BACKGROUND: Real-world data (RWD) are a valuable data source for oncology drug development and regulatory decision-making. Heterogeneity in insurance coverage by payer can impact patient care.

OBJECTIVE: To leverage 2 novel measures of insurance coverage of supportive medications for oncology treatment–related toxicities to characterize access to care in RWD cancer cohorts.

METHODS: Using Tempus AI, Komodo, and Tufts Center for the Evaluation of Value and Risk in Health data, we linked patient-level clinical and health claims data to temporally accurate coverage policies. We evaluated patients with non– small cell lung cancer (NSCLC), colorectal cancer (CRC), and breast cancer (BC) treated with chemotherapy. Evaluable patients had 1 year of continuous insurance coverage after their first chemotherapy treatment. We assessed coverage policies for specialty medications used to treat nausea and vomiting (NV) and neutropenia per National Comprehensive Cancer Network toxicity guidelines in the year following treatment initiation. We used 2 metrics of coverage: the Coverage Restriction Summary Score (range 0-8 measuring restrictions above the US Food and Drug Administration label accounting for restrictions based on patient subgroups, combination therapy, prescriber qualifications, and step therapy) and the Complexity Score (range 0-10 measuring word count, line breaks, documentation requirements, and number of policy documents). A higher score represents higher restriction or complexity. We explore the impact of these scores on receipt of the National Comprehensive Cancer Network–recommended treatment, accounting for toxicity type, using Fisher’s exact test.

RESULTS: The cohort included 9,545 patients (4,211 NSCLC, 2,368 CRC, and 2,966 BC) across 12 payers. We considered 3 medications for NV and 7 medications for neutropenia. Neutropenia treatments had a wider complexity range and SD (0-10, SD=3.0) compared with NV medications (0-8, 2.2), whereas NV medications had a wider restrictiveness range (0-4, 1.2) than neutropenia meds (0-3, 1.0). Complexity, but not restrictiveness, drove a statistically significant difference in receipt of antiemetics (P<0.05 for each medication). Four of the 7 treatments for neutropenia were significantly impacted by complexity, whereas restriction impacted 2 of 7.

CONCLUSIONS: Clinicians are faced with complex and restrictive policies when providing care. We show that policy complexity impacts receipt of guideline-recommended treatments for neutropenia and NV. Future work should explore whether the relationships identified impact clinical outcomes including persistence and survival.

VIEW THE PUBLICATION

VIEW THE POSTER