Improving Innovation in Health Services Through Better Payment Reforms

I. Medicare’s Dominant Role

The Medicare program heavily influences the ways in which hospitals, clinics, physician practices, and other service providers organize themselves to care for patients. Although Medicare’s payments represent only about 20 percent of the total amount spent on health services, the program is the single largest payer in most markets (the combined payments from private insurers is often larger than what Medicare pays to providers, but private insurance payments are divided up among many employers who do not usually act in concert with each other). As a consequence, hospitals and physician groups have no choice but to be responsive to the regulatory demands of their largest source of revenue.

Moreover, private insurers typically choose to piggyback onto Medicare’s payment systems. It is an intensive and expensive exercise to develop a payment system for hospital stays and physician services; instead of starting from scratch, insurers have tended to start with Medicare’s system and then increase the payments (with some refinements) to make the rates more acceptable to the providers with whom they have contracts. The basic methodology for making these payments is driven by federal rules. If the government decides to make a modification to the payment methodology used by Medicare, the change will usually cascade through private insurance payments to providers as well.

The reliance of private insurers on Medicare’s payment systems means Medicare’s influence goes well beyond the care structures devoted primarily to the elderly. Indeed, because Medicare’s payment systems set the basic terms for provider payments across the entire insured population, the program is by far the most important influence on the organization of medical service delivery for all patients in the United States.

Since the program’s enactment in 1965, Medicare’s orientation has been toward fee-for-service insurance. Under fee-for-service, the government has traditionally played the role of a passive insurer, administering payments for the services provided to the program’s participants without taking an active role in the care decisions of practitioners, or even decisions about the processes by which practitioners care for patients. For most the program’s history, if a licensed practitioner provided services covered by Medicare to a patient, then the program would provide a payment for those services, with generally no other questions asked.

Over the years, the rules for making payments to providers have become the source of endless revision, complexity, and sometimes controversy. The overall trend has been toward payments that lag what private insurers pay but which remain sufficient to allow the continued autonomy of those providing the services.

An important and regrettable consequence of this payment model has been the fragmentation of the care delivery system in most communities. Hospitals, physician groups, outpatient clinics, specialty facilities, labs, and drug makers all get reimbursed through separate payment systems, making it unnecessary for them to become explicitly dependent on a relationship with a more organized system of care delivery.

II. The “Delivery System Reform” Approach of the ACA

To their credit, the authors of the ACA recognized the important role Medicare has played in shaping today’s care delivery arrangements, and they realized that the same clout that has led to pervasive fragmentation and disorganization might be harnessed, with different payment rules, to encourage providers to begin managing the care process more efficiently.

Several provisions were inserted into the Affordable Care Act (ACA) to give the executive branch the authority to begin revising Medicare’s payment systems to encourage this kind of reform and evolution. The admirable goal of these provisions — often called “delivery system reforms” — was to both lower costs and improve the quality of services provided to patients.

Among the most high-profile of these initiatives is Accountable Care Organizations, or ACOs. ACOs are provider-driven managed care entities that can earn bonus payments from the government by holding costs below what would have occurred in unmanaged fee-for-service. Hospitals and physician groups that form ACOs presumably have an incentive to provide better oversight of the care delivery process to earn these bonus payments.

An important feature of the ACO program is the method by which an ACO becomes responsible for managing the cost and quality of the care provided to a beneficiary. Medicare beneficiaries are not given the opportunity to enroll in an ACO; instead they are assigned to them based on their claims experience. When a physician joins an ACO, all of the beneficiaries who use that physician as their main point-of-contact for health services are designated as enrolled in the ACO. The costs of caring for these patients then becomes part of the evaluation of the ACO’s performance.

The ACA also gave the executive branch the authority to test other approaches to increasing the value of services provided to patients, such as “bundled” payments for episodes of care. With bundled payments, Medicare makes one payment to one provider for a full episode of care (such as joint replacement surgery), and that provider is responsible for paying the other providers involved in successfully completing the service for the patient. Quality is measured by improvement in the health of the patient from the procedure. The law also authorized penalties for hospitals that incur higher than normal readmissions of discharged patients.

