Stacie Dusetzina is an associate professor of health policy at Vanderbilt University School of Medicine and a member of the 2021 Tradeoffs Research Council. Her research focuses on measuring and evaluating the costs of prescription drugs.

Cancer care costs in the U.S. were around $200 billion in 2020, with much of this spending concentrated among older adults insured by Medicare. That eye-popping price tag is part of why the Centers for Medicare and Medicaid Services (CMS) rolled out the Oncology Care Model (OCM) in 2016. With 3,200 oncologists at 200 practices participating, the OCM is the largest alternative payment model for cancer care. OCM practices agreed to offer cancer patients enhanced care coordination — with tools like 24/7 access to clinicians and individualized care plans — in exchange for a $160 per patient per month payment and the chance to receive bonus payments if they reduced costs while maintaining or improving quality. But has this model cut costs and improved quality for Medicare beneficiaries with cancer?

A study published this month in JAMA, authored by Harvard’s Nancy Keating and an outstanding team of researchers, presented results from the first three years of the OCM. The researchers compared OCM practices with a similar group of non-OCM practices treating similar patients to see if there were differences in spending and a long list of patient outcomes. The study found that average 6-month spending per patient went up around $5,000 in both groups (driven mostly by higher drug spending), and while Medicare spending went up about $300 less in OCM practices, those savings were wiped out by the $704 in average incentive payments those practices received from Medicare. In total, the OCM resulted in net losses to Medicare of $315.6 million over the three years. And the authors found no differences between OCM and non-OCM practices in almost any of the quality outcomes they studied including hospitalization, mortality or patient satisfaction.

The authors did find a few potential pockets of savings that might hint at how future models could be more successful. First, they found savings were greater among beneficiaries with “higher-risk” cancers. The high-risk group excluded beneficiaries with low-risk breast, prostate and bladder cancer for which many people receive lower cost treatments. The authors note that total payments, clinical status and severity, and treatment strategies can vary quite a bit among beneficiaries in the higher-risk group, creating more opportunities to achieve savings through the model. Second, they found the savings in OCM practices were largely generated by a reduction in the use of drugs used to minimize chemotherapy symptoms (as opposed to cancer drugs themselves). The authors argue this finding could suggest clinicians were more comfortable swapping these supportive care drugs out for cheaper alternatives than messing with highly individualized treatment regimens.

There are limitations to this study — primarily that practices chose to join the OCM, so it was not a randomized trial. But that doesn’t do much to limit the disappointment that the OCM didn’t appear to help in the ways that many had hoped. Not only did Medicare lose millions on a program that was supposed to break even or save money, but the substantial increase in spending across the board (driven mostly by higher drug spending) did not improve outcomes and is a cause for concern for future attempts to control cancer costs.