Physician participation in accountable care organizations (ACOs) reached a milestone in 2018. More than half of physicians reported their practice was part of an ACO, less than a decade after the payment model debuted as part of the Affordable Care Act enacted in 2010.
“Overall, 53.8% of physicians reported participation in at least one ACO type in 2018, up from 44.0% in 2016,” according to a recently released AMA research paper. By ACO type, 38.2% of physicians belonged to a Medicare ACO, 26.3% to a Medicaid ACO, and 39.0% to a commercial ACO. Data was also collected on medical home participation, with 31.9% of physicians reported working in a practice that belonged to one.
This and other related data and analysis was presented in the AMA Policy Research Perspective report, “Payment and Delivery in 2018: Participation in Medical Homes and Accountable Care Organizations on the Rise While Fee-for-Service Revenue Remains Stable.”
Results were based on data from the ongoing Physician Practice Benchmark Survey, conducted by the AMA Division of Economic and Health Policy Research. The division conducts independent research to support AMA federal, state and private sector advocacy.
The research examines physician participation across medical homes and ACOs in terms of practice type, practice specialty mix and practice ownership. Here are some key findings, with comments from the report.
Solo physicians were the least likely to participate. Only 11.1% of these physicians belonged to a medical home, 22.6% to a Medicare ACO, 14.6% to a Medicaid ACO, and 27.2% to a commercial ACO.
“Participation rates for each model were between 6 and 14 percentage points higher for physicians in single specialty practices compared to solo practices—the difference was greatest for medical homes and Medicare ACOs.”
Multispecialty practice physicians overall had the highest participation rates. Among multispecialty practice physicians, 46% belonged to a medical home, 51.6% to a Medicare ACO, 34.9% to a Medicaid ACO, and 49.5% to a commercial ACO.
“Participation rates for each model were between 13 and 22 percentage points higher for physicians in multispecialty practices compared to those in single specialty practices.”
Physicians in practices with at least some primary care doctors had markedly higher participation. “Overall, the data suggest that whether a practice has primary care physicians is a critical factor in the decision to participate in care delivery models such as medical homes and ACOs, a finding which is consistent with literature on this topic.”
Medical home involvement demonstrated the greatest difference, with physician participation of 40.8%—26 percentage points more than for practices without any primary care physicians.
“This gap is smaller but still substantial for Medicare ACOs (18 percentage points), Medicaid ACOs (11 percentage points) and commercial ACOs (12 percentage points).”
Hospital-owned practices had much higher physician participation than doctor-owned practices. Participation was more than 20 percentage points higher in Medicare ACOs and medical homes among physicians in hospital-owned practices, compared with those in physician-owned practices. Participation in Medicaid and commercial ACOs was 18 and 11 percentage points higher, respectively.
The report also examines the proportion of revenue coming from fee-for-service (FFS) and alternative payment methods (APMs) such as pay-for-performance, bundled payments, shared savings and capitation.
“In 2018, 87.0% of physicians reported that their practice received revenue through FFS and 63.1% reported payment from at least one APM,” according to the report, which notes that participation rates for certain APMs rose from 2012 to 2018. “However, in part because many APMs build on FFS rather than replace FFS, roughly 70% of practice revenue still came from FFS in 2018; this share has been stable since the first Benchmark Survey was conducted in 2012.”
The Benchmark Surveys data is drawn from physicians who provide at least 20 hours of patient care per week, are post-residency, and are not employed by the federal government at the time data is collected. The latest survey was conducted in 2018 with 3,500 respondents.