Presentation (CTI)

OBJECTIVES: Geographic distance-to-care may impact cancer outcomes by influencing access to treatment and support, especially in community settings. However, existing evidence is mixed, potentially due to mitigators like telehealth or methodological limitations. To address this gap, we developed a structured framework for incorporating distance-to-care metrics into outcomes research and applied it to metastatic breast cancer (mBC) patients.
METHODS: A literature review identified common approaches and gaps, which was then used to develop a conceptual framework to guide study design. Using ONCare Alliance electronic medical records, we applied it to a real-world cohort of mBC patients, assessing associations between distance and outcomes via descriptive and survival analyses.
RESULTS: We identified four framework domains: 1) measurement (e.g., straight-line vs. travel time; primary vs. satellite clinics), 2) categorization (e.g., data-driven vs. clinically meaningful thresholds; non-linearity), 3) outcomes (e.g., clinical outcomes vs. treatment patterns vs. resource use) and 4) analytics (e.g., confounding). Applied to 576 mBC patients (mean age 63.1 years; 97.6% female; 78.3% White), the median distance was 12.3 miles (range: 0-968.8), with 519 (90.1%) residing within 50 miles. Distance distributions showed clustering and outliers. Iterative application revealed that finding statistically significant relationships varied by outcome (overall survival [OS], time to treatment initiation, visit frequency, systemic therapy receipt) and distance cutoffs. For example, using quartiles, there was a significant relationship with number of office visits within 90 days (p=0.0063) but no relationship with OS (p=0.2082) or likelihood of initiating systemic treatment (odds ratio for farthest vs. nearest quartile = 0.77 [95% CI: 0.27-2.19]), and patterns appeared non-linear. Furthermore, distance correlated strongly with region (p<0.0001), insurance (p<0.0268), and race (p<0.0001), highlighting confounding risks.
CONCLUSIONS: Associations between distance and outcomes are highly sensitive, but the proposed framework provides practical guidance. Systematic application of this framework can help avoid biased conclusions, improve comparability across studies, and support equity-focused policy.