Abstract Background: Some breast cancer patients (pts) refused treatment despite their providers’ recommendations. Treatment refusals can be detrimental to these pts’ short- and long-term health outcomes. Limited research has investigated the national trends in treatment refusals and their impacts on the survival of breast cancer pts. Methods: We analyzed data collected from breast cancer pts in the 2004-2020 National Cancer Database (NCDB). Four treatment modalities were assessed: chemotherapy (CT), hormone therapy (HT), radiotherapy (RT), and surgery. The CT cohort included stage I-IV pts. The HT cohort included stage I-IV, hormone receptor-positive pts. The RT or surgery cohort was limited to only stage I-III pts. Refusal status was categorized as “yes/no,” and its correlates were assessed using multivariable logistic regression. Estimated rates of overall survival (OS) were calculated using the Kaplan-Meier method. The association between refusal status and OS was examined using log-rank tests, followed by multivariable Cox regression models. Results: In the CT cohort, 9.6% of 1,296,488 pts who were offered the treatment refused. In the RT cohort, 6.1% of 1,635,916 pts refused. In the HT cohort, 5.0% of 1,893,339 pts refused. In the surgery cohort, only 0.6% of 2,590,963 pts refused. Significant increasing trends in treatment refusals from 2004-2020 across the four treatment modalities were observed (all p-trends < .001). We found significant differences in age, race, AJCC stage group, molecular subtype, tumor grade, and care access indicators (e.g., insurance, median household income, facility type, and rural/urban area) by refusal status. Black pts were more likely than White pts to refuse surgery (adjusted odds ratio [AOR] 2.01, 95% CI: 1.89-2.14). Asian or Pacific Islander pts were also more likely to refuse surgery (AOR 1.29, 95% CI: 1.15-1.44) and CT (AOR 1.21, 95% CI: 1.16-1.27). Uninsured pts were more likely than privately insured pts to refuse surgery (AOR 4.83, 95% CI: 4.22-5.51), RT (AOR 1.97, 95% CI: 1.83-2.12), CT (AOR 1.61, 95% CI: 1.51-1.72), and HT (AOR 1.61, 95% CI: 1.49-1.73). Compared with pts who did not refuse treatment, those who refused had lower rates of 5-year OS in the cohorts of HT (81.4% vs. 88.4%), CT (74.9% vs. 84.4%), RT (74.4% vs. 90.8%), and surgery (42.0% vs. 88.1%). When stratified by stage, similar patterns of 5-year OS rates were observed across all cohorts. After adjusting for sociodemographic and clinicopathologic factors, pts who refused surgeries had a higher mortality risk than those who did not (aHR 2.91, 95% CI: 2.82-3.01). Pts who refused RT had a higher risk of dying than those who did not (aHR 1.97, 95% CI: 1.93-2.01). Pts who refused CT had a greater risk of dying than those who did not (aHR 1.86, 95% CI: 1.83-1.90). Pts who refused HT had a greater risk of death than those who did not (aHR 1.56, 95% CI: 1.53-1.59). Black pts had higher mortality risk than their White counterparts across all cohorts (HT: aHR 1.15, 95% CI: 1.13-1.17; CT: aHR 1.14, 95% CI: 1.12-1.16; RT: aHR 1.11, 95% CI: 1.1.09-1.14; surgery: aHR 1.10, 95% CI: 1.08-1.11). Conclusions: In this sample of breast cancer pts, the rate of treatment refusal was highest for CT and lowest for surgery, and there were significantly increased trends in refusals over time. Age, race, stage, molecular subtype, tumor grade, and care access measures were independently associated with treatment refusals, suggesting that differential refusals not only are affected by biological factors but also may reflect disparities in socioeconomic status. Furthermore, pts who refused treatment experienced worse OS, regardless of treatment modality. These findings suggest that stressing the importance of recommended treatment and interventions tailored for this patient population may be needed to improve their survival outcomes. Citation Format: Jincong Freeman, James Li, Susan Fisher, Katharine Yao, Sean David, Dezheng Huo. Prevalence of Refusal of Recommended Cancer Treatments and Survival Differences in Breast Cancer Patients: Analysis of the National Cancer Database [abstract]. In: Proceedings of the 2023 San Antonio Breast Cancer Symposium; 2023 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2024;84(9 Suppl):Abstract nr PS18-02.
Read full abstract