States May Be Interested in Value-Based Payments, but Commercial Insurers Are Not

CMS has issued the fourth of five planned annual reports on its Medicaid State Innovation Models (SIM) Initiative. Under the SIM initiative, six “test states” – Arkansas, Maine, Massachusetts, Minnesota, Oregon, and Vermont – have been awarded funds and are using policy levers to facilitate the creation or spread of innovative models and integrating population health and broader stakeholder perspectives into health care delivery and payment redesign models. The latest report describes the experiences of providers, health systems, consumers, payers, and state officials during the final full implementation year.

The evaluation, conducted by RTI International, working with The Urban Institute, National Academy for State Health Policy, and The Henne Group, found all six states have introduced value-based payment models (VBP models) in their Medicaid programs and are offering technical assistance to providers, social service and community-based organizations, and others to implement new delivery system models. States are also offering services, such as health IT and data analytic investment, that “enable or improve model effectiveness,” according to the Year 4 Annual Report.

However, the evaluation found “limited interest” among private payers and insurers in aligning their existing VBP models with state models. “All states struggled with effective engagement of private payers and insurers to expand VBP models beyond existing efforts and to achieve alignment across multiple payers,” the report finds. “Although private payers and insurers were willing to discuss the states’ conceptualization of VBP models, most did not make changes to the VBP models they offered to providers.”

As a result of limited interest among commercial payers, “states ultimately focused on Medicaid or state employee health plans over which they had control.”

There are several factors contributing to this lack of multi-payer alignment around common payment models, including a difference in business goals between Medicaid and commercial payers. Commercial payers in Maine, for example, reported issues related to value-based insurance design and multi-payer measure alignment; this was due to “insufficient engagement with payers” when the SIM goals were established, the report says. In addition, commercial payers in Maine are reluctant to change the design of their insurance products “in response to a single state’s recommendations,” and there is a “preference for making product design changes in response to their clients’ needs.”

If you are a business, your client’s needs come first. If a state becomes a client of an insurer, perhaps this VBP model alignment problem would be easier to solve. But other state’s experiences would indicate that is unlikely.

The RTI evaluation identified several other factors contributing to the lack of payer alignment in payment models including: the proprietary nature of information (e.g., commercial payers in Minnesota preferred not to share details on quality and utilization measures and performance reports for providers, which “limited the type of dialogue necessary to advance multi-payer payment reform”) and competitive concerns (e.g., payers that have invested in changes in payment reforms are “concerned that the returns on those investments are accruing to other parties”).

Many of these concerns are age-old problems; in the policy world, we encounter these types of issues frequently: (1) the complaint that “you didn’t engage us from the beginning;” (2) the fact that health care delivery is competitive in a capitalist market; and (3) the free rider problem.

We could add two more issues or concerns to the list above: (4) “You don’t understand clinical issues” (e.g., physicians in Arkansas felt that “state decision-makers were too far removed from daily clinical practice to understand what would work effectively”) and (5) We would have to create multiple products – e.g., products tailored to the needs of the Medicaid population vs. commercial population.

In the six test states, participation by health care practitioners in innovative models varies wildly. For example, in testing an integrated care model, the number of participating providers varied such that in Massachusetts, 10% of the Medicaid population was reached, while in Oregon, the comparable figure was 85%, as noted in the chart below (from the evaluation report):

For tests of the patient-centered medical home (PCMH) or health home model in these six states, 17% of the Medicaid population was reached in Maine, compared with 70% in Vermont and 75% in Oregon.

Engaging private payers in innovative payment models is important for health care system transformation. Given that some payers in these six test states were reluctant to change the design of their insurance products in response to a single state’s recommendations, states might consider combining forces for the purposes of designing more responsive VBP models. Another alternative is to ask insurers, along with representatives from other health care entities, consumer groups and employers to participate in system design change from the beginning of the process. VBP may lead to health cost reductions if implemented widely and well, but such achievements will be nearly impossible if entities that control so much of a population’s health care coverage choices are not involved.

Nearly Every State Engaged in Value-Based Payment Models

Value-based payment is not just for Medicare and private payers, states are also engaged in a multitude of value-based payment initiatives. While variation exists in the scope, leadership commitment, and resources devoted to these efforts, “more than 40 states have a state-initiated plan or strategy” to move toward value-based payment instead of fee-for-service (FFS) payment, and “almost half of those initiatives are multi-payer in scope,” according to the report by Change Healthcare, based on a national study of  publicly available information compiled from May through October 2017.

Of note:

• Of the states engaged in VBP, nearly all are using patient-centered medical homes (PCMH) or health homes (HH).
• For the most part, these tests are being tried in Medicaid; 31 states are testing one of these value-based approaches in Medicaid alone. In three states, they are testing at least one of the two in Medicaid and with their state employees: Oklahoma, Tennessee, and Washington. Thirteen states have multi-payer efforts: Arkansas, Colorado, Connecticut, Delaware, Idaho, Iowa, Maryland, Michigan, New York, Ohio, Pennsylvania, Rhode Island, and Vermont.
• Fourteen states have chosen the ACO model: seven in Medicaid (Illinois, Maine, Massachusetts, Minnesota, New Hampshire, New Jersey, Oregon), two in Medicaid and with state employees (Oklahoma, Washington), and five in multi-payer arrangements (Colorado, Iowa, New York, Rhode Island, Vermont).
• Episodes of care (EOC) is another approach being used; currently in 12 states though some are considering it and some only use EOC in a single service area: Colorado, Maine, and South Carolina are considering EOC; New York is using it only in maternity and chronic care; Washington is using it only for total joint episodes of care.
• Pay for performance is also being tried in the following 12 states: California, Colorado, Connecticut, Delaware, District of Columbia, Florida, Iowa, Maryland, Missouri, New Hampshire, Texas, and Wisconsin.
• Seven states have “little to no activity around value-based payment” (Georgia, Indiana, Louisiana, Mississippi, North Dakota, South Dakota, Wyoming).

The following map (created by M2 based on data in the Change Healthcare report) shows the details of each state’s initiatives in this area:

For a few states, testing alternative payment models is not new. Some states have been engaged in these efforts for nearly a decade. Minnesota was the first state to engage in a value-based payment approach of some sort, in 2008. Colorado and Maryland began their efforts in 2011; Oregon in 2012, Arkansas and Vermont in 2013.

States are often the leaders in testing new ideas. Alternative payment models for health care are no different. As health care costs continue to rise, and most states need to balance their budgets every year, it makes sense that finding different ways to pay for Medicaid and state employees’ health care is something nearly every state is focused on.

VBP is not dead, as evidenced by nearly all states are taking action to transition from a FFS approach to one that is focused on value. To slow health care costs, we have to stop paying for what doesn’t work. The more we can reward outcomes in health care, the more likely it is that providers and patients will make decisions based improvements in care, rather than adverse financial incentives for low-value care.