concern for the medical community. Colombotos and Kirchner (1986) published a study based upon a survey of physician attitudes in which physicians linked the DRG concept for treatment and the direct control of physician fees by the government as the two most distasteful proposals for the future practice of medicine. They suggested that DRGs would result in explicit protocols and standards for care, which would in turn limit the clinical autonomy of physicians. Direct government control of fees would obviously limit their economic autonomy (Colombotos and Kirchner, 1986). Their prediction was that physicians would experience both forms of infringement on their historic autonomy in the 1990s. They projected that “during the next decade clinical protocols and standards, spearheaded by the DRG concept, will probably exercise an increasing influence on the clinical decision-making of physicians. In addition, the fees of physicians will probably be fixed, first under Medicare, and then under other government-financed programs, such as NHI” (Colombotos and Kirchner, 1986). They then proceed to construct a specific scenario for the future of clinical autonomy and its economic counterpart and state that “the clinical autonomy of physicians—and their pocketbooks—are likely to fare better if clinical protocols and physicians' fees are negotiated between government and organized medicine than if they are left to the whim of market forces, a market in which the for-profit chains would have the upper hand over individual physicians competing with each other. Collective autonomy would replace individual autonomy in both clinical decisionmaking and in physician reimbursement” (Colombotos and Kirchner, 1986).

This statement, of course, refutes the conservative ideology for a classical economic model of physician competition at the level of multiple small providers and purchasers. Instead, the authors make the ironic proposition that physicians will find greater remnants of their autonomy preserved by cooperation with government than with less benign powerful large payers who concentrate economic power against the profession.

Deficit Reduction Act of 1984

With hospital payment reform under its belt, Congress again turned its attention to physician payment. Developing a strategy for reform of physician payment, however, would prove to be far more difficult and would be years in the making.

In 1984 Medicare defrayed only 49 percent of the medical care costs incurred by the average beneficiary. This left substantial out-of-pocket expenses for premiums, coinsurance, charges by physicians in excess of Medicare payments, and uncovered services (drugs, long-term care).

In the Deficit Reduction Act of 1984, Congress imposed a freeze on physician fees and established the Participating Physician and Supplies Program (PAR), under which physicians could agree to accept assignment (the Medicare-approved charge as payment in full) on all claims. In return, they would be listed in a directory available to beneficiaries and would receive expedited claims processing. Moreover, they were permitted to raise their submitted charges during the freeze, which affected their charge profile in determining future payments but not those during the freeze.

In voluntarily accepting assignment, physicians gave up the ability under Part B to “balance bill” the patient for the full fee. This feature of the program conflicted directly with deeply held values of the medical profession regarding economic autonomy. The 1987 Report to Congress of the Physician Payment Review Commission (PPRC) notes that 80 percent of all physicians surveyed who initially refused to participate believe that physicians should have the right to set their own fees (Physician Payment Review Commission, 1987). The establishment of the PAR represents the first effort to move away, albeit by incentives, from physician control of their price or fee—a key element of economic autonomy. In deference to the historic autonomy claims of the profession, however, participation was strictly voluntary. As the program developed, participation rates increased steadily over the decade of the PAR program's existence. Whereas 30.6 percent of practitioners had signed participation agreements on January 1, 1987, this percentage had increased to 52.2 percent as of January 1, 1992 (Physician Payment Review Commission, 1992).

Despite restraints, Medicare outlays for physician services continued to outpace growth in the gross national product (GNP), national health expenditures, and Medicare Part A expenditures. Spending in Part B increased from $10.1 billion in 1980 to $21.3 billion by 1985 and $41.3 billion by 1990. Spending for physician services increased at a rate of 16.4 percent per year from 1980 to 1985 and 13.8 percent per year for 1985-90. Medicare physician costs per enrollee nearly doubled (in constant 1991 dollars) from $658 to $1,205 (Physician Payment Review Commission, 1992).

The importance of the clinical autonomy of the physician is evident in analysis of the factors responsible for the continuing rise in Medicare expenditures for physician services. During the 1980s, the change in the number and average age of Medicare beneficiaries accounted for only about 2 percent of annual Part B growth. From 1981 to 1986, increases in fees represented about 6 percent of total growth in expenditures per enrollee, and rising volume accounted for about 7 percent.

Clearly these policies preserving the clinical autonomy of practitioners had direct economic ramifications. Medicare spending on physician services from 1983 to 1986, the period during which fees were frozen, increased nearly 30 percent. Almost three-quarters of this growth was attributable to more services per beneficiary and changes in the mix of services (Physician Payment Review Commission, 1987).

