Wide Variation In Payments For Medicare Beneficiary Oncology Services Suggests Room For Practice-Level Improvement
- Jeffrey D. Clough ( [email protected] ) was a medical officer, Center for Medicare and Medicaid Innovation, at the Centers for Medicare and Medicaid Services, in Baltimore, Maryland, when this study was conducted. He is now an internal medicine physician in the Department of Medicine at Duke University, in Durham, North Carolina.
- Kavita Patel is managing director for clinical transformation, Engelberg Center for Health Care Reform at Brookings, in Washington, D.C.
- Gerald F. Riley was a senior researcher at the Center for Medicare and Medicaid Innovation when this study was conducted. He is now a senior researcher at Actuarial Research Corporation, in Annandale, Virginia.
- Rahul Rajkumar is acting deputy director of the Center for Medicare and Medicaid Innovation.
- Patrick H. Conway is the deputy administrator for innovation and quality at the Centers for Medicare and Medicaid Services.
- Peter B. Bach is the director of the Center for Health Policy and Outcomes at Memorial Sloan-Kettering Cancer Center, in New York City.
Abstract
In recent years many policy makers have recommended alternative payment models in medical oncology in order to reduce costs and improve patient outcomes. Yet information on how oncology practices differ in their use of key service categories is limited. We measured annual payments for key service categories delivered to fee-for-service Medicare beneficiaries receiving care from 1,534 medical oncology practices in 2011–12. In 2012, differences in payments per beneficiary at the seventy-fifth-percentile practice compared to the twenty-fifth-percentile practice were $3,866 for chemotherapy (including administration and supportive care drugs), $1,872 for acute medical hospitalizations, and $439 for advanced imaging. Supportive care drugs, bevacizumab, and positron-emission tomography accounted for the greatest percentage of variation. Average practice payments for service categories were highly correlated across years but not correlated with each other, which suggests that service categories may be affected by different physician practice characteristics. These differences, even when clinical guidelines exist, demonstrate the potential for quality improvement that could be accelerated through alternative payment models.
