Enhancing Primary Care Payments Without Adding Financial Risk

Abstract

The delivery of primary healthcare in the USA is threatened on multiple fronts. To preserve and strengthen this critical part of the healthcare delivery system, a rapid and broadly accepted change in the basic payment strategy is needed. This paper describes the changes in the delivery of primary health services that demand additional population-based funding and the need to provide sufficient funding to sustain direct provider-patient interaction. We additionally describe the merits of a hybrid payment model that continues to include some level of fee-for-service payment and point to the pitfalls of imposing substantial financial risk on primary care practices, particularly small- and medium-sized primary care practices lacking the financial reserves to sustain monetary losses.

The delivery of primary care in the USA is seriously threatened by declining numbers of new providers,¹ low compensation,² an aging workforce,³ high levels of burnout,⁴ and the growing needs of an aging population with significant chronic disease burden⁵ occurring at the same time as the important but unfunded mandate to transform to team-based care.⁶ The costs to practices⁷ to establish team-based care⁸ are considerable and include a high level of fixed staffing and information technology infrastructure expenses that will need to be reflected in payment models.

The recent National Academies of Sciences, Engineering, and Medicine (NASEM) report Implementing High-Quality Primary Care: Rebuilding the Foundation of Health Care describes primary care as a common good that, as such, should be given the necessary funding and support to ensure the health of the population being served.⁹ The first implementation recommendation from the report calls to “pay for primary care teams to care for people, not doctors to deliver services.” The report suggests a hybrid form of payment that marries (1) fee-for-service payments that reward specific patient-provider interactions with (2) prospective population-based payments sufficient to allow practices to have the resources to meet the whole-person needs of the populations they serve. The report emphasizes the importance of appropriate incentives in the hybrid payment system but does not direct policy makers and payers to use substantial financial risk as an incentive for better performance. Nevertheless, current Center for Medicare & Medicaid Innovation sponsored models for primary care payment impose substantial financial risk, generating ongoing controversy among proponents of adding population-based payment in a hybrid payment model for primary care.¹⁰

In this perspective, we first describe the merits of a hybrid payment model that attempts to combine the better aspects of fee-for-service and population-based payment while also addressing mostly overlooked operational challenges in both payment approaches. We then challenge the common assumption that achieving cost savings requires applying substantial financial risk to all types and sizes of primary care practices. We argue that primary care practices will assume some operational and internal performance-based risk under a hybrid payment model but should not be expected to assume insurance risk, which is more appropriately applied to significantly larger populations than those served in small- and medium-sized primary care practices

REFORM OF BASIC PRIMARY CARE PAYMENT

Many proponents of value-based payment have adopted the Health Care Payment Learning & Action Network taxonomy¹¹ that presents a value continuum that assumes fee-for-service is “valueless” and concludes that population-based payment—payment for population served rather than for services rendered—produces the greatest value. We challenge this overly simplistic continuum, believing that every payment model has strengths and weaknesses and that the decision about which payment model to adopt depends on the context of adoption, available design options, and operational challenges. Furthermore, the Learning & Action Network’s value continuum is specifically insensitive to the realities of the contemporary provision of primary care.

It has become commonplace to assert that fee-for-service inherently does not produce value but only volume. However, we hold that how physicians spend their time and what services they provide and order for their patients are integral to whether they are producing high value. The Medicare Physician Fee Schedule (MPFS) produces unnecessarily poor value because of misvalued fees that influence and distort physician provision of care. The MPFS exaggerates the use of procedures, tests, and imaging and shortchanges time spent with and about patients’ care and preferences.¹² However, these distortions do not represent an inevitable characteristic of fee-for-service and can and should be addressed. Better-designed fee schedules need not produce unsustainable service volume, which drives up costs, as evidenced by most developed countries relying on fee-for-service payment for physicians yet generating total per capita costs far below what the USA produces—without demanding financial risk-taking.^13,14

Population-based payment currently enjoys the policy high ground, but during the managed-care backlash in the 1990s, population-based payment—then called capitation—was heavily criticized for possibly leading to stinting on care. Although population-based payment has strong logic in many situations, it is not necessarily a panacea and when applied as global capitation far exceeds the ability of most small- and medium-sized practices to manage. These practices should not be left with their only options being to join larger organizations and/or ACOs in order to be provided sufficient payment to provide the desired level of care or seeking re-insurance for risk-taking. We think preserving small, independent practices that are not expected to take risk should be an important objective of payment reform and can be advanced using with a mix of fee-for-service and population-based payment, which would not involve insurance risk for total cost of care.

