How the U.S. health care system pays for health services shapes the care patients receive. Traditional fee-for-service (FFS) payment incentivizes a higher volume of services, often prioritizing quantity over quality. As states seek to promote value-based, equitable care for Medicaid members, primary care population-based payment (PBP) models are one promising approach that have been recommended by the National Academies of Sciences, Engineering, and Medicine and adopted by multiple states.
Primary care PBP models pay primary care providers (PCPs) an upfront, set amount for each patient served and incorporate provider accountability both for quality and cost of care. This payment approach discourages excessive service volume. Instead, it provides predictable “budgets,” offering providers flexibility to deliver more personalized care and support non-billable services that promote well-being (e.g., increased patient communication through phone calls, texting, or the patient portal; some health-related social needs interventions). When designed with an explicit health equity lens, primary care PBP models can be a key part of efforts to reduce health disparities.
Primary Care PBP Models in Medicaid: Six Essential Design Choices
The Center for Health Care Strategies’ (CHCS) recently released resource, Developing Primary Care Population-Based Payment Models in Medicaid: A Primer For States, guides states through key primary care PBP model design choices. While each Medicaid program will make model design choices based on unique state context, the primer outlines options for states to consider in six key areas:
- Model Goals. Clearly defined model goals are an essential component of any primary care PBP model. The goals can present a “north star” for the model — defining the state and stakeholders’ vision and purpose, providing guiding principles for model design choices, and determining how success and failure are measured. Engaging with stakeholders, including Medicaid enrollees and PCPs, is critical for developing goals that address community needs and promote health equity. States should select a limited number of measurable, accountable, and attainable goals for their program.
- Model Scope. The scope of a primary care PBP model includes covered patients, provider types, and services. Medicaid primary care PBP models might focus on Medicaid-covered adults, children and adolescents covered by Medicaid or CHIP, people dually enrolled in Medicare and Medicaid, or some subset of these populations. To determine the types of providers eligible to participate in the model, states can use either a broad or narrow definition of a PCP. Finally, while the core services of primary care should be paid for through the PBP, some services may continue to be paid for through FFS to encourage their provision. The scope of services should reflect the populations served and the care delivery transformation goals supported or required by the model.
- Payment Approach. Primary care PBP models are typically paid on a per member per month (PMPM) basis, but payment approaches vary greatly by model. States must decide whether their PBP approach will be a “full” or “hybrid” PBP. Full PBPs pay for all covered services on a PMPM basis, while hybrid approaches pay for a percentage of expected FFS revenue as a PBP, while the rest is paid via a reduced FFS rate. All primary care PBP models tie payment to quality metrics, typically through a quality withhold or “clawback” of a portion of the capitated payment. Quality measurement and incentives can be tied to health equity using stratified or equity-focused quality measures.
- Patient Attribution. Attribution methodology is a critical design choice for gaining PCP buy-in for the model, as PCPs may not participate if they do not trust that their attributed patients reflect whom they actually care for. Prospective payment in a primary care PBP model typically requires prospective attribution, but other important attribution considerations include: (1) whether attribution occurs at the individual PCP level, practice level, or site of care; (2) how the patient is attributed, such as via patient selection or historical utilization; and (3) how frequently attribution occurs.
- Rate Setting. PBP rates should be fair, equitable, actuarily sound, and incentivize participating PCPs to provide the best possible care. PBP model rates are typically based on historical utilization, though states should be mindful that relying solely on past utilization to set rates may reinforce historic inequities in access to care. During rate development, states can consider boosting the underlying FFS payments for PCPs to promote greater investment in primary care and encourage provider participation. Rates may also be risk adjusted to provide adequate payment for a patient’s care needs, and to avoid incentivizing adverse selection or withholding of needed care. Many states are beginning to investigate social risk adjustment, which can reflect the greater intensity of care that may be needed by patients with health-related social needs, to set more equitable rates.
- Care Delivery Standards. Primary care PBP models often include specific requirements or incentives for improving care delivery, many of which are informed by primary care medical home models. Two popular methods are: (1) creating tiers of standards, which are tied to increased payment levels; and (2) requiring a base set of capabilities that practices must demonstrate to participate in the model. Care delivery redesign is a key opportunity to improve quality and promote health equity in a primary care PBP model by focusing on person-centered, culturally appropriate care.
Helping States Advance Primary Care PBP Models
Primary care PBP models show promise for improving how primary care is financed and delivered by reducing the negative incentives present in FFS payment. When designed with an explicit focus on health equity, these models may be part of a broad strategy to promote equitable health care and outcomes for all Medicaid enrollees. States can use CHCS’ primer to help guide critical choices as they design or refine their primary care PBP models.
What are thoughts about the use of SDOH z-codes as a way to get to standardized data for risk adjustment including SDOH needs/social risk? Are there any active pilots/experiments by states or coalitions of other payers (health plans, employers, etc.) to use z-code data as a source of risk adjustment? I like the potential of validated SDOH tools to capture patient reported information within the EMR (ex: PRAPARE) and consistent application of a short list of relevant z-codes — but I worry about the already heavy documentation/coding burden for providers that seems generally driven by payment motivations more than clinical… Read more »