September 21, 2014

The Stubborn Rise in Medicare Physician and Contractor Expenditures

By David Koitz

This is Issue #4 in The Concord Coalition Series on Health Care and Medicare. For the rest of the series, click here.

Two prominent areas where Medicare has had little success in controlling costs are payments to physicians and contracts with private group health care providers. With combined expenditures of more than $170 billion, these two components of Medicare will account for one-third of the program’s total expenditures in 2009.[1] Over the period from 1997-2005, payments for care provided by physicians under the traditional fee-for-service portion of Medicare (applicable to about 75 percent of the program’s recipients) grew almost three times faster per enrollee than all other covered services. This year they will reach an estimated $62 billion. [2] Reflecting the impact of that rise, the standard enrollee premium for Part B of Medicare, most of which helps fund physician services, rose from $50 a month in 2000 to $96.40 a month in 2009.[3] In the aggregate over the 1968 to 2005 period, Medicare physician spending grew at an average annual rate of 8.9 percent. In contrast, inflation averaged 4.1 percent.[4]

Medicare coverage provided by health care companies that have contracted with the government to enroll and serve recipients in private group or managed care plans has also experienced significant growth. Over the years, per-participant costs under those plans have routinely exceeded those of Medicare recipients who have remained in the traditional fee-for-service program. Managed care plans have participated in Medicare since the 1970s when HMOs were allowed to provide a broad range of services to program recipients in exchange for a monthly per-person charge paid by Medicare, or what is referred to as a capitation payment. Other forms of group arrangements were authorized in 1997 under what was labeled the "Medicare + Choice" program, and further expansions were made under the current "Medicare Advantage" program authorized in 2003. The latter expanded the types of plans that could participate, altered the bidding process to be followed by contractors, and increased payments to them to increase provider and enrollee participation. These expenditures are estimated to reach $110 billion in 2009, and account for 25 percent of aggregate Medicare spending.

Physician Payments

When Medicare began, its payments to physicians were derived from the fees they charged, mirroring the practice then of private insurers. While not always covering 100 percent of the submitted bill, under the law, any amount not reimbursed by the program could be recovered from the patient--what was referred to as “balance billing.” Collectively, the two forms of payment created an incentive for doctors to increase their fees routinely, a condition that led to Medicare’s physician expenditures growing at an average annual rate of 13 percent from 1967 through 1974. Attempting to limit that growth, lawmakers mandated that annual fee increases not exceed an economic index that tracked the inflation faced by physicians and general wage levels--what is referred to as the Medicare Economic Index (or MEI).[5]

Limiting increases to the MEI, however, was not enough, and further constraints were enacted in 1984. Fees were frozen from 1984 through 1986, and for the period from 1987 through 1991, specified increases were set in law. But even those actions proved ineffective, largely because limits on fees alone did little to control the volume of services physicians provided (i.e., the quantity and type). By 1990, Medicare's total payments to physicians were more than three-and-a-half times greater than they had been 10 years earlier, and the average physician was receiving more than two-and-a-half times as much in Medicare payments. For the entire period from 1975 to 1991, Medicare physician spending grew at an average rate of 15 percent annually.[6]

Reacting again to the unchecked growth, Congress authorized a new fee schedule in 1992 that based physician payments on measures of the relative resources used to provide each service. It now applies to more than 7,000 forms of services.[7] The formula for each fee had two parts: one was a weight--or "relative value"--that indicated the resource costs of each service relative to all other services (a CAT scan, for instance, was given a higher relative value than a routine office visit). The other part was a fixed dollar amount known as the conversion factor, which would be multiplied by each relative weight to calculate the fee to be paid for each service.

By itself, this new schedule was not designed to control Medicare's spending. It merely redistributed spending among physicians' services as a means to promote equity. Its purpose was to correct for what previously were considered to be under- and over-priced services, and by so doing, it would alter the propensity of physicians to supply services based on how much Medicare paid for them. What did control spending was a mechanism that tied the annual update of the fees to an overall spending target. Under this approach, the conversion factor was updated annually to reflect increases in physicians' costs and then adjusted by another factor to counteract changes in the volume of services provided. This second adjustment, known as the volume performance standard (VPS), was an attempt to keep total spending for physicians' services within budgetary targets.