Operational challenges are mostly ignored in value-based payment advocacy. Fee-for-service is best suited for procedural, technical services with clear-cut beginnings and endings; codes should be concise, unambiguous, and stable. Yet these requirements are the antithesis of what is needed to support high-quality primary care services, which often lack clear-cut beginnings and endings—especially with the new whole-person, team-based care expectations assigned to primary care, as reflected in the NASEM report. Indeed, evidence shows that already upwards of 25% of the activities primary care clinicians perform are not coded or recognized for payment under fee-for-service-based fee schedules.¹⁵

Accordingly, there is a strong case to also adopt population-based payment for primary care to provide necessary payment flexibility for adoption of care teams, 24/7 patient engagement involving telehealth and other routine forms of communication (email and phone), efforts to tackle social determinants of health and health disparities, and many other activities that defy fee-for-service payment for practical, operational reasons. Furthermore, current fee-for-service payment levels fall far short of supporting the time and effort primary care providers dedicate to providing needed care during face-to-face visits. The office visit payment should approximate the marginal costs to practices of providing an additional visit to neutralize current incentives to increase profits by increasing volume. Finally, it is commonly overlooked that billing costs often exceed fee-for-service payment for high-frequency, low-intensity services. Population-based payment would instead better assure a range of activities, like routine blood pressure monitoring, is performed. These considerations support the NASEM call for hybrid payment for primary care.

SHOULD PRIMARY CARE PHYSICIANS ASSUME FINANCIAL RISK?

Population-based payment inherently involves performance risk, or the obligations the practice directly assumes to care for patients assigned to them—costs that mostly reflect the value of a clinician’s own time and efforts. However, insurance risk (i.e., the total cost of care), which mostly lies outside a physician’s responsibility and control, need not be part of the payment reform package. The recent iterations of these programs for primary care, such as Primary Care First,¹⁶ have placed a substantial portion (up to 10%) of total primary care payments at downside risk. Although the programs provide a potential asymmetric upside bonus, the threat that downside losses could negatively affect the financial stability of small- and medium-sized primary care practices and lead to a disruption in care persists. Primary care consumes approximately 5% of the total cost of care (TCOC) across all payers—even less in Medicare¹⁷—and therefore has only limited ability to control the remaining 95% (e.g., for high hospital, skilled nursing home, and pharmaceutical spending), much of which results from high prices primary care has little or no control over.

Placing the weight of reducing the TCOC on primary care also opens the door to scrimping on care and/or shifting referrals to the lowest-cost providers without regard to the quality of care. Primary care practices should be expected to spend prudently to manage societal resources but focus on services they can directly influence.¹⁸ We are not suggesting that primary care providers do not have an influence on the referral and care patterns of their patients, but only that making the lowest-funded segment of the industry responsible for managing overall costs and holding that segment at financial risk for the performance of the entire healthcare delivery system is an overreach. A concrete operational concern about putting primary care practices at risk for TCOC relates to prices, as alluded to earlier. Recent research shows that prices, not service utilization, are driving health care spending increases; prices now are responsible for about two-thirds of spending increases.¹⁹ Insurers can account for price increases by raising premiums, spreading risk over large populations, negotiating payments, and/or the creation of narrow networks. In contrast, primary care providers when faced with external pricing differentials have limited tools to mitigate the impact under TCOC models. Holding primary care providers at financial risk for price increases beyond their control may overpower providers’ decision-making on the choice of facilities to send their patients without regard to quality. In our view, accountability should be limited to service that primary care practices can directly influence such as avoidable hospital and ED visits, chronic disease outcomes, referral rates, and preventive care.

Financial risk-taking has a place for large healthcare organizations that deliver primary care services and have sufficient financial resources to gamble on meeting benchmark performance without the potential downside loss negatively affecting their staffing or supports for care delivery. This is not the case for most independent primary care practices. Policy makers should avoid payment models that assume all primary care is delivered by large organizations, such that they may inadvertently drive primary care toward consolidation, which could further reduce the independent primary care workforce.

We are unaware of studies demonstrating that substantial financial risk for primary care drives better performance or improved equitable clinical outcomes. On the other hand, some convincing evidence shows that providing bonuses for targeted outcomes does in fact produce positive results in reductions in unnecessary hospital and emergency department utilization, even within the short 5-year CPC+ model.²⁰

CONCLUSION

In designing future primary care payment models, policy makers should be well advised to start by understanding the funding needed to provide high-value care and should avoid payment schemes that favor consolidation and the loss of independent practice while still demanding accountability for outcomes and quality and rewarding superior performance.

The best financial incentives are neutral on the overall cost of providing care and drive toward neither scrimping on care nor providing excessive, often clinically inappropriate care. Bolstering neutral incentives with bonus payments linked to desired and achievable measures will likely produce greater effort and better results than the consequences of imposing substantial financial risk on payments to primary care providers. The NASEM-endorsed hybrid payment model could achieve incentive neutrality, as other countries have achieved,²¹ while affirmatively supporting primary care.

Declarations

Conflict of Interest

Author Howard Haft is employed by and serves as Senior Medical Advisor to the Maryland Department of Health. The opinions expressed here are his personal views and not necessarily those of the Maryland Department of Health.

Author Berenson is a fellow at the Urban Institute. He thanks the Commonwealth Fund for supporting his contribution to the paper.

The authors declare that they do not have a conflict of interest.