It was increasingly evident that Medicare payment policies were contributing to the cost increases. The reliance on historical fee patterns resulted in a payment system of pricing that came to be considered irrational, confusing, and unfair. Over the years, wide payment differentials were perpetuated among types of procedures, specialties, geographic areas, and practice sites that could not be explained by differences in the costs of physicians' practices.

Two distortions were particularly noteworthy. First, because payments were based on past charges, two physicians providing identical services could receive markedly different payments. Second, the value of surgical and technical procedures became increasingly distorted relative to visits and consultations.

Consolidated Omnibus Budget Reconciliation Act of 1985

In the Consolidated Omnibus Budget Reconciliation Act (COBRA) of 1985, Congress began to take steps to realign the pattern of payments to physicians. Applying this concept of “inherent reasonableness,” it authorized the Secretary of Health and Human Services (HHS) to identify services for which Medicare-allowed charges were out of line with relative costs and to depart from the CPR methodology in adjusting payments for those services. In addition to providing a mechanism to change payments for selective services, COBRA created a framework for more comprehensive reform. The legislation directed the HHS Secretary to develop a resource-based relative value scale (RBRVS).

Congress also created the PPRC to advise on changes in the methods of paying physicians under Medicare. The creation of the PPRC signaled both the intention of Congress to reshape physician payment policy and the need for independent analytic support and policy advice. The commission began its work in the fall of 1986 and issued its first Report to Congress in the spring of 1987 (Physician Payment Review Commission, 1987).

Omnibus Budget Reconciliation Acts of 1986 and 1987

In the years after the establishment of the PPRC, Congress continued to squeeze physicians, particularly in the area of their economic autonomy. In the Omnibus Budget Reconciliation Act (OBRA) of 1986, Congress placed maximum allowable actual charge (MAAC) limits on the amounts non-participating physicians could bill above the Medicare-approved charge. The MAACs were only intended to be a transitional solution to controlling balance bills (bills in excess of Medicare-allowed amounts), but the establishment of charge limits set an important precedent for payment reform. Beginning with OBRA 1986, Congress began to take steps to both realign the pattern of relative payment and achieve budget savings by reducing prevailing charges for cataract surgery and anesthesia during cataract surgery.

The PPRC agreed with this approach and argued that Congress should move policy in the direction of longer term reform by reducing payments for overvalued procedures. It followed the principle of inherent reasonableness to identify 12 families of procedures it considered to be overvalued in relation to Medicare payments for other services. In OBRA 1987, Congress continued this pattern by reducing prevailing charges and imposing special limits on these services.

By 1988, the PPRC had endorsed the concept of replacing the CPR system with a fee schedule for Medicare. The study of an RBRVS, commissioned by HCFA and conducted under the direction of Professor William Hsaio of Harvard, was well under way, and Congress had begun to incrementally adjust relative payments and to strengthen beneficiary protection from balance billing (Hsaio et al., 1988).

In 1989, the PPRC submitted a set of proposals to Congress to rationalize the pattern of payments of physicians, to improve beneficiary financial protection, and to control program spending without diminishing access and quality of care. The cornerstone of the payment reform proposal was replacement of the system of payment of fees based upon usual, customary, and prevailing fee structures with a Medicare fee schedule based primarily on resource costs.

The commission recommended that the RBRVS be resource-based and composed of three elements: (1) physician work, reflecting the time and intensity of physician effort in providing a service; (2) practice expenses, including costs such as office rent, salaries, equipment, and supplies; and (3) a separate malpractice-expense component that reflects professional liability insurance premium expenses (Physician Payment Review Commission, 1989).

The RBRVS is translated into a fee schedule when multiplied by a dollar conversion factor. The PPRC recommended that the initial conversion factor be budget-neutral so that outlays for physician services projected under the fee schedule would be the same as those under the current system.

The second element of the PPRC proposal was a limit on charges for unassigned claims at a fixed percentage of the fee schedule amount. The charge limits would replace the physician-specific MAAC limits with a single limit applied to all physician services. This element of the package directly impinged on the economic autonomy of physicians by creating for the first time a fee limit for all physicians.

The third and most controversial piece of the PPRC package was its recommendation to base annual updates in the conversion factor on a comparison of actual increases in expenditures with a target rate of increase. The expenditure target (ET) would reflect projected increases resulting from inflation and growth and aging of the beneficiary population along with decisions concerning how much expenditure growth could exceed these factors to allow for increases in volume of services.

The ET proposal became the major obstacle to agreement. Not surprisingly, the AMA was strongly opposed to ETs. The American College of Surgeons, in contrast, supported this approach. It may be argued that this opposition was based on the possibility of infringement of Medicare into the clinical realm. In the face of this opposition from significant elements of the profession, Congress compromised and established a more complicated approach, called the volume performance standard (VPS).