While it worked, the VPS system led to updates that were volatile. The adjustments depended heavily on the historical-volume trend, and a decline in that trend in the mid-1990s led to large increases in Medicare's payments. Congress then attempted to offset those effects by making successively larger reductions to the updates. Between 1992 and 1998 the annual fee increases ranged from a low of 0.6 percent to a high of 7.5 percent. These actions clearly had a restraining impact--Medicare's spending for physician services grew at an average annual rate of only 3.2 percent during the 1992-1998 period--but the changes in overall physician spending varied substantially from one year to the next, ranging from a reduction of 2.6 percent in 1992 to increases of almost 10 percent in both 1994 and 1995. That volatility led Congress to modify the VPS in 1997, replacing it with what was labeled a “sustainable growth rate” (SGR) adjustment.

Like the VPS, the SGR method uses a spending target to adjust future physician payment rates and to control growth in expenditures. In contrast to the VPS, however, the SGR target is tied to growth in the nation’s per capita Gross Domestic Product (GDP)--in effect, it’s tied to the broadest measure of the nation’s economic growth--a measure that is considerably less volatile and generally smaller than changes in the volume of physicians' services.

What is important and notably different about the SGR mechanism is that it has a self-correcting adjustment feature. If cumulative spending for physician services over a period of years deviates from the expenditure targets--i.e., if growth in Medicare physician spending is faster or slower than the change in the per person growth of the economy--the annual updates to the payment rates will eventually be adjusted so that over time spending will be brought back into line with the cumulative target.[8] In effect, if cumulative spending exceeds the cumulative target (as it currently does), the SGR mechanism is suppose to reduce payment rates each year until spending in the most recent year is below the expenditure target for the year. “At that point, the updates to payment rates may become positive, but the increases are to be set to keep annual spending below the year-by-year expenditure targets until cumulative spending and the cumulative target converge.”[9]

The problem with the SGR system is that since 2001, Congress--confronted by heavy lobbying--has repeatedly overridden the reductions the SGR mechanism would have put in place, and thereby allowed cumulative spending to exceed the targets. Consequently, while holding physician payments to the growth of the economy in general might have been conceived as an effective mechanism to help rein in Medicare spending, and still could be, the ad hoc fashion in which Congress has interceded has made it a cost control mechanism is name only. Over the period from 1998 through 2005, total physician spending grew at an average of 7.4 percent annually, well above the growth of GDP.[10] The Congressional Budget Office (CBO) compared the cumulative amount of that rise over the 1997 to 2005 period--showing that per enrollee spending on physician services rose by 65 percent in contrast to a 35 percent per enrollee rise for the rest of Medicare’s fee-for-service expenditures.[11]

The political complexity of achieving real restraint with physician payments is illustrated by testimony given by CBO in 2004--

Spending for physicians' services subject to the SGR mechanism has grown at an average rate of about 6 percent a year since the 1996/1997 base year… By the end of 2002, such spending had exceeded the cumulative target by about $17 billion… in the next few years, expenditures in excess of the target would have grown by another $10 billion. As a result, physician payment rates for 2003 were scheduled to drop by 4.4 percent (after falling by 5.4 percent in 2002). In the Consolidated Appropriations Resolution, 2003… Congress responded to that imminent reduction by allowing the Administration to boost the cumulative target, thereby producing a 1.6 percent increase in payment rates…

Through 2003, that spending exceeded the higher target by about $6 billion… If it had been allowed to operate, the SGR method would have reduced payment rates again, this time for 2004. However, the Medicare Modernization Act replaced that scheduled reduction in rates with increases of 1.5 percent in both 2004 and 2005—but it left the cumulative target intact, specifying that those increases were not to be considered changes in law or regulation for the purpose of adjusting the expenditure target. Thus, spending for physicians' services will continue to exceed the cumulative target…[12]

But leaving the SGR system in place then was mostly pro forma. Little has changed over the past four years in Congress’ ad hoc fashioning of physician fee increases. The Deficit Reduction Act of 2005 held 2006 payment rates at their 2005 level, overriding an impending reduction of 4.4 percent. The Tax Relief and Health Care Act of 2006 raised the physician fee schedule for 2007, increased the target for one year, and specified that the 2008 physician fee schedule conversion factor be computed as if the 2007 physician fee schedule update had not been changed. Again, a pro forma move. Similarly, the Medicare, Medicaid, and SCHIP Extension Act of 2007 raised the physician fee schedule update for the first six months of 2008, increased the target for half a year, and again specified that the physician fee schedule be computed in the second half of 2008 and 2009 as if the update had not changed. But then the Medicare Improvements for Patients and Providers Act of 2008 canceled the reduction that went into effect on July 1, 2008, and increased payment rates for 2009.

