In recent days there has been a lot of action around CMS’ Comprehensive Primary Care Initiative (CPCI). First, the next phase of the program was announced, expanding the program in size and scope. Several days later, an evaluation of the first two years of the initiative was published in the New England Journal of Medicine.
The original CPCI demonstration began in October 2012 and included 502 practices in seven regions (states or smaller areas within states). The regions were determined largely by payer interest, as commercial and state health insurance plans are essential partners in this multi-payer model. The CPCI involves risk-stratified care management fees for participating practices and the possibility of sharing in net savings to Medicare (if any). In turn, the practices must invest in practice redesign around: access and continuity, chronic disease management, risk-stratified care management, patient and caregiver engagement, and care coordination across a patient’s providers, e.g., managing care transitions and ensuring close communication and collaboration.
Most aspects of the CPCI are familiar to followers of patient-centered medical home (PCMH) models. What set it apart was the size of the care management fees, which ranged as high as $40 per beneficiary per month (PBPM) for the highest-risk patients. In the first two years, practices received a median of $115,000 per clinician in care management fees . This was enough money for practices to really invest in transformation, including robust data collection and quality management, and care coordination.
The results of the two-year evaluation showed modest results – practices had made progress in improving care management for high-risk patients, but quality measures and patient experiences indicators haven’t yet changed, and practices did not show significant savings to Medicare net of care management payments.
These findings came on the heels of CMMI’s announcement of CPCI 2.0, or Comprehensive Primary Care Plus (CPC+). CPC+, beginning in January 2017, doubles down on CPCI principles. It maintains the original model as “track 1”, but also proposes a new option, “track 2”. The new option involves an ambitious investment in primary care transformation, including a departure from fee-for-service in track 2 (CPCI care management fees are layered on top of FFS). CPC+ escalates the care management fees (track 1 average of $15 PBPM, track 2 average of $28 PBPM), adds Medicare performance-based incentive payments which are paid at the beginning of the year but can be clawed back if targets are not met ($2.50 PBPM in track 1, $4 PBPM in track 2). Most significantly, track 2 introduces “CPCP” – Comprehensive Primary Care Payments. Practices still receive a portion of payment as FFS, but have a percentage of their projected E&M reimbursements paid prospectively on a quarterly basis, with a commensurate reduction in reimbursement for claims filed.
CPC+ faces a number of challenges. A first-order question is whether payers other than Medicare will get on board with the model. The CPC models are meant to be multi-payer payment initiatives, which matters since they seek to transform practice operations, and patients can’t easily be sorted into different systems at the point of care. Another potential challenge is that practices applying to CPC+ must have a letter of commitment from their EHR vendor, agreeing to support data collection needs of the demo – those with experience trying to pull data from EHRs will know this can be a deceptively painful and expensive proposition.
Then, there is the open question of whether the model can actually deliver on quality improvement and cost savings – the two-year evaluation is not encouraging on this front, however, I’m not entirely pessimistic yet. It’s worth remembering that most of the health system is still on a learning curve when it comes to value based care. Even large and high-functioning primary care practices have had to do a lot of capacity building in CPCI, including building skills in quality improvement (QI), changing workflows to support chronic disease management and care coordination (for example, forming teams or adding new roles), collecting data (as mentioned, not as easy as it should be) and then using it as an effective tool for QI. I don’t mean to let practices off the hook for the urgent need to improve the quality and value of care, but I think it is possible that the first two years may include a capacity-building phase as practices develop population health management skills.
Finally, there is an argument to be made that CPC+ simply makes primary care whole. Primary care is poorly paid relative to specialties, unlike many other mature health systems. Primary care activities are undervalued in traditional FFS, which assigns more weight to procedures. Some have noted that CPC+ is considerably more generous than the ACO payment model, in terms of supplemental payments and risk (or lack thereof), and wonder, especially in light of lukewarm initial data, what primary care practices’ incentives are. Having spoken with dozens of primary care practices, including some participating in CPCI, I’d argue that the greatest incentive doesn’t come from the Comprehensive Primary Care model itself, but the desire to avoid reverting to the status quo payment structure in primary care (most ACOs are testing new options, not fleeing their current situation). From the perspective of a patient, I hope that primary care will be able to deliver on better outcomes and better value.