III. Excessive Reliance on the Creativity and Expertise of Administrative Agencies

The delivery system reforms undertaken under the authorities of the ACA may produce some positive improvement in terms of more efficiency and quality in care delivery. But this progress is wholly dependent on the ability of the same federal agency that issued rules contributing to excessive fee-for-service costs — the Centers for Medicare and Medicaid Services, or CMS — to now write new rules that will reduce the wasteful spending.

It is not possible for one centralized government agency to have a full and complete understanding of a system as complex as the provision of medical care. Well-intentioned changes will frequently have unintended and unanticipated effects. For instance, with ACOs, CMS sought to create within Medicare the option of a provider-led managed care alternative to unmanaged fee-for-service, modeled on plans previously built in the private sector, such as the large multi-specialty physician group Sharp Healthcare in San Diego. Sharp was eager to help CMS launch its new effort, and signed up immediately as one of CMS’ “Pioneer ACOs” in 2011. The Pioneer option was designed explicitly to allow mature systems like Sharp to take full advantage immediately of the opportunities ACOs would supposedly provide for implementing cost-cutting efforts. By 2014, though, Sharp had dropped out of the ACO program altogether and returned to the traditional unmanaged fee-for-service Medicare program. Sharp argues that the Pioneer program was poorly designed because it penalized plans that had already had success in cutting costs. Sharp was not alone in abandoning the Pioneer ACO experiment. Of the original 32 Pioneer participants, only 8 were left as of December 2016.

Further, regulating medical care delivery is susceptible to the same kind political pressures that political scientists have identified in many other areas of government regulation. Those who are regulated by the government expend much effort to influence the rulemaking process and ensure their businesses can survive and perhaps prosper within the government’s framework. Indeed, a typical phenomenon is for those regulated by the government to work the regulatory process to the point where the rules become a barrier to outside competition and disruptive creativity.

In the context of health care, steady improvement in productivity and quality would require moving patients and market share away from high-cost and low-quality providers toward those who can deliver more for less. The tendency of Medicare’s payment regulations has been to treat all providers equally, regardless of the value they provide patients. The reforms set in motion by the ACA try to steer away from this tendency, but there are powerful interests pushing back against any kind of change that would scale back Medicare payments to large numbers of licensed medical providers. Over the medium and long run, it will be difficult for a regulatory process alone to create the sustained improvements in productivity necessary to keep Medicare costs growing at a sustainable rate.

IV. A Better Approach

The architects of the Affordable Care Act (ACA), and those who implemented the law in the Obama administration, were right to consider what can be done through Medicare to improve the overall effectiveness of health care in the United States. Without changes in Medicare, it will be difficult to move away from the fragmentation and lack of coordination that remains typical in most communities today.

But it is not necessary to rely solely on the federal government to set in motion the changes needed to improve productivity and eliminate unnecessary spending. Some progress can be made in that way, but this approach relies too heavily on the government acting against its natural tendency toward inertia and protection of incumbents.

A better approach would rely on economic incentives to produce continual improvements in productivity. The key is to get the direct involvement of the beneficiaries in selecting low-cost and high-value care delivery options. The operating assumption of this approach is that if consumer incentives can be built into the process of selecting from among care delivery options, the benefits could be continual, as they are in most sectors of the economy, and less dependent upon the regulatory insights and creativity of a federal agency.

There are a number of different ways to bring beneficiaries more directly into the process of seeking higher value care. All of them should be pursued as far as possible through existing authority, and then through changes in legislation as needed.

