Medicare's oncology models went from 122 practices to 23 in one generation. The collapse is the most instructive fact about them.
Oncology is where episode payment meets a problem it was not designed for: care that lasts months, costs are dominated by drugs the practice does not control the price of, and clinical variation is legitimate rather than wasteful.
The Oncology Care Model
OCM ran from 2016 to June 30, 2022. Practices “entered into payment arrangements that included financial and performance accountability for episodes of care surrounding chemotherapy administration to cancer patients,” and CMS “also partnered with commercial payers in the model.”
The payment structure had two parts: “a per-beneficiary Monthly Enhanced Oncology Services (MEOS) payment for the duration of the episode and the potential for a performance-based payment for episodes of chemotherapy care.” The MEOS payment was $160 per beneficiary per month and “assisted participating practices in effectively managing and coordinating care for oncology patients during episodes of care.”
At its conclusion, “there were 122 practices and 5 commercial payers participating.”
The Enhancing Oncology Model
EOM is “a nationwide voluntary payment model that incentivizes oncology practices to deliver coordinated, high-quality, patient-centered care for Medicare beneficiaries receiving systemic chemotherapy for seven specific cancers: high-risk breast cancer, lung cancer, chronic leukemia, small intestine/colorectal, lymphoma, multiple myeloma, and high-risk prostate cancer.”
It “began on July 1, 2023 and added a second cohort on July 1, 2025,” with both cohorts ending June 30, 2030. EOM “incentivizes oncology providers to take on accountability for the total cost of care and the quality of care during oncology episodes.”
Current participation: 23 participants and 1 payer.
Reading the drop honestly
| OCM at conclusion | EOM currently | |
|---|---|---|
| Practices or participants | 122 | 23 |
| Participating payers | 5 | 1 |
The obvious explanation is the one the field generally offers: EOM required downside risk from the outset and reduced the monthly enhanced services payment, so practices that had participated under more generous terms declined to continue.
That explanation is plausible and this course cannot confirm it from the sources cited here, because the participation decision is not something CMS publishes reasons for. What can be said with confidence is that a successor model with more risk and less support attracted a small fraction of its predecessor’s participation.
Worth remembering: this is the clearest available illustration of a tension running through the entire curriculum. Voluntary models with generous support attract wide participation and struggle to save money, because the support payments consume the savings. Voluntary models with real risk and less support may save money per participant and attract almost nobody. The CPC+ case study in the introductory course documented the first failure mode. EOM is a live example of the second. A voluntary model has to be attractive enough to join and demanding enough to work, and the space between those constraints is narrower than model designers usually assume. Mandatory models like TEAM exist precisely because that space kept turning out to be empty.
Why oncology is structurally hard
Three features make oncology resist episode payment:
- Drug costs dominate and are largely exogenous. A practice cannot negotiate what a novel therapy costs, and a single new agent can move a benchmark built on prior-year spending.
- Episodes are long and overlapping. Six-month chemotherapy episodes behave more like a population contract than a surgical bundle.
- Appropriate variation is large. Unlike joint replacement, where discharge destination is genuinely discretionary, most oncology spending variation reflects tumor biology and regimen choice.
Drug economics therefore determine oncology results more than care delivery choices do, which is an uncomfortable finding for a model that asks practices to change how they deliver care.
Key takeaways
- OCM paid a $160 monthly enhanced services payment plus performance-based payments, ending June 2022 with 122 practices and 5 payers.
- EOM covers seven cancers, runs to June 2030, and currently has 23 participants and 1 payer.
- The participation collapse illustrates how narrow the viable design space is for voluntary models.
- Oncology resists episode payment because drug costs are exogenous, episodes are long, and variation is often appropriate.
Sources
Check your understanding
What happened to participation between the Oncology Care Model and its successor?
The successor model required downside risk from the start and reduced the monthly enhanced services payment. Practices that had participated under upside-only terms largely did not continue, which is a finding about the price of mandatory risk rather than about oncologists' interest in value.