The Urgent Need to Reform Medicare's Physician Payment System

Report Health Care Reform

The Urgent Need to Reform Medicare's Physician Payment System

December 5, 2006 19 min read Download Report
John O'Shea
John O'Shea
Former Senior Fellow, Health Policy
Dr. John O’Shea addressed the pressing physician payment and doctor practice issues in the health sector as senior fellow at Heritage.

The 5.1 percent reduction in Medicare payments scheduled to begin in January 2007 could lead many physicians to stop accepting new Medicare patients, to defer investments in new equipment and technol­ogy, or both. These problems are merely symptoms of a deeper pathology within Medicare and highlight the need for Congress to adopt comprehensive Medicare reform. Without congressional action to reform this profoundly flawed system, America's seniors could face a decline in access to high-quality health care.

Before Medicare was enacted in 1965, it was vigor­ously opposed by the American Medical Association (AMA). [1] Given the political realities, the final com­promise on the Medicare law concerning physician payment was designed to ensure that a maximum number of physicians would participate in the pro­gram and that America's seniors would have access to the highest-quality medical care.

These goals are now being undone. Following sev­eral unsuccessful attempts to reform Medicare's phy­sician payment methodology over the past four decades, access to quality health care for America's seniors is once again threatened. The current system of administrative pricing and price controls is anti­quated and irrational. Congress should replace it with a system that allows physicians and their patients to decide on proper treatment options based on the free-market principles of value, choice, and competition.

A New Policy

Congress should start reforming Medicare by re­placing the existing system of central planning and price controls that governs physician payments and almost every other provider reimbursement in the program. Specifically, Congress should:

  • Promote value in the delivery of medical ser­vices by introducing free-market forces. Con­gress should improve quality of care by pro­moting transparency of price and outcomes, allowing the market to function and thus re­warding superior physician performance and superior results in the treatment of disease. A number of positive policy changes to accom­plish this objective are embodied in the Medicare Physician Payment Reform and Quality Im­provement Act of 2006 (H.R. 5866), sponsored by Representative Michael C. Burgess (R-TX).
  • Reject bureaucratic schemes that increase Medicare's regulatory burden on doctors. Specifically, Congress should reject pay-for-performance proposals that would force physi­cians to comply with government guidelines in the practice of medicine. The medical profes­sion, not government officials, should set stan­dards for medical practice.
  • Start moving Medicare to a defined-contribu­tion system. Instead of trying to determine the "right prices" for thousands of medical treat­ments and procedures and enforcing them with thousands of pages of rules and regulations, Congress should start to move new retirees to an entirely new system based on defined contribu­tions and powered by the free-market principles of choice, competition, price transparency, and information that would drive improvements in quality. At the very least, Congress should start a new system of defined contributions for new retirees, allowing them to take the health plan of choice with them into retirement.

Evolution of the Medicare Physician Payment Structure

The initial Medicare physician payment struc­ture--known as the customary, prevailing, and rea­sonable (CPR) payment system--was implemented in 1966 without the threatened backlash from the medical profession. However, the system not only was complicated,[2] but also contained serious weaknesses that soon became evident. Because it was based initially on charges rather than on the value of the medical services, procedural services tended to be reimbursed at much higher levels than "evaluation and management" services. In addition, prevailing charges led to wide variations in reim­bursement rates for the same services from region to region. Critics often cited the CPR payment sys­tem as the major reason for the rapid rise in the cost of Medicare services because it created incentives for physicians to raise their prices.

By the mid-1970s, rising costs and the irrational­ity of the original Medicare physician payment sys­tem prompted congressional action. In an attempt to limit costs, prevailing fees were linked to the Medicare Economic Index (MEI), a measure that was supposed to reflect changes in the medical marketplace, including practice costs.

Not surprisingly, the resulting system did noth­ing to curb the increasing cost of medical services. The MEI methodology, implemented in 1975, lim­ited charge inflation but placed no control on the volume of services that physicians deliver. As Medi­care Part B expenditures continued to grow, physi­cians were once again blamed, this time for increasing the volume and intensity of services--an entirely predictable outcome given the incentives built into the payment system.

