Steady Progress in Multi-Payer Primary Care Transformation

Focus Area:
Primary Care Transformation

The release on October 17, 2016 of 2015 Shared Savings and Quality Results for year two of the Center for Medicare and Medicaid Innovation’s (CMMI) Comprehensive Primary Care (CPC) Initiative marked another milestone in efforts to align payers and transform primary care.  Observers looking for an airtight identification of the right mix of payment, reporting, practice transformation, and payer alignment efforts that conclusively deliver high-quality care and lower costs will find much to criticize. Those with more modest—and appropriate—aspirations of learning what works to improve the efficacy of primary care will be quite pleased.

Positive Results Continue

The Shared Savings and Quality Results report compares 2015 performance to expected performance in each of the seven markets in which CPC operates. As dictated by the program design, the unit of accountability here is the market—those markets that outperform their benchmark get to share in the savings. Four of the markets (AR, CO, OK, and OR) met that standard in 2015 and three (NJ, NY, and OH) did not; across all markets, overall program savings covered the additional expenses on care management and technical assistance for the practices. This is a notable improvement from 2014— when only one of the seven regions showed net savings and gross savings (before care management fees and technical assistance) were half the 2015 figure.

Quality results were even more definitive. Admission rates for selected conditions were lower than expected in five of seven regions, and patient-reported experiences were higher in CPC practices than in non-participating practices.  Performance on electronic Clinical Quality Measures (eCQMs) also exceeded national benchmarks, particularly on preventive health measures.

Variation and Accountability

CPC—and it successor, Comprehensive Primary Care Plus (CPC+)—is complex. CMMI designs and oversees the initiative and hires the contractors. Payers agree to work together to align payments, measures, and reporting to primary care practices. These practices have to go through an expensive and time-consuming effort to change their day-to-day work in order to improve the value of the care they deliver. If these efforts are successful, patients engage with these practices and their medical care differently.

Results need to be assessed at the national, market, and practice levels. At the market level, the variation between the regions is telling.  Oklahoma was the big saver, beating its benchmark by 7 %—while New Jersey missed its mark by 4 %. On the quality side, hospital readmission rates varied by 16 % between the lowest market (Oregon) and the highest (New York).

The savings report from CMMI did not release provider-level data, where variation is undoubtedly even greater. In response to CPC participant feedback that market-level shared savings have too little connection to practice-level activities, CPC+ does not have a shared savings component.

We think the market-level variation needs to be explored further. Is technical assistance to practices more effective in some markets than others? Are payers in the markets that beat their benchmarks aligning in a different way? Perhaps they select more important common measures, communicate with practices and technical assistance providers with more clarity, or have better practice reporting and feedback. Maybe they just play together better in the sandbox.

Winning or Learning?

No other payer airs its laundry like Medicare does. While that subjects Medicare’s payment reform efforts to intense scrutiny and criticism that its private sector colleagues avoid, it also means that Medicare leads health care delivery system transformation efforts.  Other participants—providers and payers—benefit from it.

As results from various CMMI-led payment reform efforts become public, there is a temptation to render definitive verdicts on the efficacy of the efforts. The release of equivocal results leads to a rush to judgment—usually negative or discouraging, particularly from observers who have themselves never developed and implemented health care delivery system transformation

A second camp—more research oriented—considers these efforts to be rigorous social science interventions under carefully controlled conditions with clear comparisons. If carefully done, the results will be conclusive or, at the least, reliable.

Both of these views, however, ignore the messy reality of CMMI’s payment reform efforts. They are multi-level interventions, in varying environments with significant mid-period shifts in personnel and activities. The appropriate blend of payment incentives remains a topic of mere speculation. Hope for quick or definitive verdicts is illusory.

Instead what is needed is persistence, measurement, and a willingness to learn. The stronger and earlier positive results from CPC compared to its predecessor—the Multi-Payer Advanced Primary Care Practice (MAPCP) demonstration—show that learning is occurring; the more directive set of standards employed in CPC appears superior to the policy of Medicare following the local lead, as was the case in MAPCP.

We know from international comparisons that no delivery system can perform well without strong primary care. We know from economics that actors respond to consistent financial signals. We know from engineering that systems need feedback to improve. With CPC, and soon with CPC+, we are learning how to apply those lessons in the diverse, multi-payer, dynamic, and relentlessly local environments that comprise US health care. These results show clearly that we are making steady progress in multi-payer primary care transformation, moving toward better population health, lower population costs, and more satisfied patients.