2035 - Differences In Sociodemographic and Treatment Patterns By Medicaid Enrollment Among Patients with Hepatobiliary Cancers
Presenter(s)
J. S. Chiang1, S. Jackson2, K. S. Jin2, K. Achuck3, B. K. Neilsen4, D. Klebaner1, E. Rahimy2, G. S. Ford1, A. Dawes5,6, and E. L. Pollom2; 1Department of Radiation Oncology, Stanford University, Stanford, CA, 2Department of Radiation Oncology, Stanford University School of Medicine, Stanford, CA, 3Thomas Jefferson University, Philadelphia, PA, 4University of Nebraska Medical Center, Omaah, NE, 5Department of Surgery, Stanford University School of Medicine, Stanford, CA, 6Stanford Surgery Policy Improvement Research and Education Center (S-SPIRE), Stanford University School of Medicine, Stanford, CA
Purpose/Objective(s): Medicaid enrollment may mark patients at higher risk for delays in cancer evaluation and treatment, with potential downstream differences in care delivery and outcomes. We compared sociodemographic and treatment patterns among patients with hepatobiliary (HB) cancer with versus without Medicaid enrollment.
Materials/Methods: Using linked population-based cancer registry-Medicaid claims data, we identified patients diagnosed with HB cancers (2016-2019) with or without Medicaid enrollment in the 12 months before diagnosis. The non-Medicaid group included Medicare, private, self-pay, and uninsured patients. Variables included cancer subtype (hepatocellular carcinoma [HCC], cholangiocarcinoma, other), age, sex, race/ethnicity, stage, rurality, neighborhood socioeconomic status tertile (SESTERT; 3=highest), and receipt of radiation (RT; including radioembolization), surgery, chemotherapy (CTX; including TACE), and immunotherapy. Group comparisons used ?² and t-tests. Predictors of NCI-center care were assessed with multivariable logistic regression, and cumulative incidence rates (CIR) of initial treatment were estimated using Fine-Gray methods with death as a competing risk (censored at one year).
Results: Among 16,842 patients, 6,701 (39.8%) and 10,141 (60.2%) patients were Medicaid and non-Medicaid enrolled, respectively. Medicaid enrollees were younger (65.5 vs 67.8 years; p<0.01) and more often Asian/Pacific Islander, African American, and Hispanic, with fewer White patients (p<0.01). HCC was more common (82.3% vs 74.1%) and cholangiocarcinoma less common (9.6% vs 16.7%; p<0.01) among Medicaid enrollees. Stage was similar (p=0.64). Medicaid enrollment was associated with greater representation in the lowest SES tertile (51.2% vs 29.9%; p<0.01). Surgery (19.8% vs 25.4%; p<0.01) and CTX (36.4% vs 41.3%; p<0.01) were less common, while RT was similar (12.4% vs 13.3%; p=0.11). Medicaid enrollees were more likely to be treated at an NCI-designated center (27.0% vs 23.1%; p<0.01). In adjusted logistic modeling, receipt of care at an NCI-designated center was not independently associated with Medicaid enrollment; HCC, metastatic stage, older age, and lower SES associated with lower likelihood of NCI-center care. Medicaid-enrolled patients had delayed treatment initiation (6-month CIR, 51.3% vs 61.7%; p<0.001) and RT initiation (6-month CIR, 10.0% vs. 11.3%; p=0.027).
Conclusion: Medicaid enrollment among patients with HB cancers was associated with younger age, minoritized race/ethnicity, lower neighborhood SES, differences in cancer subtype, and select treatment patterns. Although Medicaid enrollment was not independently associated with receipt of care at an NCI-designated center after adjustment, Medicaid-enrolled patients were delayed in initiating treatment, including RT, compared to non-enrolled patients, highlighting modifiable care pathways and the need for targeted interventions to improve equity in HB cancer care.