e18017 Background: The Oncology Care Model (OCM) is Medicare’s first bundled payment program for patients with cancer. Because practices voluntarily enrolled in the OCM in 2016, there may be differences between OCM-participating and non-participating oncologists that impact the OCM’s generalizability. We examine baseline characteristics of OCM participants and markets with high OCM physician participation. Methods: In this cross-sectional study, we identified characteristics of US medical oncologists practicing in 2016 using a national telephone-verified physician database. We linked this data with Dartmouth Atlas and Medicare claims data to identify market characteristics of areas with high OCM participation. We used logistic regression to examine relationships between market characteristics and OCM participation. Results: Of 10428 US medical oncologists, 2605 (24.9%) were listed on an OCM-participating practice’s website. There were no differences in sex or medical training between OCM participants and non-participants, although OCM participants were younger. OCM participants were more likely to be affiliated with large, group, and urban practices that were not part of a health system (Table). Southwest, Southeast, and mid-Atlantic markets had higher OCM participation. Markets with high OCM physician participation had higher specialist density, Hospital Care Intensity Index, and acute care utilization at the end of life (all p < 0.001). Market penetration of Accountable Care Organizations (adjusted odds ratio [aOR] 4.65, 95% CI 3.31-6.56, p < 0.001) and Medicare Advantage (aOR 2.82, 95% CI 1.97-4.06, p < 0.001) were associated with higher OCM participation. Conclusions: We found differences in provider and practice demographics, care intensity, and prior exposure to alternative payment models between OCM-participants and non-participants. Differences in practice and market characteristics influence oncologists’ participation in alternative payment models and should be accounted for in Medicare’s OCM evaluation. [Table: see text]