Abstract C019: The mediating effect of hormone therapy compliance in the relationship between race and survival among breast cancer patients in South Carolina: A causal mediation analysis

Introduction: Among women, breast cancer is the most common type of cancer diagnosis. Hormone-receptor positive (HRP) cancers compose approximately 80% of all breast cancer diagnoses and are commonly treated with hormone therapies. An estimated 50% of women do not comply with their prescribed hormone therapies due to incorrect dosage or discontinued use. Empirical evidence has shown that hormone therapies have a significant effect on cancer prognosis and patient survival. Purpose: This study was conducted to investigate the mediating effect of hormone therapy compliance in relation to race and breast cancer survival. Methods: A retrospective cohort study was conducted on women identified through several administrative databases and the South Carolina Central Cancer Registry (SCCCR). A causal mediation analysis was conducted among women in South Carolina regarding their compliance to prescribed hormone therapy and their survival. Participants in this study were African American (AA) or European American (EA) and less than 65 years old. Counterfactual notation was applied to estimate the controlled direct effect (CDE) and the proportion mediated regarding the relationship between race, hormone therapy compliance, and overall survival. The CDE can be estimated by comparing EA and AA women on their survival while holding the mediator constant as having complied with their hormone therapies. The proportion mediated estimates the mediated effect of compliance on the survival of breast cancer patients in South Carolina. Results: There were a total of 1,188 women from South Carolina who had HRP cancers. Among this cohort, 72% were EA and 28% were AA women. Approximately 87% of the participants were compliant with their prescribed hormone therapy. AA women were more likely to die from breast cancer compared to EA regardless of age. The odds ratio and 95% confidence interval for the CDE was 1.05 (1.02, 1.09). Finally, the proportion mediated through compliance was estimated to be 26%. Conclusion: Hormone therapy compliance has been observed to significantly mediate the relationship between race and breast cancer survival among HRP women in South Carolina. The CDE showed a significant association between race and survival when participants were compliant to hormone therapies. Compliance to therapies and physician recommendations may influence the difference in survival that is commonly observed between these two groups. The controlled direct effect estimations can be useful for future policy evaluation. Policy implications include the implementation of effective means of the timely distribution of information concerning the potential side effects as well as education regarding management strategies for these hormone therapies. Finally, understanding the obstacles of compliance to hormone therapies will inform policymakers towards regulations that minimize the impact of these obstacles. Citation Format: Christian R. Alvarado, Samantha C. Truman, Oluwole Babatunde, Sue P. Heiney, Jessica M. Escareno, Swann Arp Adams. The mediating effect of hormone therapy compliance in the relationship between race and survival among breast cancer patients in South Carolina: A causal mediation analysis [abstract]. In: Proceedings of the Eleventh AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2018 Nov 2-5; New Orleans, LA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2020;29(6 Suppl):Abstract nr C019.