Predictors of Breast Cancer Outcome in a Cohort of Women Seeking Care at a Safety Net Hospital

Alexander Boyko; Muhammad Mustafa Qureshi; Michael D C Fishman; Priscilla J Slanetz

doi:10.1016/j.acra.2023.11.037

Predictors of Breast Cancer Outcome in a Cohort of Women Seeking Care at a Safety Net Hospital

Acad Radiol. 2023 Dec 12:S1076-6332(23)00672-4. doi: 10.1016/j.acra.2023.11.037. Online ahead of print.

Authors

Alexander Boyko¹, Muhammad Mustafa Qureshi², Michael D C Fishman¹, Priscilla J Slanetz³

Affiliations

¹ Division of Breast Imaging, Department of Radiology, Boston Medical Center, and Boston University Chobanian & Avedisian School of Medicine, 820 Harrison Avenue, Boston, Massachusetts 02118, USA (A.B., M.D.C.F., P.J.S.).
² Department of Radiation Oncology, Boston Medical Center, 830 Harrison Avenue, Boston, Massachusetts 02118, USA (M.M.Q.).
³ Division of Breast Imaging, Department of Radiology, Boston Medical Center, and Boston University Chobanian & Avedisian School of Medicine, 820 Harrison Avenue, Boston, Massachusetts 02118, USA (A.B., M.D.C.F., P.J.S.). Electronic address: priscilla.slanetz@bmc.org.

PMID: 38087721
DOI: 10.1016/j.acra.2023.11.037

Abstract

Rationale and objectives: This study aimed to identify predictors associated with lower mortality in a population of women diagnosed and treated for breast cancer at a safety net hospital.

Methods: From 2008 to 2014, 1115 patients were treated for breast cancer at our academic safety net hospital. 208 were excluded due to diagnosis at an outside facility, and the remaining 907 (81%) formed the study cohort. Retrospective charts and imaging reviews looked at race, ethnicity, insurance status, social determinants of health, screening utilization, treatment regimen, and 7-13-year follow-up care, including the cause of death. Multivariable logistic regression modeling assessed mortality, and adjusted odds ratios (aOR) with 95% confidence intervals (CI) were computed.

Results: Of the 907 women, the mean age was 59 years (inter-quartile range 50-68 years), with 40% White, 46% Black, 4% Asian, and 10% Other. Increasing age (aOR=1.03, p = 0.001) and more advanced stage at diagnosis (aOR=6.37, p < 0.0001) were associated with increased mortality. There was no significant difference in mortality based on race or ethnicity (p > 0.05). Of 494 with screening prior to diagnosis, longer screening time was observed for patients with advanced stage (median 521 days) vs. early stage (median 404 days), p = 0.0004. Patients with Medicaid, insurance not specified, and no insurance were less likely to undergo screening before diagnosis than privately insured (all p < 0.05). Shorter screening time was associated with lower all-cause mortality (aOR=0.57, 95% CI=0.36-0.89, p = 0.013).

Discussion: In a safety net population, a more advanced stage at diagnosis was associated with higher mortality and lower odds of undergoing screening mammography in the two years prior to a breast cancer diagnosis. Early screening was associated with lower mortality. Finally, given no racial or ethnic differences in mortality, the safety net infrastructure at our institution effectively provides equitable cancer care once a cancer is confirmed.

Keywords: Breast cancer; Breast cancer mortality; Health disparities; Health equity; Insurance; Safety net hospital.