Abstract
Purpose: Because of shared mechanisms such as decreased access to health care, rurality and poverty may act synergistically to decrease colorectal cancer (CRC) survival. Methods: We conducted a retrospective cohort analysis of SEER data (22 registries) with census tract-level measures of poverty/rurality for the period 2006-2015. Multivariable Cox proportional hazard regressions were applied to examine the independent and intersectional associations of persistent poverty and rurality on 5-year cause-specific CRC survival across five racial/ethnic groups. Results: Among 532,868 CRC patients, non-Hispanic Blacks (NHB) demonstrated lower 5-year survival probability (64.2% vs. 68.3% in non-Hispanic Whites [NHW], 66.5% in American Indian/Alaska Natives [AI/AN], 72.1% in Asian/Pacific Islanders, and 68.7% in Hispanic groups) (p-value < 0.001). In adjusted analysis, CRC patients living in rural areas with poverty were at a 1.2-1.6-fold increased risk of CRC death than those who did not live in these areas in five racial/ethnic groups. In particular, AI/AN patients living in rural areas with poverty were 66% more likely to die from CRC (95% CI, 1.32, 2.08). Conclusions: CRC patients who live in rural or poverty areas in SEER areas in the U.S. have a poorer survival compared with those who do not live in such areas regardless of race/ethnicity. Significantly greater risk of CRC death was observed in AI/ANs. Impact: Patient navigators, community education or screening, and other health care system interventions may be helpful to address these disparities by socioeconomic status, race, and geographic residence. Multi-level interventions aimed at institutional racism and medical mistrust may also be helpful.