Evaluation of CMS’s Oncology Care Model

The Centers for Medicare & Medicaid Services’ Oncology Care Model (OCM) included more than 4,500 oncologists across 33 states, serving more than 600,000 people. Abt Global led a team that evaluated OCM’s ability to lower care costs while maintaining care quality.  

OCM reduced episode payments by an average of 2.1 percent, with reductions notably increasing in the last two years of the program.  Reductions were found only for higher-risk cancer types, which also accounted for 67 percent of all OCM episodes. More specifically, reductions were concentrated in treatment of high-risk breast cancer, lung cancer, colorectal cancer, and lymphoma. Most reductions were attributable to reductions in spending on non-chemotherapy drugs. Although Part B chemotherapy and Part D drug spending (predominantly oral chemotherapy medications) account for the bulk of episode payments, OCM did not generate reductions in spending for such care.  

Ultimately, OCM resulted in net losses for Medicare. Losses were lower in the last two years than in prior periods, with the model nearly breaking even in the final performance period. The greater reductions in the last two years suggest that it takes time for changes to be fully implemented, while the reductions by specific cancer types highlight the fact that opportunities for reductions may vary across cancers. That higher-risk episodes broke even in the most recent OCM performance periods indicates future models may achieve net reductions.  

The goal of OCM was to incentivize the use of higher-value treatment alternatives and better adherence to clinical guidelines. We found the model did increase the use of higher-value supportive care therapies to prevent neutropenia (a very low white blood cell count) and cancer-related bone fractures. These changes accounted for nearly one-third of the reductions in payments attributable to OCM. OCM was also associated with adoption of three higher-value biosimilar anti-cancer treatments, which also contributed to reductions in episode payments.  

To explore the potential impact of OCM on health equity, we assessed four historically underserved populations: patients who were Black, Hispanic, had dual eligibility for Medicare and Medicaid, or lived in significantly resource-poor neighborhoods. We found that, prior to OCM, patients from historically underserved populations had higher acute-care use and episode payments but were less likely to have timely initiation of chemotherapy after surgery, adhere to oral treatment, or receive hospice care. While OCM improved adherence to oral treatment for all four populations, there was no evidence of consistently improved quality of care. In terms of reductions of payments, only the treatment of Hispanic patients registered improvements.  

Overall, OCM did not achieve CMS’s goals of net savings or improved care quality. While participating practices did change their procedures, those changes did not demonstrate meaningful improvements in quality as we measured it, relative to a comparison group. Additionally, the estimated reductions in episode payments did not cover the cost of Model incentives. 

However, OCM holds promise for forthcoming models. For example, reductions in episode payments increased substantially over time such that OCM had nearly broken even by the last performance period. Lessons learned under OCM may allow these types of reductions to occur earlier in future programs. 

• OCM At-A-Glance Report
• OCM Final Evaluation Report Executive Summary
• OCM - Final Evaluation Report Appendices
• Cancer Care Experiences Among People Covered by Medicare