The association between residential natural vegetation exposure at diagnosis and colorectal cancer-specific mortality among colorectal cancer cases in Kentucky.

Living near green areas with high vegetation may contribute to an active lifestyle, stress reduction, and enhance social cohesion, and consequently may be associated with improved cancer survival. Epidemiologic studies focused on greenness exposure and cancer-specific mortality, specifically colorectal cancer (CRC)-specific mortality, are sparse. We examined the association between residential greenness at diagnosis and CRC-specific mortality. Using the Kentucky Cancer Registry database, 35,868 primary CRC cases diagnosed between January 2000 and December 2018 were followed until death or November 2020. Residential greenness exposure at diagnosis was estimated using Normalized Difference Vegetation Index (NDVI) with 270m and 1230m buffers. Using Cox models, we estimated adjusted hazard ratios (aHRs) of CRC-specific mortality and 95 % confidence intervals (CIs) by quartiles of greenness. We assessed effect modification using the Likelihood Ratio Test (LRT). We observed 9805 CRC-specific deaths over 208,098 person-years of follow-up. While crude assessments suggested an inverse relationship between NDVI quartiles and CRC-specific mortality, the association was null after adjustment for individual- and neighborhood-level characteristics (270m Buffer aHR Quartile 4 versus Quartile 1: 1.01, 95 % CI: 0.95, 1.08, p-trend = 0.65; 1230m Buffer aHR Quartile 4 versus Quartile 1: 1.03, 95 % CI: 0.96, 1.10, p-trend = 0.26). No effect modification was observed by urban/rural status, race, or neighborhood-level income, while modification by stage was observed for the 270m buffer (LRT p-value = 0.03). Our findings suggest that residing in greener areas was not associated with CRC-specific mortality. Future studies should consider additional greenness measures, greenness interactions, and residential changes over follow-up.