In late April, the Centers for Medicare & Medicaid Services (CMS) announced its latest attempt to promote primary care and encourage the move to a value-based system: the Primary Care First Model. CMS hopes this initiative will not only succeed but also inspire state Medicaid programs and commercial insurers to follow suit and adopt similar payment systems. But success may prove elusive.
Under the Primary Care First Model, primary care physicians will receive additional payments for meeting certain target metrics in treating their Medicare patients with the hope that the improved treatment processes will reduce the patients’ downstream health costs. The model comprises a set of voluntary five-year payment options designed to support delivery of advanced primary care. It is based on the existing Comprehensive Primary Care plus (CPC+) model principles. CMS says these principles are focused on prioritizing the physician-patient relationship, enhancing care for patients who have complex chronic needs or who are seriously ill, reducing administrative burden and financially rewarding improved health outcomes.a
How CMS aims to test the model
The Primary Care First Model will start in 2020 and will be offered in 26 regions, including 22 statewide regions. It focuses on advanced primary care practices that are ready to assume financial risk in exchange for reduced administrative burdens. CMS will initially test two model options: the basic Primary Care First option and a separate track option focused on the seriously ill population (SIP).
The Primary Care First option. This track has a risk-adjusted per-person payment and a flat primary-care visit fee. CMS also will pay practices bonuses amounting to as much as 50% of a practice’s revenue if the practice meets certain target metrics, which reduce hospitalizations and improve patient satisfaction measures. The practices also must provide evidence that patients have adopted advanced care planning practices.
Under the Primary Care First option, physicians assume two-sided risk, meaning they can either gain increased payments or lose money. Physicians in participating practices can lose as much as 10% of their revenue if their patients end up sicker than expected, but they can gain as much as 50% of their revenue if their patients do well.
The SIP option. In addition to being focused on physicians who treat patients with high healthcare needs and those seriously ill, this option also is targeted at physicians who frequently deliver hospice or palliative care services. Under the model, these physicians will receive larger payments that are risk-adjusted for the expected high spending of the seriously ill patients. The model’s success will depend in large part on how well CMS can accurately risk-adjust for the seriously ill patients and for the variability that may occur among patients for particular physicians. Practices that have or attract a sicker population of patients will be placed at added risk. It will be especially important for participating physician practices to have patient populations large enough to enable them to accommodate a few very sick patients, yet this prospect will pose a challenge because physician practices tend to have patient populations much smaller than those served by hospitals.
Expectations for participation
In addition to providing care to patients under either of these options, primary care practices will be required to collect and report data to CMS. Despite this data-reporting requirement and the risks participating primary care physicians will face, CMS expects that 25% of smaller primary care practices will opt to participate in the Primary Care First Model. The model assumes that giving primary care physicians increased flexibility in how they deliver care will enable them to use their superior knowledge of their own patient populations to innovate their care processes. It also assumes that CMS will be able to adequately risk-adjust to account for the increased spending associated with sicker patients and not penalize the physicians caring for these patients.
Outlook for success: Primary Care First compared with CPC+
The Primary Care First model and the CPC+ model have different underlying strategies, which may have a bearing on their ability to succeed. The CPC+ model offers primary care practices increasingly larger per person payments to reflect increasingly sophisticated care practices for complex patients.
It remains an open question whether the Primary Care First Model will be any more successful than either the CPC+ model or its predecessor, the Comprehensive Primary Care (CPC) Initiative. These successive demonstrations appear to be getting results that are moving in the right direction, but they also point to challenges.
The CPC pilot, a four-year study that began in the fall of 2012, was designed to determine how offering care management fees affects the cost and quality of care delivered in 500 primary care practices. The mid-term assessment found that, despite having received a median of $115,000 per clinician in care-management fees over two years, the participating practices did not demonstrate any savings, after netting out the cost of theincentive payment. They had also not shown any appreciable quality improvements in the treatment of their overall patient population. However, care of high-risk diabetes patients saw some appreciable improvement.
Results of the first-year assessment of the CPC+ evaluation, which are all that are currently available, are not any more promising. CPC+ partnered with public and private payers across various sites and included health IT vendors to help the practices use IT tools to improve primarycare. Like the CPC pilot, the amounts paid seem substantial: $88,000 in one of the tracks and $195,000 for primary care physicians in the other track, in addition to their traditional payments.
Although more than 90% of the physicians said the CPC+ model improved quality, few measurable differences in either service use or qualityof- care outcomes were reported. Perhaps the out-years of the CPC+ will show a more positive impact, but to date, there is little evidence that any of CMS’s various primary-care initiatives have actually improved the outcomes of primary care. Thus, CMS’s hopes for success with the Primary Care First Model do not have a strong foundation in precedent. The funding for the Primary Care First Model, as with the CPC+ model, seems sufficient to get primary care physicians’ attention.
But the multiplicity of pilot projects could make them question whether this model will be any more permanent than the pilot projects that preceded it. And the temporary nature of these pilot projects could cause at least some physicians to feel disinclined to change their practice patterns, even though they might need to do soto be prepared for whatever comes next.
a. CMS.gov, “Primary Care First Model options,” page last updated June 4, 2019.