The 1989 Physician Payment Reform. Faced with the evident failings of Medicare's physician payment system, Congress embarked on a new ver­sion of administered pricing that was far more complex and cumbersome than anything previ­ously prescribed. In the Omnibus Budget Reconcil­iation Act (OBRA) of 1989, the House Committee on Ways and Means and the Senate Committee on Finance created a new system based on an even more rigorous model of central planning, resulting in the adoption of the resource-based relative value scale (RBRVS) for Medicare physician payments. In a dramatic reversal of the Reagan Administration's position on the proposed Medicare fee schedule, the George H. W. Bush Administration supported the change, along with limits on the right of physi­cians to balance-bill.[3]

The RBRVS, a complicated formula developed by Harvard economist William Hsiao, based Medi­care payments on a social science measurement of the time, energy, and effort involved in the work done by the physician and the associated practice and malpractice costs. It attempted to link physi­cian payment to the resources, or "inputs," that were used in providing medical services. However, the formula did not adequately account for input in terms of physician skill or expertise.

"Outputs," or results, were another matter. The fee schedule also failed to assess either the value of a medical service or its benefit to Medicare patients. Moreover, the methodology of computing the value of work units used in the formula has been subject to serious question.[4]

Characteristically, complex systems based on central planning tend to resolve one problem in a fashion that invariably generates new ones. The RBRVS system was a way to ratchet down fees for individual medical services, but it left in place the powerful incentives for physicians to increase their volume of services and drive the aggregate level of Medicare Part B spending even higher. Congress anticipated this development and included yet another administrative "fix" by limiting the volume of Medicare services under Part B.

In an attempt to control total spending for phy­sicians' services driven by volume increases, OBRA also tied the annual update of the fee schedule to the trend in total spending for physicians' services relative to a target that was based on historical trends in volume. This method, effective in 1992, became known as the Medicare Volume Perfor­mance Standard (VPS). If overall volume went up, physician reimbursement was adjusted downward.

Of course, the obvious problem with this scheme was that aggregate volume increases do not give physicians an incentive to curb their own appetite to increase the volume of their individual practices. Not surprisingly, adoption of this administrative fix led to unstable and unpredictable physician pay­ment updates. Congress responded with the Bal­anced Budget Act of 1997, which replaced VPS with the sustainable growth rate (SGR) mechanism that governs physician payment updates today.[5]

From VPS to SGR. Unlike the VPS, the SGR tar­get is tied to growth in the nation's gross domestic product per capita and adjusts physician payments by a factor that reflects cumulative spending rela­tive to the target. The fundamental problem with this method is that the vagaries of the general econ­omy may have no relation whatsoever to the cost of providing services to Medicare beneficiaries. The SGR has done little to reduce volatility in the system, does not control costs, and--if left unfixed--prom­ises to jeopardize Medicare beneficiaries' access to care by discouraging physicians from accepting new Medicare patients and by driving physicians out of the program altogether.

How Administrative Pricing Affects Doctors and Patients

Physicians face a 5.1 percent cut in Medicare payments beginning in January 2007, with cumu­lative cuts projected to reach 34 percent by 2015. Practice expenses are expected to increase by 15 percent to 20 percent during the same period. (See Chart 1.)

Chart 1

The real effect of Medicare fee cuts on physi­cians needs to be understood in terms of the over­all medical practice environment. In addition to the cuts in Medicare payments, physicians face declining reimbursements from all payers, public and private alike. For example, the fee schedule for TRICARE--the health care system for military personnel, retirees, and their families--is linked directly to the Medicare fee schedule, and many private payers follow Medicare's lead in setting their fees. In addition, the growing number of uninsured puts a disproportionate burden on physicians, who provide at least 18 percent of uncompensated care to the uninsured and, unlike hospitals and community health centers, are not compensated for their contribution to the health care safety net.[6]

The net effect of these economic forces has been a significant decline in physician income. According to a recent study by the Center for Study­ing Health System Change, the average physician's real (inflation-adjusted) net income from practicing medicine declined 7 percent between 1995 and 2003. This was in contrast to income trends for other professionals, who experienced an increase of about 7 percent during the same period.[7] The constraints on physicians' time have also affected their ability to respond to increases in patient demand, and the need to comply with thousands of pages of Medicare regulations under threat of criminal prosecution only adds to the problem.[8]

Patient Access to Care. Although Medicare beneficiaries are not currently reporting major problems finding a doctor, a closer look at the data suggests that access to health care for America's seniors is unstable and that projec­tions for the future are even more ominous. From 1996-1997 to 2004-2005, the percentage of phy­sicians accepting all Medicare patients declined only slightly, from 74.6 percent to 72.9 percent, but the percentage of surgical specialists accepting all Medicare patients dropped significantly, from 81.5 percent to 73.1 percent. Interestingly, physicians with the lowest incomes actually increased their acceptance of Medicare patients from 65.2 percent in 2000-2001 to 72.2 percent in 2004-2005.[9] Some analysts suggest that this was to offset stag­nant revenues from private payers.

