AREA-LEVEL SOCIAL AND STRUCTURAL DETERMINANTS OF HEALTH AND CERVICAL CANCER STAGE AT DIAGNOSIS.

INTRODUCTION: Cervical cancer is highly preventable, yet disparities in stage at diagnosis persist. While individual-level factors affect access to screening and diagnostic care, less is known about how area-level social and structural determinants shape stage at diagnosis. Area-based measures offer a multidimensional approach to capturing structural disadvantage but remain underutilized in cervical cancer research. METHODS: A retrospective cohort study of 352 patients diagnosed with cervical cancer between 2015 and 2023 at Smilow Cancer Hospital Network at Yale was conducted. Residential addresses at diagnosis were geocoded to census tracts and linked to area-level datasets, including ACS poverty estimates, NCI Social Determinants of Health measures, and CDC Social Vulnerability Index. Stage was categorized as early (I-IIA) or advanced (IIB-IVB). Bivariate analyses assessed associations between individual- and area-level characteristics and stage at diagnosis. Multivariable logistic regression models were fit separately for each area-level measure, adjusting for individual-level covariates. RESULTS: Overall, 51.7% of patients were diagnosed with advanced-stage disease. Higher area-level poverty was associated with higher odds of advanced-stage diagnosis (global p = 0.01). Compared with the lowest-poverty neighborhoods, patients in moderate-poverty areas had over twice the odds of advanced disease (aOR 2.19, 95% CI 1.05-4.58), while those in highest-poverty neighborhoods had nearly threefold higher odds (aOR 2.99, 95% CI 1.43-6.24). Higher household composition and disability vulnerability was also independently associated with advanced-stage diagnosis (global p = 0.04). CONCLUSIONS: Neighborhood social and structural disadvantage is independently associated with advanced cervical cancer stage, underscoring the importance of incorporating area-level context into strategies to reduce cervical cancer disparities.