End of life (EOL) care in head and neck squamous cell carcinoma (HNSCC) compared to other solid tumors (OST) in Washington (WA) State.

11147 Background: Real world data describing EOL care in HNSCC is limited. We performed a retrospective study evaluating EOL in HNSCC vs OST in a population-based sample of patients in WA. Methods: We used a database linking WA state cancer registry records with claims records from Medicare, Medicaid, and two large commercial insurers. Patients with HNSCC (oral cavity, oropharynx, hypopharynx, or larynx) were compared to OST (any solid tumor diagnosis). Adults with AJCC stage II-IV or SEER stage Regional/Distant who died in 2011-2021 with continuous insurance enrollment 6 months before death were included. We compared proportions of patients with >1 ED visits in the last 30 days of life, ICU admission in the last 30 days of life, chemotherapy in the last 14 days of life, and hospice enrollment at least 3 days prior to death. We performed multivariate regression analysis to determine factors associated with hospice enrollment. Results: 1,389 patients with HNSCC and 41,412 patients with OST were identified. Demographics for HNSCC vs OST included median age 68 vs 73, white race 91.5% vs 90.1%, stage IV 57.5% vs 45.3%. Insurance types for HNSCC vs OST were commercial 8.4% vs 9.8%, Medicaid 16.6% vs 8.8%, Medicare 59.1% vs 65.5%, multiple 15.8% vs 15.9% (p<0.0001). HNSCC patients had lower rates of >1 ED visits compared to OST (14.1% vs 16.4%, p=0.02); there was no significant difference in ICU admission (25.1% vs 24.0%, p=0.3) or chemotherapy receipt (6.3% vs 5.1%, p=0.056). Hospice enrollment was significantly lower in HNSCC patients (49.0% vs 56.4%, p<0.0001). A stratified analysis limited to stage IV patients yielded consistent findings in all categories. HNSCC patients died more at home and less in hospice (20.2% vs 12.7% and 41.5% vs 52.0%, p<0.0001). In a multivariate regression analysis (Table) HNSCC, black race, living with a partner, or living in neighborhoods with higher area deprivation index (ADI) was associated with lower hospice enrollment. Patients with Medicare or multiple insurance types were more likely to enroll in hospice; AJCC stage was not significant. Conclusions: In this population-based sample, compared to OST, a greater proportion of HNSCC patients were insured by Medicaid at EOL and were less likely to enroll in hospice prior to death. The reasons driving these observed disparities in hospice enrollment warrant further study to optimize EOL care among patients with HNSCC. [Table: see text]