Health care payers around the country are testing whether models such as patient-centered medical homes can strengthen primary care delivery, reduce the cost and use of acute care services, and improve patient outcomes. Yet many of these efforts have had mixed results. A new Health Affairs article, related report, and blog post by Mathematica’s health experts inform the research base on primary care delivery, offering cautious optimism while identifying the challenges to improving patient outcomes. The research looks at cumulative results over four years for Mathematica’s evaluation of the Comprehensive Primary Care (CPC) initiative, a health care delivery and payment model developed by the Centers for Medicare & Medicaid Services (CMS) as one of its largest efforts to strengthen primary care.
The research tested whether requiring primary care practices to implement a new approach to delivering primary care, and providing financial and technical support to help them do so, reduced Medicare spending and improved quality over the four-year intervention. Mathematica’s experts evaluated CPC’s effects on care delivery and outcomes for Medicare fee-for-service beneficiaries attributed to the 497 practices that participated in CPC relative to those in 908 matched comparison practices. Building on a June 2016 article published in the New England Journal of Medicine on the first two years of CPC, this new article adds to the analysis Medicare claims and qualitative data from the last two years, data from a survey of Medicare beneficiaries in CPC’s final year, practice survey data from the last two years, and information from a survey of physicians in the study’s first and last years.
The final findings show slight improvements from CPC as well as continuing challenges, including the following:
- CPC practices reported improved primary care delivery, such as care management for high-risk patients, enhanced access, and improved coordination after care transitions.
- CPC slowed growth in emergency department visits by 2 percent relative to the comparison group.
- CPC also slowed growth in hospitalizations by 2 percent relative to the comparison group, though this finding was statistically significant only at the 10 percent level.
- Similar to the results for the first two years, Medicare expenditures grew more slowly for the CPC group than for the comparison group, but this change in Medicare expenditures was not enough to offset the initiative’s care management fees.
- CPC did not appreciably improve or worsen physician or beneficiary experience or practice performance on a limited set of Medicare claims-based quality measures.
“As CMS and other payers pursue value-based purchasing, the full four-year results of this evaluation are particularly relevant,” said Deborah Peikes, senior fellow and project director. “Primary care is an integral part of an efficient and effective health care system and these findings can be helpful to the payers, practices, and other participants in the Comprehensive Primary Care Plus model that more than 3,000 practices are now implementing.”
Mathematica is continuing research on new approaches to primary care, including an evaluation of the Comprehensive Primary Care Plus model, to help inform efforts to transform primary care. Early results from this evaluation will be reported in 2019.