Overall, Congress has overridden the SGR reductions in seven of the past eight years. The summation of those actions has basically been to increase physician payment rates with little or no control for the volume of services or other measures physicians employ that have historically contributed to Medicare’s expenditures outstripping the growth of the economy. In its budget outlook for the 2008 to 2018 period issued in January 2008, CBO estimated that if the SGR mechanism were allowed to work as scheduled, it would have reduced payment rates for physicians’ services by about 10 percent in July 2008 and by about 5 percent annually through 2018. At the end of that time, cumulative physician spending would be nearly back in line with the formula’s cumulative targets, but payment rates for physicians in 2018 would be less than three-quarters of what they were in 2008.[13]

These incremental and temporary adjustments have resulted in a growing distortion of projected Medicare costs. The Medicare actuaries recently estimated that “Because actual physician-related spending has exceeded the target spending levels by progressively larger annual amounts since 2001, cumulative actual spending is greater than the cumulative target amount by about $64 billion after 2008, and by a projected $70 billion after 2009.”[14] When CBO or the Medicare Trustees forecast future expenditures they do so based on “current law,” meaning that they assume the SGR targets will be enforced. This assumption results an improbable view of projected spending since phantom savings are built in. As stated by the Trustees’ in their 2009 report on Medicare’s finances, “It is important to note that projected…expenditures are significantly understated because future reductions in physician payment rates, required under current law, are unrealistic and very likely to be overridden by Congress.”[15]

CBO recently estimated in its analysis of the Adminstration’s budget that the SGR mechanism would require a 21 percent reduction in physician payments in 2010, and a six percent annual reduction to follow through 2019. Given the duration and dimensions of those cumulative adjustments, the pressure from interest groups representing providers and enrollees would be intense, and there is little history to suggest that Congress will acquiesce to allowing the SGR mechanism to work in the years ahead. Thus, the probability is great that the SGR mechanism too will soon be relegated to the graveyard of Medicare’s attempts to control physician expenditures. Indeed, President Obama’s budget is based on the assumption that the SGR formula will not be enforced. According to CBO, his proposal to freeze rates at their 2009 level would increase spending relative to the SGR level by $285 billion through 2019. Allowing rates to rise unrestrained (i.e., keeping pace with increases in the MEI) would cost over $400 billion in CBO’s estimation.[16]

What is also apparent from the history of physician spending is that it is not entirely clear how alterations of fee schedules and payment practices affect program costs. There is a great deal of evidence that to the extent altering them results in a large impact on physician incomes, physicians will react by changing the volume of services provided, such as to increase them if their incomes drop or stagnate. Other evidence suggests, however, that setting fee levels to control costs will alter the amount of aggregate payments to physicians, with more services being provided with higher prices and less with lower prices. The latter findings suggest that while there may be some offsetting change in the volume of services to compensate for lower fees, setting fee levels for the purpose of lowering payments would achieve program savings.[17]

While that debate and Congress’ reticence to take action raises uncertainty about the effectiveness and viability of imposing physician payment constraints, what is clear is that striking a balance between continuing enrollee access to physician care and controlling the public cost of physician services will require new thinking. At least in recent years, policymakers have been reluctant to test the premise that limiting payments will not materially affect physicians accepting Medicare payments for services. Physicians do not have to participate in Medicare, and the specter of physicians dropping out has made lawmakers hesitant to press them. However, since one out of every five dollars spent nationally for physician care comes from Medicare, the degree to which physicians would flee the program remains an open question. That threat, for instance, might be mitigated by the inability to achieve the same income from patient out-of-pocket payments and other insurers. Moreover, to the extent major shifting did occur, it would result in higher premiums for private insurance and greater out-of-pocket expenditures from all who are insured. Repeatedly over the years, Congress has shown little patience with billing practices that attempt to recover large fees from the Medicare population. Ultimately, as cost pressures on the Medicare program intensify in the coming decade, a large exodus of physicians from the program could lead lawmakers to impose even greater constraints on physician charges, either through tightening Medicare payment rules as they’ve done in the past or even imposing limits on physician charges in general.[18]