A recent AMA physician survey reported that 45 percent of physicians will either decrease or stop seeing new Medicare patients if the projected 5.1 percent cuts go into effect in January 2007. Faced with the projected 34 percent cuts by 2015, 82 per­cent of physicians say that they will need to make significant changes in their practices that will affect access to care. More than half of physicians also say that they will forgo investments in their practices, including deferring purchase of new medical equipment and information technology (IT).

This finding is troubling, particularly since offi­cials in Congress and the Administration are advo­cating increased use of information technology, such as electronic medical records, as a way to improve patient safety and quality of care. How­ever, adopting, installing, and using health IT sys­tems could cost as much as $44,000 per provider initially and $8,500 per provider per year, an investment that many physicians are unwilling to make given the uncertainty about future income.[10]

In addition to complaints from the medical pro­fession, the Medicare Payment Advisory Commis­sion in its 2006 report to Congress called the SGR formula a "flawed, inequitable system for volume control" and expressed its concern that "such con­secutive annual cuts would threaten access to phy­sician services over time, particularly primary care services."[11] In other words, the emerging situation would be disastrous for Medicare patients.

Serious Medicare Reform

Physicians, like every other class of profession­als, respond to incentives. Reversing four decades of misguided administration of the Medicare pay­ment system will not happen overnight. Physicians practicing medicine today should not be held accountable for congressional mismanagement and micromanagement of Medicare. Physicians are also not responsible for the financial problems that Congress has imposed on Medicare, including the massive entitlement expansion created by the pre­scription drug benefit, adding an estimated $8 tril­lion of unfunded liabilities in the form of benefits promised but not financed in the program.

Nor should physicians be held responsible for Medicare spending that is beyond their control. Medicare is providing seniors with modern health care services, and much of the increase in spending attributed to physician services is the result of the development of better and more expensive tech­nology and imaging procedures. (See Chart 2.)

Chart 2

Current Proposals

Much recent legislative action regarding Medi­care physician payment has taken the form of averting impending fee cuts. However, in July 2006, Representative Burgess introduced the Medi­care Physician Payment Reform and Quality Improvement Act of 2006 (H.R. 5866). This bill goes beyond a mere moratorium on the scheduled fee cuts and recommends terminating the SGR and replacing it with a single conversion factor: the MEI minus 1 percent. In addition, it proposes putting in place by 2009 a system of evidenced-based quality measures developed by the physician specialty organizations. Under this legislation, physicians would be allowed to balance-bill certain patients above normal charges as limited by current law as long as the physician participates in the quality reporting program.

The bill takes the right approach in authorizing balance billing because it allows market forces to operate in a way that gives physicians an incentive to provide medical services that are of the highest value. It would also motivate patients to demand transparency about price and quality. The legisla­tion correctly specifies that quality measures should be developed by the individual specialty groups. Basing payment on measures imposed by an administrative agency is unlikely to produce any meaningful gain in quality for the money spent.

Although the MEI is theoretically more sensitive to the medical marketplace, its role as an indicator of the true cost of services is questionable. Con­tinuing to link the Medicare fee schedule to the MEI would not address the basic financing struc­ture that has plagued Medicare since its inception. Without additional major structural reforms, these limited modifications would likely result in another round of increased spending followed by a call for further cuts.

Congress cannot make up for its previous mis­calculations or policy decisions by simply cutting Medicare payments to physicians. Avoiding cuts in the short run is not a substitute for real Medicare reform, and this dalliance will only exacerbate the problem. Medicare can no longer func­tion with a payment system that is linked to external measures that bear little or no relation to the true cost or effectiveness of providing medical care. Physician payment needs to be based on free-market competition that pro­vides the proper incentives for quality, efficiency, and transparency.

Step 1: Congress should start moving Medicare to a defined-con­tribution system.

The present payment structure, which defines benefits, provides no financial incentives for patients to learn more about the value of their health care and no incentives for physicians to be transparent or to compete on price and quality.

Government should define the contributions to be used for health care and let patients and health care professionals decide on the benefits, treatments, and medical procedures. If Congress is unwilling or unable to overhaul the entire Medicare pro­gram, it can at least start giving new retirees a defined contribution to the health plan of their choice, enabling them to take their private health plans into retirement with them.

