The long-term influence of adverse social determinants of health on clinical outcomes in breast cancer survivors: Insights from a propensity score–matched cohort study using federated electronic health records.

e12745 Background: Social determinants of health (SDOH), including housing instability, food insecurity, and socioeconomic challenges, are increasingly recognized as modulators of cancer outcomes. However, their long-term effects on breast cancer survivors remain underexplored. This study examines associations between adverse SDOH and clinical outcomes in breast cancer patients using real-world data. Methods: Utilizing the TriNetX federated network, we conducted a retrospective analysis to compare outcomes on adult breast cancer patients (ICD-10-CM: C50) from 88 healthcare organizations. Cohort A (Breast_SDOH; n = 21, 947) comprised patients with documented adverse SDOH (e.g., homelessness [Z59.0], food insecurity [Z59.41]), while Cohort B (Breast_nonSDOH; n = 1, 031, 237) excluded such factors. Propensity score matching (1:1) balanced cohorts (n = 21, 184 each) on demographics, comorbidities, and BMI. Outcomes assessed starting 1-day post-index event (first breast cancer or SDOH diagnosis), included mortality, severe sepsis, lymphedema, pain, malaise/fatigue, memory loss, and angiosarcoma. Analyses included measures of association, Kaplan-Meier survival, and instance counts, with statistical significance at p < 0.05. Results: Adverse SDOH were associated with significantly higher risks and hazards for several outcomes. Mortality risk was elevated (15.9% vs. 13.4%; risk ratio [RR] 1.186, 95% CI 1.132-1.243; p < 0.001), with a hazard ratio (HR) of 1.849 (95% CI 1.755-1.948; p < 0.001). Severe sepsis showed increased risk (4.5% vs. 3.6%; RR 1.248, 95% CI 1.134-1.373; p < 0.001; HR 1.872, 95% CI 1.695-2.068; p < 0.001). Malaise and fatigue (23.1% vs. 22.0%; RR 1.048, 95% CI 1.001-1.097; p = 0.047; HR 1.520, 95% CI 1.441-1.603; p < 0.001) and memory loss (13.5% vs. 11.7%; RR 1.161, 95% CI 1.099-1.228; p < 0.001; HR 1.895, 95% CI 1.783-2.014; p < 0.001) were also more prevalent. Conversely, lymphedema (5.9% vs. 8.3%; RR 0.713, 95% CI 0.663-0.768; p < 0.001) and angiosarcoma (19.1% vs. 38.4%; RR 0.497, 95% CI 0.475-0.521; p < 0.001; HR 0.505, 95% CI 0.479-0.532; p < 0.001) risks were lower, potentially reflecting differential healthcare access or documentation biases. Pain exhibited mixed findings, with marginally lower risk (19.3% vs. 20.4%; RR 0.944, 95% CI 0.899-0.991; p = 0.021) but higher hazard (HR 1.508, 95% CI 1.426-1.594; p < 0.001), suggesting accelerated onset in the SDOH group. Conclusions: Adverse SDOH exacerbate long-term risks of mortality, sepsis, and neurocognitive symptoms in breast cancer survivors, underscoring the need for integrated social support interventions. Lower risks for certain treatment-related outcomes may indicate barriers to care or underreporting. These findings call for SDOH screening and mitigation strategies in oncology to improve equitable outcomes.