A.  Redesign Medicare Basic Choice Structure and Options, and Changing Medigap

Medicare has evolved over the years into a partial beneficiary choice program. Program enrollees have the option of taking their entitlement to benefits in the form of enrollment in private health insurance plans — called Medicare Advantage (MA) plans — or getting their coverage through the traditional, unmanaged fee-for-service program. In 2016, about 30 percent of beneficiaries opted for an MA plan. Beneficiaries who remain in fee-for-service have the option to enroll in a supplemental insurance plan — called Medigap — to pay for some or nearly all of the costs that Medicare does not cover. Further, all Medicare beneficiaries — those in MA plans and fee-for-service — can choose to enroll in separate, privately-administered coverage for prescription drugs.

The introduction of ACOs by the Obama administration created yet another option for getting care, although it was done without providing an explicit enrollment option for the beneficiaries.

The process by which beneficiaries decide what options will work best for them should be reformed to encourage transparency and strong price competition. During a period of open enrollment (which might be biannual), beneficiaries should be given the option to enroll in an MA plan, a successor plan to ACOs (perhaps called an Integrated Delivery Network, or IDN), or unmanaged fee-for-service. IDNs would be provider-driven managed care networks; they would be subjected to less regulation than what has been applied to ACOs. The benefits covered by each of these options would be actuarially identical, but the premiums would differ based on the relative efficiency of the insurance plans. Beneficiaries enrolling in an IDN would pay reduced premiums based on the ability of the IDN to deliver Medicare services for costs below unmanaged fee-for-service.

Beneficiaries selecting an IDN or unmanaged fee-for-service would have the option of enrolling in a Medigap plan too. But those in fee-for-service would not be allowed to enroll in a plan that covered all of the cost-sharing required by Medicare, while those in an IDN would be given greater latitude to get more expansive supplemental insurance.

Enrollment in Medigap insurance, in combination with unmanaged fee-for-service Medicare, increases overall Medicare costs because the beneficiaries pay little or nothing for using more services. The most popular Medigap plans are those that fill in as much of the program’s cost-sharing requirement as is allowed by law. One study found that beneficiaries enrolled in Medigap insurance incurred Medicare costs that were, on average, 22 percent above the expenses incurred by those without the supplemental insurance. The added utilization of services that comes with enrollment in Medigap policies is not reflected fully in the premiums charged by Medigap plans because Medicare pays for the majority of most medical bills (for instance, Medicare pays 80 percent of the cost of physician services). In effect, it is federal taxpayers, through the subsidies provided to Medicare, who pay for most of the added health care utilization driven by Medigap enrollment. A policy that restricted the kinds of Medigap plans beneficiaries can select while enrolled in unmanaged fee-for-service is appropriate in this context.

After choosing an MA plan, an IDN, or fee-for-service, beneficiaries would also pick drug coverage. Drug plans that are coordinated with an MA plan or an IDN are likely to be far less expensive than plans tied to fee-for-service.

The beneficiaries would be presented with these options in a clear and transparent selection process so that the price differences of various combinations of coverage would be clear and understandable.

This kind of beneficiary choice model would create strong incentives for MA plans and IDNs to drive down their premiums to attract beneficiary enrollment, which in turn would mean that the managed care plans would need to work aggressively with hospitals and physician groups to reorganize their business practices to increase their productivity. Plans that found innovative ways to cut costs and provide better services to patients would be rewarded with higher enrollment by the program’s enrollees.

B.  Reference-Based Pricing

Although a better choice structure would go a long way toward changing the incentives in Medicare and beyond, it is also possible to bring more beneficiary choice directly into the unmanaged fee-for-service option of the program.

Currently, Medicare pays administered prices for hospital, physician, and related services in separately-regulated payment systems. Under the ACA, tests of “bundled payments” are underway to see if combining the separate fees into a larger payout will improve efficiency and quality for high-frequency procedures.

