Accountability in Value-Based Payment Needs to Go Beyond Primary Care

In a research article published last year in Annals, investigators set out to quantify the variation in readmission rates among primary care physicians (1). Using a retrospective cohort design to analyze 100% of Medicare claims from Texas, the authors found little variation in 30-day readmissions among primary care physicians. The average readmission rate was 12.9%, with 10th (12.4%) and 90th (13.4%) percentile values tightly clustered around that value. Even the 99th percentile value for readmission rates was only 14.0%, just 1.1 percentage points higher than the average. These findings suggest an important lesson as the Centers for Medicare & Medicaid Services continues to drive the nationwide shift toward value-based payment: Accountability in value-based payment reforms needs to go beyond primary care.

The agency has made primary care transformation a top policy priority, implementing value-based payment models, such as Comprehensive Primary Care Plus and Primary Care First, that hold primary care physicians and their practices accountable for the quality and costs of care. Although specifics vary by payment model, each is based on the rationale that a strong primary care foundation can improve the quality of or address health care utilization or spending (that is, improve value).

As an internal medicine physician, I believe in the benefit of strong primary care systems. I have been fortunate to experience how primary care interventions can improve patients’ health. The data are also clear: Gaps in primary care access and functions (for example, care coordination) can lead to care fragmentation and suboptimal care experiences and outcomes. These issues are particularly relevant for more than two thirds of older Medicare beneficiaries with multiple chronic conditions, who account for more than 90% of all spending. Primary care is a central part of a better health care system.

However, their limitations notwithstanding (for example, retrospective design and data from 1 state), analyses like those reported in the Annals article also provide a sober-minded view about what we can reasonably expect to achieve through payment models focused exclusively on primary care. As noted by the Annals authors, “detecting a 1.1–percentage point difference” in readmissions “would require more than 3500 admissions per [primary care physician] per year” (1). The idea of using primary care to reduce readmissions is hard to argue against, but it is also hard to meaningfully operationalize that goal in primary care alone given such little variation. My experience as a health system medical director for payment strategy suggests that this dynamic is not limited to Medicare payment models or readmissions. Similar issues can and often do exist for different payer and patient populations, as well as other performance measures.

One alternative would be to ensure that in value-based payment programs, accountability for readmissions and other measures goes beyond primary care. Only some of the many factors that drive readmissions and other utilization can be linked to and effectively addressed through primary care; others arise from patients’ health care interactions with other clinicians in other care settings. Holding other clinicians (for example, subspecialist physicians) and organizations (for example, acute and postacute facilities) accountable for quality and costs alongside primary care physicians may help value-based payment models achieve their intended benefits. Although not conclusive, it is nonetheless instructive that, to date, payment models that spread accountability across primary and non–primary care clinicians and facilities (for example, accountable care organizations) have demonstrated benefits, whereas primary care–focused models (for example, comprehensive primary care) have largely not.

As we await more data from ongoing and future payment models, policymakers and practitioners would do well to heed data, such as those reported in the Annals article. Although many factors may be at play, one for policymakers to carefully consider is if and how to extend quality and cost accountability beyond primary care.

References

1. Singh S, Goodwin JS, Zhou J, et al. Variation among primary care physicians in 30-day readmissions. Ann Intern Med. 2019;170:749-755. [PMID: 31108502] doi:10.7326/M18-2526