Download the full report with sources
Connecticut Medicaid is considering reforms to primary care delivery and payment. The CT Health Policy Project is collecting evidence from other states and programs to help inform that planning. Part 1 focused on Connecting with community services to improve health.
A major part of DSS’s planning for reform is to consider changing how Connecticut Medicaid pays for primary care. DSS’s goal in reconsidering the current payment model is to:
Provide sufficient payment to enable and integrate care delivery redesign and performance measurement opportunities and ensure that payment adequately supports and advances biopsychosocial health and drives accountability for outcomes
In Connecticut and nationally, there is a great deal of interest in moving all healthcare payment, including for primary care, away from a system that pays for individual services to value-based purchasing (VBP) that rewards quality and places providers at financial risk to lower costs. Unfortunately, despite many trials across the US devoting years and massive resources to the idea, there is no evidence that value-based purchasing (VBP), including primary care payment reform, either improves the quality of care or reduces spending.
Findings from the literature:
The largest test of primary care payment reform so far evaluated, is the federal Center for Medicare and Medicaid Innovation’s (CMMI) CPC+ program engaged 2,610 primary care practices across the US in multi-payer payment and care delivery reform. The final CPC+ program evaluation was very disappointing. Independent evaluators found ED visits and acute hospitalizations down slightly, however those impacts were offset by increases in other services. When the costs of the program’s enhanced payments were included, the program lost over $400 million. CPC+’s record on quality improvement was small and mixed.
- Berwick, et. al. conclude that “CMMI has devoted more attention to testing new payment models than to fostering specific care models.” They recommend “Rebalance CMMI model tests toward delivery system redesigns, not just new payment models.”
- Berenson, et. al. notes that “Despite more theoretical, often creative, proposals and a range of primary care payment demonstrations, we are not much closer to a consensus payment model”.
- Milad, et. al. stated, “evidence from this review suggests that success in improving quality, reducing spending, and improving appropriate utilization after a shift to greater risk-sharing is far from guaranteed.”
- Markowitz, et. al. “conclude that CPC+ did not improve spending or quality for private-plan enrollees in Michigan, even before accounting for payouts to providers. This analysis adds to existing evidence that CPC+ may cost payers money in the short term, without concomitant improvements to care quality.”
The evidence points to first supporting specific primary care functions that are proven effective, such as sharing raw data, care coordination resources, and learning collaboratives, and then designing payment and accountability to support the functions. (Friedberg, et. al., Berwick, et. al., Pandey, et.al.)
The literature also highlights the need for monitoring of unintended, but predicable, consequences such as stinting on care (Ubl). Especially troubling is evidence of reduced primary care utilization, including in Medicaid ACO programs (Rosenthal, et. al., McConnell, et. al.). This is especially troubling as improving access to primary care is a prime goal of primary care reform across all stakeholders. While moving away from paying fees for each service to per-patient payments could theoretically reduce unnecessary office visits, it was expected that those new care openings would be filled by patients with unmet needs. This would result in no change in utilization but better access to care for more patients. As that didn’t happen, it is very possible that de-linking payment from service delivery and paying regardless of whether patients got care led to reduced access.
In a related concern, Catel, et. al. urges carefully designed and monitored risk adjustment systems to avoid “perverse incentives” for providers to select more lucrative patients to drive profits.