Abstract C004: Associations between county-level implicit racial bias and survival disparities in uterine cancer

Abstract Introduction: Racial disparities in uterine cancer outcomes may, in part, stem from pervasive discrimination and biased attitudes. Aggregate-level implicit bias may gauge discriminatory attitudes against Black populations. Younger patients may be especially vulnerable to such bias due to higher rates of uninsurance and lower access to care relative to older, Medicaid-insured groups. In this study, we examined whether county-level aggregate implicit bias corresponds with survival, by race and age groups, among uterine cancer patients. Methods: This retrospective cohort study included non-Hispanic Black (NHB) and non-Hispanic White (NHW) patients with uterine cancer diagnosed from 2003 to 2019, from 17 Surveillance, Epidemiology, End Results (SEER) registries. We used county-level aggregate estimates of anti-Black implicit bias from Project Implicit, a large-scale, web-based platform that administers the Implicit Association Test. Scores were aggregated by county and year, standardized as Z-scores, and dichotomized: county-years with Z-scores >0 were classified as anti-Black bias (reference), and those ≤0 as pro-Black bias (coded as 1). Implicit bias scores were linked to SEER patients by county year. We used Cox proportional hazards models to estimate age-stratified (≥65 vs. <65) hazard ratios (HRs) and 95% confidence intervals (CIs) for multivariable-adjusted associations between implicit bias (main exposure) and overall survival (outcome). Within the two age strata, multiplicative interaction between race and the binary implicit bias score was tested to examine whether the association between aggregate implicit bias and survival varied by race. Results: A total of 12,774 NHB (median [IQR] age at diagnosis, 62 [56-69] years) and 100,537 NHW (median [IQR] age at diagnosis, 62 [56-70] years) patients with uterine cancer were included. Among patients diagnosed with uterine cancer, survival patterns differed by patient race and the level of anti-Black implicit bias in their surrounding social context, across both age groups (<65, ≥65 years). In Cox proportional hazards models stratified by age and adjusted for demographic, clinical, and treatment factors, NHB patients <65 in pro-Black counties had a 31% lower hazard of death compared with NHB patients in anti-Black counties (estimate= –0.37; HR= 0.69; 95% CI: 0.47, 1.03). Conversely, NHW patients <65 in pro-Black counties exhibited higher hazard compared with NHW patients in anti-Black counties (estimate= 0.21; HR= 1.23; 95% CI: 0.98, 1.53). Among those <65, the coefficient of statistical interaction between race and implicit bias (estimate= –0.48, p= 0.02) indicated that the protective effect of pro-Black environments was specific to NHB patients. Among older patients (≥65 years), we did not observe statistically detectable associations between the exposure and outcome across race strata or interaction estimates. Conclusion: Our findings suggest that the survival disadvantage experienced by <65-year-old NHB uterine cancer patients may be attenuated in more racially inclusive community contexts. Citation Format: Parvati Singh, Caitlin E. Meade, Ashley S. Felix. Associations between county-level implicit racial bias and survival disparities in uterine cancer [abstract]. In: Proceedings of the 18th AACR Conference on the Science of Cancer Health Disparities; 2025 Sep 18-21; Baltimore, MD. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2025;34(9 Suppl):Abstract nr C004.