Payment Reform: Innovation, Evaluation, and Perspiration

My e-mail inbox is full of innovations—solicitations for health care meetings, most offering “unique opportunities” to hear from “national leaders,” all of whom are busy innovating.

Much of the conference fodder focuses on payment reform. Thanks largely to the Affordable Care Act (ACA), the federal government—of all places—is an agent of innovation, paying for new services in new ways through initiatives, programs, and payment mechanisms such as episodes of care, accountable care organizations, health homes, and comprehensive primary care.

The commercial insurance industry is following suit, and we are awash in a veritable tsunami of innovations. Value-Based Payments and Alternative Payment Mechanisms (known as VBPs and APMs for the acronym-inclined)—involving population-based payments, risk-sharing arrangements, bundled payments and pay for performance—are swamping the industry.

Innovation is not the same as improvement, however. Payment innovations may create employment opportunities for consultants, but do they work?

In a highly competitive market, that question is answered by customers who reward payment initiatives that improve service value. No such luck in health care—where patients use other people’s money to pay for services they need greatly and understand poorly.

So innovators and imitators in complex nonmarket-based systems like education and social services must rely on evaluations of their work. In health care, this is the domain of health services research—a burgeoning if less lucrative field than VBP consulting.

Programmatic evaluations are less authoritative than the market. There are multiple methods for evaluating an innovation—the gold standard of randomized controlled trials can rarely be created in real world conditions of health care payment. Even when superior methodologies are employed, the conclusions that evaluations deliver are directional and not definitive. Given this, there is a robust debate about the extent to which evaluations should be insisted upon in health policy innovations.

This inability of evaluations to provide a definitive answer on whether a payment innovation “works” is illustrated in a recent report issued by the Milbank Memorial Fund, evaluating eight evaluations of multi-payer primary care transformation projects participating in the Fund’s Multi-State Collaborative. Researchers from Mathematica Policy Research found ample reasons to question positive and negative findings in all eight evaluations, noting ways that study designs, comparison groups, study power, and statistical methods could all be improved. Findings like these can create problems for payment innovators—and especially for their financiers or sponsors, who are looking for a definitive answer about whether an innovation is worthy of further investment. Congress explicitly forbids Medicare from implementing payment innovations developed by the Centers for Medicare and Medicaid Innovation (CMMI) unless the Office of the Actuary can certify that the innovation reduced costs and maintained quality or improved quality at no incremental cost. Such authorization is a high bar.

On the private sector side, there is no independent arbiter of effectiveness. Insurers, locked in private negotiations with providers, often lack the leverage, capacity, and focus to commit to and test a specific payment reform. Employers are advised by experts about how to evaluate payment reforms but do not have the ability to render a consensus judgment. Many choose to remain passive purchasers of health insurance, relying on insurance companies to figure out what payment reform works. Some large self-insured companies choose not to participate in payment reform efforts, letting other payers bear the burden of investing and experimenting.

So, in our multi-payer, financially fragmented health care environment, how will we come to consensus on when a payment innovation works? In the wake of the ACA, we are learning a lot of things very quickly:

• Public sector leadership. We would not have the level of innovation and ferment around payment reform without the initiatives from CMMI and state Medicaid programs. It may run counter to our country’s principles of market competition and private sector innovation, but when it comes to health care payments, Medicare remains “the straw that stirs the drink,” as baseball star Reggie Jackson proclaimed about himself 45 years ago.
• High-quality third-party evaluations made quickly available. We are learning how to evaluate better. There is a science here; it should not be performed by the providers or payers who are involved in the innovation, and the findings should be made available quickly and transparently.
• A learning attitude. Even the best evaluations will return qualified findings—there will not be definitive verdicts. Instead we are engaged in a collective discovery of what works. In addition to formal evaluations, learning what works means measuring performance in public, transparent ways. It means expecting improvement, not objective answers and accepting the particularity of local cultures and settings that will change the implementation of payment reforms.

At the Fund, we see these lessons in our work with multi-payer primary care transformation. After five or more long years, many of the projects in the Fund’s Multi-State Collaborative are showing real improvements in population health and costs. Is there innovation in this work? Certainly. But more importantly there is also evaluation and education, persistence, and perspiration.