Contractor Spending

As an alternative to its traditional fee-for-service system of payment, Medicare has taken a variety of steps over the years to encourage the use of private health care entities to provide and manage a broad range of services to enrollees who elect to participate in their plans. These include HMOs, PPOs, and other private group providers. When added to the program as an option for enrollees, the expectation was that this approach would result in better coordination of services and treatment, and allow enrollees and the program to benefit from efficiencies that are harder to achieve with the existing fee-for-service program--most notably among HMO and PPO providers. Here too, however, Medicare has had little success with cost containment. A recent review of those efforts concludes--

Twenty-five years of experience with private plan contracts has shown no significant cost savings. In fact, studies show Medicare has paid private plans more than their costs and more than it would have paid for beneficiaries had they remained in the traditional Medicare program. Where private plans have excelled is in delivering enhanced benefits and lower out-of-pocket costs for beneficiaries compared with what they would get under traditional Medicare.[19]

Medicare contractors—operating under what is called the Medicare Advantage program (also known as Part C of Medicare)--will serve an estimated one-fourth of the Medicare population in 2009.[20] The program’s financing of these arrangements centers around a series of per-person spending benchmarks that are calculated on a county-by-county basis. Contractors then submit bids showing how much they want Medicare to pay per enrollee, and the government compares those bids to the benchmarks. The benchmarks are set so that they represent at least as much per person as the program spends for enrollees who stay under the traditional Medicare program, and the approved contractors are paid their bids up to the benchmark plus 75 percent of the amount by which the benchmark may exceed their bids. If there is such an excess, they are required to return it to their participants in the form of additional benefits, such as reduced cost sharing on services or as rebates for premiums enrollees pay for Part B of Medicare or prescription drug benefits (Part D). More than 500 companies were contracted under the Advantage program in 2008.

In many counties, the program benchmarks have turned out to be consistently higher than the per person costs of those who stay in the traditional Medicare program. For instance, for 2007, CBO calculated that the benchmarks would be 17 percent higher than the national average of Medicare’s per capita fee-for-service costs, and that the program’s per enrollee spending under the Advantage program would be about 12 percent higher.[21] As a result, enrollment in private group plans has been growing rapidly as the Medicare population recognizes their broader range of services or lower premiums for Part B or prescription drug coverage. While HMOs and PPOs enroll the greater number of group-plan participants, the fastest growing segment are private fee-for-service plans that often provide greater benefits than the traditional Medicare fee-for-service program. Overall group-plan enrollment is projected to rise from 21 percent of the Medicare population in 2008 to 26 percent in 2010, with most of the increase attributable to those seeking group coverage from fee-for-service entities.[22]

While the Advantage program plans offer enrollees a wider choice of coverage that may better suit their medical needs than the traditional program, and a greater range of potential services, their higher overall cost has become a major source of concern. Increasingly, reducing the resources used to support them--notably for the burgeoning fee-for-service plans--are being seen as a potential means of funding other federal benefits or coverage (e.g., coverage of the uninsured), and as well to help reduce Medicare’s future financing strains.

Conclusion

There is a propensity among many who have studied the nation’s health care problems to throw up their hands and urge policymakers to completely revamp the system… an entreaty of sorts for a silver bullet. The problem they see is a health care closet so full of issues it’s hard to decide where to start. However, to really solve our health care problems, we have to tackle them at their roots. The two issues discussed in this report are difficult political issues. They’re Medicare issues, but they reflect the larger problem of where to begin health care reform. Unwillingness to address them now is unwillingness to engage in real change under whatever umbrella we chose to label as “reform.” Why, for instance, would physicians be any more willing to accept less compensation under a new nationalized program or a single-payer system? And while the Medicare Advantage program may have been conceived in part as a way to save money by instilling competition into the program, it hasn’t succeeded. It should be little wonder that offering higher benefits and lower premiums would make it popular, but it is now working against the goal of constraining future spending. Indeed, the growing cost of Advantage plans has become a serious budgetary concern, adding to Medicare’s precarious fiscal outlook.

