We obtained data from SEER 18 registries for individuals aged 20 years and older with a first primary malignant brain cancer from 2001 to 2011. Rural/urban residence at diagnosis was defined using both metropolitan/non-metropolitan county classifications and individual RUCC categories. We used Cox proportional hazards regression to compute adjusted hazard ratios (HRs) and 95% confidence intervals (CIs) for the association between rural/urban residence and brain cancer survival.
Among 37,581 cancer cases, 77.9% were non-Hispanic White, 56.5% were male, and 88.7% lived in a metropolitan county. Brain cancer patients living in the most rural counties had a significant increased risk of cancer death compared to those living in the most urban counties (HR 1.15; 95% CI 1.01-1.31). Those living in non-metropolitan counties had a similar risk of cancer death compared to those living in metropolitan counties (HR 1.01; 95% CI 0.97-1.06). Effect modification was observed overall by cancer type, with non-specified oligodendroglioma (HR 1.35; 95% CI 1.01-1.81) showing the greatest effect.
After adjusting for confounding factors, our results suggest that rural residence has a modest effect on brain cancer survival, and that this disparity may vary by cancer type. Future research should explore differences in treatment strategies between rural and urban brain cancer patients.