Colorectal cancer survival disparities in the five regions of Georgia

Background/objective
The objective of this study was to examine 5-year colorectal cancer survival rates. We also determined whether demographics, tumor characteristics, and treatment modality were associated with 5-year CRC survival in the Clayton, West Central, East Central, Southeast, and Northeast Georgia regions because the significant higher CRC mortality rates in these regions in comparison to the overall rates in the State of Georgia.

Methods
We conducted a retrospective cohort analysis using data from the 1975–2016 Surveillance, Epidemiology, and End Results program aggregated CRC patients to these five regions. Five-year CRC survival was calculated and stratified by the five regions of Georgia, using the Kaplan-Meier method with log-rank test. Cox proportional hazard regression was used to examine the mentioned association in these five regions.

Results
Among 11,023 CRC patients, 5-year CRC survival was lowest in Clayton (65.9%) compared to the West Central (69.0%), East Central (68.2%), Southeast (70.5%), and Northeast regions (69.5%) (p-value = 0.02). In multivariable analysis, greater risk of CRC death was found in the Clayton region compared to the West Central (HR, 1.12; 95%, 1.00–1.25) region when adjusting for demographics, tumor characteristics, and treatment modality. Among Clayton Georgians, age of 75+ years (HR, 2.13; 95%, 1.56–2.89), grade 3 & 4 tumors (HR, 2.22; 95%, 1.64–3.00), and distant stage (HR, 20.95; 95%, 15.99–27.45) were negatively associated with CRC survival.

Conclusion
We observed place-based differences in CRC survival with significantly lower survival rates in the Clayton region. Factors associated with higher risk of CRC death include older age at diagnosis, high-grade tumors, and distant stage CRC among Clayton Georgians. Our study provides important evidence to all relevant stakeholders in furthering the development of culturally tailored CRC screening interventions aimed at CRC early detection and improved outcomes.