To effect real change, we have to be willing to address the problems the current health care system has created. And they have to be taken on one by one, issue by issue. The easier course is always to address the least contentious ones first, but if the closet is to be better organized, the tough issues can’t be suppressed. Whether we embark on creation of an entirely new system or decide we can fix what we have, success can only be achieved by wading into the swamp. Medicare’s problems are tough and the potential remedies are conspicuous--45 million constituents makes for a large and attentive audience. But if for no other reason, the program’s looming insolvency--now projected to be only eight years away--demands that we confront them head on.


[1] CBO, “Supplemental Data” to the 2009-2019 Budget Baseline, Medicare baseline, March 24, 2009.

[2] Excludes physician services provided under group contract arrangements (e.g., HMOs and the like funded under the Medicare Advantage program), ancillary facilities, and outpatient hospital settings.

[3] In addition to the standard premium, some higher-income enrollees now have to pay an income-related supplemental charge. These higher premiums are being phased-in over three years, which started in 2007. About 5 percent of current enrollees will pay them this year. Overall, Part B premiums from all enrollees account for about 25 percent of the financing of Part B expenditures, with the other 75 percent coming from the government’s general receipts.

[4] “Cost Containment in Medicare; A Review of What Works and What Doesn’t,” prepared by The Urban Institute and Health Policy Alternatives, Inc., for AARP (written by Robert Berenson, Michael Hash, Thomas Ault, Beth Fuchs, Stephanie Maxwell, Lisa Potetz and Stephen Zuckerman), and Sarah Thomas, project manager for AARP’s Public Policy Institute. December 2008.

[5] The Medicare Economic Index includes a bundle of the various things that comprise the cost of physicians’ services such as the value of physician’s own time, non-physician employees’ compensation, rents, medical equipment, etc. It measures year-to-year changes in prices for these various components based on appropriate price proxies. Most of the components come from the Bureau of Labor Statistics. Changes in the costs of physicians’ time are measured through changes in nonfarm labor costs. Changes in productivity also are factored directly into the index. Source: Centers for Medicare and Medicaid Services.

[6] See statement of Douglas Holtz-Eakin, Director, CBO, “Medicare's Physician Fee Schedule,” before the Subcommittee on Health, Committee on Energy and Commerce, U.S. House of Representative. May 5, 2004.

[7] Laura A. Dummit, “Updating Medicare’s Physician Fees: The Sustainable Growth Rate Methodology,” National Health Policy Forum, The George Washington University, November 10, 2006.

[8] Note that this adjustment factor, whether positive or negative, has to fall within a range of 3 percent to 7 percent. Thus, an adjustment in any given year may not be sufficient by itself to bring cumulative spending in line with the target level. See statement of Peter R. Orszag, CBO Director, before the Committee on the Budget, United States Senate, “Health Care and the Budget: Issues and Challenges for Reform,” June 21, 2007.

[9] Ibid.

[10] “Cost Containment in Medicare; A Review of What Works and What Doesn’t,” loc. cit.

[11] CBO, “The Sustainable Growth Rate Formula for Setting Medicare’s Physician Payment Rates,” September 6, 2006.

[12]See statement of Douglas Holtz-Eakin, May 5, 2004, loc. cit.

[13] Source: CBO, Federal budget forecast, January, 2008.

[14] Centers for Medicare and Medicaid Services, Office of the Actuary, Memo from M. Kent Clemens, Joseph M. Lizonitz, and Suguna M. Murugesan, “Projected Medicare Part B Expenditures under Two Illustrative Scenarios with Alternative Physician Payment Updates,” May 12, 2009.

[15] 2009 Report of the Medicare Trustees, p.79.

[16] CBO, A Preliminary Analysis of the President’s Budget, March 2009, p.16; CBO, Budget Options Volume I, Health Care, p.112.

[17] “Cost Containment in Medicare; A Review of What Works and What Doesn’t,” loc. cit.

[18] For a discussion of Congress’s predisposition to discourage excessive billings, see testimony of Nancy-Ann Min DeParle, Administrator of the Health Care Financing Administration, before the Senate Finance Committee, “Private Contracting in Medicare,” February 26, 1998. (The Health Care Financing Administration was the former name of the Medicare’s parent agency, the Centers for Medicare and Medicaid Services).

[19] “Cost Containment in Medicare; A Review of What Works and What Doesn’t,” loc. cit.

[20] CBO, Supplemental Data, loc. cit.

[21] CBO, “Medicare Advantage: Private Health Plans in Medicare,” June 28, 2007.

[22] CBO, Supplemental Data, loc cit.