Congressional concern about Medicare spending in the aggregate is understandable, but the deci­sions that actually affect the cost and quality of ser­vices are made during individual encounters between patients and their doctors in which dis­cussions of macroeconomics have no place. Given the proper incentives and ownership of their health care resources, patients will become knowledge­able about the value of their care. Millions of Amer­icans, including seniors, make informed decisions every day about goods and services in the economy, from financial planners and lawyers to computers and cars. They can do the same with medical ser­vices. By avoiding physicians who are not transpar­ent about quality and cost, they will drive providers to make such information available.

Step 2: Physicians should be paid for the real value of their services, not for bureaucratically defined "performance."

Physicians should be paid for the value of their services, not for their ability to perform processes that may be unrelated to patient outcomes. Current pay-for-performance (P4P) proposals and pro­grams have significant flaws and should not be considered a substitute for real quality improve­ment of the health care system.[12]

P4P advocates do not even agree on whether or not to pay for a certain level of performance or improvement above a baseline level. For example, under a quality measure such as screening women for cervical cancer, should initially poor performers that raise their screening rate from 40 percent to 80 percent be paid more than a provider that improves from 80 percent to 85 percent? Furthermore, does the increase from 40 percent to 80 percent repre­sent better medicine or better documentation by the provider? Finally, should quality measures be enforced by withholding payments from poor per­formers or by increasing payments to good per­formers, or both?

A recent review of the early experience with pay for performance reported several interesting find­ings: Most P4P programs currently under way suf­fer from significant design flaws, some measures are more difficult to affect than others, and choos­ing only a subset of measures is an inadequate eval­uation strategy.[13] Another recent study found that paying clinicians to reach a common, fixed perfor­mance target may produce little gain in quality for the money spent and will largely reward those cli­nicians who show the greatest improvement in reported performance, which raises the issue of whether better medicine or simply better docu­mentation is being rewarded.[14]

A good example of an effective quality improve­ment initiative is the National Surgical Quality Improvement Program (NSQIP) developed within another government program. During the mid-to-late 1980s, the Department of Veterans Affairs (VA) came under a great deal of public scrutiny over the quality of surgical care in the 133 VA hospitals. At issue were the operative mortality rates in the VA hospitals and the perception in Congress that these rates were significantly above the national (private-sector) norm.

To address the gap, Congress mandated that the VA report its surgical outcomes annually. In the process, the VA developed a database of preopera­tive risk factors and postoperative outcomes, allow­ing meaningful measurement of the quality of surgical care. From 1991 to 2001, through feed­back to participating hospitals and providers, the VA hospitals saw a 27 percent decline in post-oper­ative mortality and a 45 percent drop in post-oper­ative morbidity. Median post-operative length of stay fell from nine days to four days, and patient satisfaction improved. This program was then rolled out to non-federal institutions and continues to grow in membership and effectiveness.

Significantly, this program was not a quick fix. Surgical specialists took more than a decade to develop it.[15]

Step 3: Congress should junk Medicare's ex­cessive regulation.

Physicians face not only cuts in fees, but also a number of other burdensome Medicare regulations. Physicians should be given the choice to set their own prices, including the ability to balance-bill patients for charges not covered by Medicare, and let market forces decide whether the price is correct.

Competition based on value will set prices much more rationally than pricing systems administered by the government. Competition will also be a more realistic and meaningful driver of long-term quality improvement initiatives than the proposed pay-for-performance schemes. Once patients con­trol their own medical care, they will insist on transparency in price and quality, and physicians will have the proper incentives to improve the effectiveness and efficiency of their care. Physicians will also be more likely to invest in information technology and other improvements in patient care if they are not restrained by the threat of future income reductions.

Just as regulation has failed to control spending, it is unlikely to improve quality. Physicians have always had incentives through a number of mech­anisms to provide quality care to patients. Most important is their personal and professional duty to alleviate suffering. In addition, physicians need to maintain their reputations, both among peers and in their communities. Even though the medical tort system is flawed in many states, the threat of mal­practice litigation provides an additional incentive to maintain quality.

The problem with these mechanisms is that they are difficult to measure and do not provide the transparency needed in the current system. As information collection methods improve, meaning­ful data will become widely available, and lasting quality improvement will be incorporated into the system. Regulation in the form of price controls or piecemeal quality mandates will only impede this process.