Omnibus Budget Reconciliation Act of 1989

OBRA 1989 included the four components of the PPRC proposal: the Medicare Fee Schedule, charge limits, the VPS to determine updates in the conversion factor, and increased Federal support for clinical effectiveness research. The previously mentioned VPS implemented in 1990 “sets an annual volume target through Congressional action, or if the Congress does not act, through a default formula. The difference between this target and actual volume partly determines future physician rate updates, with low volume growth rewarded by higher updates” (Physician Payment Review Commission, 1994). Beginning in 1991, the newly established charge limits including limits on balance billing began to replace the MAACs, with full implementation in 1993. The Medicare Fee Schedule was fully implemented in 1996.

1996 and Beyond

Thirty years after the implementation of Medicare, physicians have found dramatic changes in their level of economic autonomy and, to a lesser extent, in their clinical autonomy. As noted in the discussion of the chronology of the program (and in Table 1), much of the activity of the Medicare program can be seen as reflecting a policy of observing the original congressional mandate of non-interference in the private practice of medicine with respect to its clinical dimension. Economic adjustments to the program have been quite subtle in their influence on clinical activity, and it may be argued that Medicare's processes of control of clinical utilization at the level of the individual practitioner have been quite limited—especially in contrast to the more heavy-handed utilization control methods of private insurers.

In the realm of economic autonomy, the picture is different. Following the initial attempts in the 1970s by policymakers to limit fee increases, Medicare has moved more directly to limit physician discretion in economic matters. The creation of the PAR Program in 1984 has led to the current limitation on balance billing that has effectively curtailed the potential for even non-participating physicians to exceed a mandated payment level in billing of Medicare beneficiaries. Although this may not appear as a major intrusion upon economic autonomy, the issue of balance billing has been an explosive one when viewed in other industrialized nations. The Ontario physicians' strike of 1986 provides a specific example of the volatility of this issue as perceived by physicians (Iglehart, 1986). Glaser (1989), whose cross-national work on physician payment policies has been widely recognized since the early 1970s, has boldly asserted that the decision to balance or extra-bill the patient beyond insured levels is “in every country … the most explosive issue between public authorities and medical profession.” Culbertson (1991) has suggested that “balance billing offers an ‘escape valve’ for the government and climate in which expenditure control is a consuming governmental objective.” With increased pressure on Medicare Part B to contain increases in expenditures for physician services, balance billing will emerge as a public policy debate between beneficiary advocates and the medical profession in the congressional consideration of medical savings accounts.

The move to the congressionally mandated Medicare Fee Schedule based on relative value units has effectively removed control of Medicare fees or prices from the hands of the medical profession. This factor, of course, violates Schulz and Harrison's (1986) definition of autonomy as inclusive of economic control by the profession. The remaining policy debate, encompassed in the heated controversy over expenditure targets and subsequent volume performance standards, centers on control of numbers of procedures performed. Economists have long debated whether physicians attempt to achieve a target income through performance of additional procedures when revenue is contained through price controls (Evans, 1984). As Evans explains this theory, “When average workloads and incomes fall, due to exogenous increases in supply, physicians change their practice patterns to increase utilization” (Evans, 1984). It is this that has made the notion of volume performance standards controversial for, on a macro level, it is suggested that physicians will lose clinical autonomy through overall programmatic budgetary limitations, which will have a detrimental effect on the clinical judgment of individual physicians.

What is debatable in this assertion is whether the economic consequences for an individual physician are sufficiently great to cause him or her to either inappropriately withhold service for fear of negatively impacting global budgets or to prescribe excessive services to make up for loss of marginal income. The experience of large physician groups such as the Permanente Group does not appear to support either of these assertions when risk is placed at the level of a larger entity such as the medical group rather than at the level of the individual physician. In its present form, it may be argued that the Medicare program and Medicare trust funds continue to be the ultimate holders of risk and therefore insulate individual physicians to some extent from their own decisions.

What of the future? It appears that much in the same way that financing and payment for health care services for individuals under 65 years of age is moving away from fee-for-service payment toward capitated managed care plans, Medicare may follow the same pattern. Indeed, some have argued that Medicare is the last bastion of fee-for-service medicine in the United States—a remarkable concession to its founders' commitment to the autonomy of physicians. If Medicare moves in this direction on a wider scale, it will in effect transfer risk from its general funds to the management of its contracting providers, as private insurers have done in the 1990s.

