Challenges for Medicare at 50

Editor’s Note: this post originally appeared in NEJM.

Medicare will be under huge pressure in the next few years from the demographic bulge and rising per-enrollee spending. We see this challenge as an opportunity for the program to take more aggressive action to improve care delivery. As the largest payer for most U.S. health care providers, Medicare has the potential to improve the quality and efficiency of U.S. health care delivery overall — but it hasn’t used this clout effectively in recent years. As the participants in a recent conference that we organized on “Strengthening Medicare for 2030” concluded, Medicare would be wise to accelerate the adoption of provider payment reforms in the traditional program and increase competition among the private health plans available to beneficiaries under Medicare Advantage. Doing both of these things effectively would improve results for Medicare and move the health care system toward higher quality and sustainable costs.


In the past, Medicare has taken bold steps to reform provider payment. Inpatient hospital prospective payment, enacted in 1983, replaced a system of reimbursement of costs subject to increasingly tight caps with an almost completely prospective system, using national rates adjusted for local factors. The shift affected more than just Medicare; Medicaid programs and some private payers followed Medicare’s lead. One result was that the length of hospital stays dropped dramatically. Subsequently, over many years, prospective payment was applied to outpatient hospital services, surgical centers, and other providers.

Meanwhile, however, a poorly conceived approach for physician payment, the sustainable growth rate formula (SGR), stymied physician payment reform for more than a decade. The SGR controlled physician reimbursement rates so stringently (without controlling service volume) that Congress repeatedly postponed its effects before finally dropping it this year. Moreover, while private payers were increasing patient cost sharing to limit premium increases, Medicare left its benefit structure essentially unchanged from 1965 on and failed to address the subsidy to private supplemental coverage, which often eliminates all cost sharing at the point of service.

Further payment reform is now clearly in order. As the population ages, patients with multiple chronic diseases account for a growing share of health spending — which increases the need to provide more effective care at lower cost. Both public and private payers are now experimenting with alternative payment models (APMs) that encourage all providers participating in an enrollee’s care to work together to deliver care tailored to that patient’s needs. These models include accountable care organizations (ACOs), bundled payment for episodes of care, and patient-centered medical homes for primary care.

The secretary of health and human services has set the ambitious goal of moving half of Medicare payments to APMs by the end of 2018, but we argue that Medicare is not currently pursuing a course that will achieve this goal — let alone with APMs that successfully improve value. The shortcomings of Medicare’s efforts often reflect limitations on the authority that Congress has granted to the Centers for Medicare and Medicaid Services (CMS), but Congress may prove more supportive now that it has taken bipartisan action to replace the SGR and encourage payment reform.

If APMs are to be broadly adopted, Medicare models need to be attractive to providers, whose active support and participation are essential to the success of any payment system. To date, many providers have found participating in APMs to be dauntingly complex and to require up-front investments that are unlikely to be justified by small and uncertain rewards. Some have joined the Medicare APM pilot projects only to later withdraw from them.

Part of the difficulty involves provider-specific benchmarks that reward improvement rather than the level of performance. Frequent updating of these benchmarks undermines the business case for investments by providers to improve the effectiveness of care delivery. If Medicare committed to avoiding the use of provider-specific spending trends to update benchmarks and shifted as quickly as possible to benchmarks reflecting regional or national standards, we believe it would have more success. The recently passed Medicare Access and CHIP Reauthorization Act (MACRA), which replaced the SGR with a merit-based incentive payment system (MIPS), opens the way for better benchmarks by providing a strong incentive for physicians to participate in APMs.2 Where evidence of their effectiveness is strong, bundled payments can be mandated. The recent proposal by Medicare to mandate bundled payment for joint replacement and transition to a regional benchmark is encouraging, although many aspects of that proposal are problematic.3

The quality measures used for APMs can also be improved and simplified, and quality data can be made available to beneficiaries in user-friendly form. Lack of alignment among Medicare, private payers, and Medicaid programs on quality measures has burdened providers and limited their incentives for improving quality. Better alignment requires expanded discussion among payers, provider associations, professional societies, and quality-measurement experts.

Medicare APMs also do not support provider’s efforts to engage beneficiaries in their care. To coordinate care effectively, providers need to be able to engage patients actively in their care, especially in chronic disease management, and steer them to other providers on their team. But under Medicare, an ACO may not even be able to identify the patients for whom it’s responsible until after the contract year ends. Shifting from an attribution model to an enrollment model — in which beneficiaries choose to participate in an ACO and have incentives to do so and to use specialists that are cooperating with their ACO — is the most effective way of engaging beneficiaries.

Medicare has offered private-plan options to beneficiaries since the 1980s, primarily to expand their choices rather than to contain costs. Indeed, Medicare has actually subsidized Medicare Advantage plans, although these subsidies were reduced by the Affordable Care Act (ACA). Medicare could reduce program spending by introducing competitive bidding by Medicare Advantage plans, requiring each plan to submit a bid reflecting the price at which it would agree to deliver the Medicare package of benefits in a given geographic area and either setting the Medicare contribution on the basis of the second-lowest bid or choosing a percentile in the distribution of bids. This approach, along with other practices used in the insurance exchanges created by the ACA, would probably lead to lower Medicare payments and make it easier for beneficiaries to compare the available Medicare Advantage alternatives.

At least for the immediate future, we would favor limiting competitive bidding to Medicare Advantage plans so that Medicare gains experience with the bidding process and can improve risk adjustment to avert favorable risk selection. If this experience is successful, we would favor including the cost of traditional Medicare as another bid and applying the resulting Medicare contribution to the traditional program as well. The federal contribution to Medicare would then be the competitively determined cost of providing the Medicare benefit package in the area (perhaps the second-lowest bid or an average of the bids) — a policy often referred to as “premium support.” We acknowledge the political difficulties of taking this step but believe it would result in higher-quality care at lower cost.4

Another area for reform involves simplifying and improving the Medicare benefit structure. A unified benefit structure with one deductible for Parts A and B and an out-of-pocket maximum would make sense, especially if some physician office visits were exempted from the deductible to avoid discouraging early treatment. Supplemental plans could also be required to leave some cost sharing for beneficiaries to pay at the point of service; MACRA took an initial step in this direction. These changes would have modest effects on Medicare spending but would remove a barrier to engagement of beneficiaries in the payment reforms discussed above.

Medicare is in a position to lead the health system toward more efficient delivery of care. To meet this challenge, we believe that CMS needs to step up the vigor of its pursuit of payment reform in the traditional Medicare program and competition in Medicare Advantage.