Congress should move quickly to reform the Medicare physician payment system. Simply avoid­ing impending cuts or tweaking the current admin­istrative payment system is insufficient. Because physician payment is inseparable from the prob­lems of Medicare itself, the reform of physician payment should be accompanied by policy changes that start to restructure Medicare.

Specifically, Congress should move toward a defined-contribution system in Medicare, at least for new retirees, allowing them to take private plans or health options with them into retirement. Over time, this change would take Medicare com­pletely out of the business of fixing prices. Second, Congress should reject any regulatory scheme that increases the already heavy regulatory burdens on physicians, including pay-for-performance schemes that would likely be counterproductive. Finally, Congress should promote value through price transparency, information, and intensified competi­tion among doctors and other medical professionals.

John S. O'Shea, M.D., is Health Policy Fellow in the Center for Health Policy Studies at The Heritage Foundation.




[1]At the AMA national convention in 1965, nine state delegations introduced proposals to boycott the new system.

[2]In the CPR system, physicians were paid the lowest of three possible fees: the actual charge submitted, the fee customarily charged by a particular physician, or the prevailing fee charged by physicians in a given locality.

[3]Balance billing allows physicians to charge patients directly for the portion of the bill not covered by Medicare.

[4]For an overview of the policy and politics surrounding the adoption of the RBRVS, see Robert Emmet Moffit, "Back to the Future: Medicare's Resurrection of the Labor Theory of Value," Regulation, Vol. 15, No. 4 (Fall 1992), pp. 54-63, at (November 20, 2006), and Physician Payment Review Commission, 1991 Annual Report to Congress.

[5]For a general description of the SGR formula and the congressional debate over the pending 5.4 percent Medicare payment cut, see Robert E. Moffit, Ph.D., "Why Doctors Are Abandoning Medicare and What Should Be Done About It," Heritage Founda­tion Backgrounder No. 1539, April 22, 2002, at (November 20, 2006).

[6]Jack Hadley, Ph.D., and John Holahan, Ph.D., "The Cost of Care for the Uninsured: What Do We Spend, Who Pays, and What Would Full Coverage Add to Medical Spending?" Kaiser Commission on Medicaid and the Uninsured Issue Update, May 10, 2004, p. 3, at /static/reportimages/86CACE57D9FCE86AC2A4809BA9997426.pdf (November 20, 2006).

[7]Ha T. Tu and Paul B. Ginsburg, "Losing Ground: Physician Income, 1995-2003," Center for Studying Health System Change Tracking Report No. 15, June 2006, at (November 20, 2006).

[8]Sally Trude, "So Much to Do, So Little Time: Physician Capacity Constraints, 1997-2001," Center for Studying Health System Change Tracking Report No. 8, May 2003, at /static/reportimages/C722A5265FDDFE93306E2AC8DA5C6055.pdf (November 20, 2006).

[9]Peter J. Cunningham, Andrea Staiti, and Paul B. Ginsburg, "Physician Acceptance of New Medicare Patients Stabilizes in 2004-05," Center for Studying Health System Change Tracking Report No. 12, January 2006, at CONTENT/811/811.pdf (November 20, 2006).

[10]American Medical Association, "2006 AMA Member Connect Physician Survey: Physicians' Reactions to the Projected Medicare Payment Cuts," at (November 20, 2006).

[11]Medicare Payment Advisory Commission, Report to the Congress: Medicare Payment Policy, March 2006, at publications/congressional_reports/Mar06_EntireReport.pdf (November 20, 2006).

[12]For a broad discussion of the Medicare "pay for performance" proposals, see Richard Dolinar, M.D., and S. Luke Leininger, "Pay for Performance or Compliance? A Second Opinion on Medicare Reimbursement," Heritage Foundation Backgrounder No. 1882, October 5, 2005, at

[13]Mark R. Chassin, "Does Paying for Performance Improve the Quality of Health Care?" Medical Care Research and Review, Vol. 63, No. 1 Supplement (February 2006), pp. 122S-125S, at /static/reportimages/BBACDC97A5CA6F2E35FD418095A1EDA0.pdf (November 20, 2006).

[14]Meredith B. Rosenthal, Richard G. Frank, Zhonge Li, and Arnold M. Epstein, "Early Experience with Pay-for-Performance: From Concept to Practice," JAMA, Vol. 294, No. 14 (October 12, 2005), pp. 1788-1793.

[15]See American College of Surgeons, "National Surgical Quality Improvement Program," Web site, at (November 20, 2006).




John O'Shea
John O'Shea

Former Senior Fellow, Health Policy