At the outset, Medicare permitted certain prepaid organizations to receive payment on a cost-reimbursement basis for their Medicare enrollees. In April 1985, before the advent of risk contracting, there were 916,000 Medicare enrollees in 109 plans receiving cost reimbursement (Health Care Financing Administration, 1996). A risk-sharing contract option for health maintenance organizations (HMOs) was instituted in 1972, 1 year before the Federal HMO act was passed. Progress was very slow at the outset. Greenlick has noted that this program began in 1978 as a demonstration at five original sites. It required 5 years, from 1982 to 1987, to enroll 1 million beneficiaries under Medicare risk contracts. In 1985 HCFA implemented changes enacted in 1982 to provide for a managed-care capitated payment option based on a prospective payment methodology. In the original risk-sharing contracts, growth continued to be slow. The second 1 million beneficiaries were enrolled from 1987 to 1991, and by 1995, the third million had entered into this arrangement (Greenlick, 1996).

Medicare HMO enrollment has increased steadily since risk contracting began in 1985. The number of beneficiaries in cost-reimbursement plans remained relatively steady from 1985 until 1996, when there was a significant decline (Health Care Financing Administration, 1996).

In 1996, nearly 9 percent of Medicare beneficiaries were enrolled in risk-contracting HMOs, and an additional 2 percent were enrolled in cost-reimbursed HMOs. In recent years, Medicare risk enrollment has grown rapidly (41 percent from December 1994 through January 1996). Enrollment in Medicare risk-contracting HMOs is particularly significant in California (36 percent), Oregon (34 percent), Arizona (31 percent), and Hawaii (31 percent) (Health Care Financing Administration, 1996).

It is not clear that managed care and at-risk payment for physician services will either limit or enhance physician autonomy. Certainly at this time, capitation is being widely touted as a means of preserving and, in some instances, expanding physician autonomy in both the clinical and economic arenas. Economically, physicians are presumed to gain autonomy under capitated arrangements through control and management of professional dollars. This control is further enhanced when physicians also control the distribution of hospital funds and are placed at risk for their expenditure as well (Sokolov, 1995). David DeValk encourages physicians to undertake capitation as it “engages the provider fully in the modification of ‘American medicine’; physicians are empowered to make decisions and changes (rather than dealing with bureaucratic hassles and 1-800-nurse-authorization lines)” (Medical Group Management Association, 1995). This is clearly a challenge to physicians to reassume clinical discretion that has arguably been lost to other organizations and indeed other professions.

The depth of emotion surrounding the clinical autonomy issue in the private sector is a result of the widely held perception in the medical community that insurers have dramatically eroded autonomy in pursuit of economic advantage. This has been accomplished through intrusive utilization controls and requirements for prospective authorization of procedures that exceed those traditionally associated with the Medicare program (Gray, 1991). These review activities, often involving other professionals in the review of physician judgments, have not been well received by the profession as having any significant impact on quality of care. Rather, the prevailing assumption among physicians is that motivation of these private organizations is purely economically driven.

Will physicians, given the opportunity to manage capitated premiums on behalf of beneficiaries, behave in a different manner? This is certainly the position of the leadership of much of the medical profession. It has been argued “that physician-led organizations delivering health care would avoid the stockholder-satisfying mentality of many for-profit insurance companies and, therefore, that physician-directed enterprises would direct more resources toward patient care and fewer to providing a return on stockholders' investments” (Goldfarb, 1995). However, findings from a study undertaken by Kerr et al. (1995) suggest that physicians may adopt behaviors that are equally detrimental to the exercise of clinical autonomy. The authors of the study conclude that “physicians are responding to capitation by using utilization management techniques, some at early stages of development, that were previously used only by insurers. This physician-initiated management approach represents a fundamental transformation in the practice of medicine” (Kerr et al., 1995). If economic judgments concerning the allocation of Medicare dollars currently exercised at a global level are placed at the level of smaller organizations, Freidson's principle of the primacy of peer review may or may not be distorted. The collegial professional group, which is the backbone of peer judgment in Freidson's typology of clinical autonomy, will then be forced to balance its clinical judgments with economic judgments when the use of limited resources is at stake.

Can clinical autonomy and economic autonomy be balanced and maintained in the future? Jones and Ethridge (1996) have argued that “operating in a rapidly changing insurance marketplace, Medicare is shifting from a social insurance model toward a private insurance model—expanding the number and type of alternative health plans it offers—and growing numbers of beneficiaries are enrolling in these plans.” If this is so, perhaps the historic commitment of the Medicare program envisioned by its founders to respect and reinforce the clinical autonomy of physicians will no longer be a relevant policy issue. Medicare was established in a political and economic climate in which the attainment of both clinical and economic autonomy for the medical profession was an economically realizable and socially supported policy objective. The test of the future will be to attain, as Reinhardt has suggested, the clinical objectives of the best in scientific achievements and traditions of the medical profession, while providing this care at an economic level that society as a whole can sustain.