CMS should establish a separate payment model for discrete procedures and services using competitive bidding and reference-based pricing to determine payment. Calpers, the manager of insurance coverage for California state employees, has built a model for this approach that could be adapted for the Medicare program. Calpers identified that its plan enrollees often encountered vast price differences for the same intervention, such as joint replacement surgery, cataract removal, and colonoscopies. An investigation found no discernible difference in quality between systems that charged $1,250 for arthroscopy on the knee and those that charged $15,500 to do same thing. Instead of attempting to steer all of its enrollees to lower cost providers, Calpers continued to allow participants to select their provider from an approved list. But if a patient chose a provider with a much higher price, then he would have to pay the difference between the “reference price,” based on a low-cost offering, and the price charged by the patient’s selected provider. Implementing this reference-based pricing reduced Calpers’ costs for arthroscopy by $2.3 million over a two-year period. More was saved in joint replacement procedures.

This model could be adapted for Medicare by giving providers the opportunity to bid on a commonly-used procedure. The government would provide a list of targeted services and procedures and ask providers to come forward with all-in prices for delivering care for the identified procedures. The reference price could be set on a weighted average of the most efficient providers, or some percentage of the weighted average. These reference prices would then be compared to what Medicare would pay for the procedure based on the current system of payment regulations. Beneficiaries would be fee to get the service from any approved provider, but if they went to one that charged less than amount that would be paid using the usual fee-for-service regulations, they would share in the savings with the government. Conversely, if they selected a provider with a price above the reference amount, and also above the amount the government would pay under the traditional rules, they would be required to pay the excess cost themselves.

Overtime, this approach would drive strong competition among providers of these services, and the government could systematically add more and more high-frequency service items to the list for which this kind of pricing would apply.

Whenever competitive bidding is considered in the medical context, providers of services worry that important quality and risk considerations will be pushed aside in the name of cost-cutting. This concern is likely overstated, and can be addressed in any case with proper design. There is very little evidence of a strong correlation between high cost and high prices and better quality in medical care. In fact, there are studies showing the opposite is true in certain contexts.

Further, it is possible to collect clinical data on patient outcomes to control for important differences in the quality of care provided to patients. In the context of reference-based pricing, Medicare should require all participating providers to collect and submit the relevant outcome data in a uniform format to allow for easy comparisons across providers. The regulatory process could then prevent any provider with unacceptable levels of quality from participating in the program.

C.  Competitive Bidding of Drugs and High-Volume Products

Medicare purchases large amounts of products that could be subjected to more discipline through competitive bidding. Part B of the program — for outpatient services — purchases large amounts pharmaceutical products for use in physician offices. For drugs with comparable therapeutic purposes, it should be possible to establish a competitive bidding process to drive down prices.

An attempt to do this beginning in 2006, the Competitive Acquisition Program, or CAP, failed to attract enough interest among physicians to work in large part because they saw no financial advantage to switching from the existing regulated system to one based on more uncertain bids from drug companies.

Another attempt at using competitive bidding to set part B drug prices would need to correct for CAP’s primary flaw by ensuring the physicians who now benefit from the inflated prices paid by the federal government for the drugs will share in the savings from moving to a competitive model. The federal government should explicitly design a program that provides most of the gains to the physicians to lower their resistance to the change. More savings for the federal government could be phased-in over time.

V. Summing Up

There is broad agreement among policy experts that Medicare plays a pivotal role in the health system. The program’s rules are important factors in how hospitals and physicians organize themselves to take care of patients. For many years, Medicare’s payment systems encouraged fragmented, uncoordinated, and costly care. Now there is strong interest in changing how Medicare works to improve the efficiency of the medical care delivery system.

The Obama administration pursued a strategy centered on writing better regulations that encourage high-value instead of high-volume care. There are elements of this approach that might work and produce better results.

But it is short-sighted to rely entirely on enlightened federal rulemaking to cut costs. The government has a long track record of writing regulations that mainly protect the status quo at the expense of cost-cutting innovation.

The current “delivery system reform” agenda should be amended to allow much greater consumer choice and economic incentives to cut costs. The Medicare beneficiaries themselves can be the strongest force for efficiency and productivity improvement if they are given the appropriate financial incentives to enroll in low-cost, high-quality options. Those supplying services to patients will find new and creative ways to cut cost and provide better service if doing so will allow them to attract higher enrollment into their systems of care, and thus boost their revenue.