Disparities in breast cancer screening: Insights from an adult female outpatient population in the Northeastern United States.

Abstract

e13721 Background: Breast cancer (BC) is the most frequently diagnosed cancer among women globally (excluding non-melanoma skin cancer), accounting for 15.4% of cancer-related deaths. Adherence to United States Preventive Services Task Force (USPSTF) recommendations for biennial screening mammography (BSM) among women remains suboptimal. We aimed to identify barriers to BC screening and characterize healthcare disparities within the adult primary care setting of a large tertiary-care hospital in Hartford, Connecticut. Methods: We conducted a retrospective observational study using Epic (Healthy Planet) electronic medical records of women aged 40-74 years referred for BSM between October 2022 and October 2025. The primary outcome was overdue status for BSM per USPSTF guidelines. Variables associated with healthcare disparities were collected for 1,223 patients and compared by overdue status using Wilcoxon rank-sum and Fisher’s exact tests. False discovery rates were controlled using Benjamini-Hochberg q-values. Multivariable logistic regression was performed to assess associations between risk factors and overdue status. Results: Among 1,223 eligible women, 637 (52.1%) were overdue. In baseline comparisons, overdue status differed significantly by age (median 60 vs 58), race/ethnicity (Hispanic/Latino 64% vs 50.7%), preferred language (English 54.6% vs 68.6%), insurance status (Medicaid 25.1% vs 35.6%), substance use (3.4% vs 10.5%), time since last primary care physician (PCP) visit (<1 year 84.1% vs 59.8%), and personal history of BC (7.8% vs 3.5%) (all q-values <.05). In multivariable analysis, Spanish-speaking and married women were significantly less likely to be overdue, whereas higher odds were associated with Medicaid insurance, substance use, and last PCP visit >1 year ago (all p-values <.05) (Table 1). Conclusions: These findings demonstrate significant sociodemographic and clinical disparities in USPSTF-concordant BC screening, identifying actionable subgroups for targeted outreach. Time since last PCP visit showed the strongest association with overdue screening. Future studies should evaluate whether addressing these barriers improves screening equity and reduces disparities. Multivariable logistic regression for overdue status (reference: up to date). Variables Category (vs reference) OR (95% CI) P Preferred language Spanish vs English 0.47 (0.36, 0.61) <.0001 Marital status Married vs Single 0.69 (0.52, 0.90) 0.007 Insurance Medicaid vs Private 1.73 (1.25, 2.40) 0.001 Substance use Yes vs No 2.35 (1.38, 4.01) 0.002 Last PCP visit >1 year vs ≤1 year 3.94 (2.82, 5.51) <.0001 No visit vs ≤1 year 2.91 (1.87, 4.54) <.0001 OR, odds ratio; CI, confidence interval; P, p-value from Wald test for OR = 1. Note: Multivariable logistic regression was obtained using stepwise selection with entry and retention criteria of α = 0.30 and α = 0.05, respectively.