Risk of emergency hospitalisation and survival outcomes following adjuvant chemotherapy for early breast cancer in New South Wales, Australia.

Abstract

To examine risk of emergency hospital admission and survival following adjuvant chemotherapy for early breast cancer. Linked data from New South Wales population-based and clinical cancer registries (2008-2012), hospital admissions, official death records and pharmaceutical benefit claims. Women aged ≥18years receiving adjuvant chemotherapy for early-stage operable breast cancer in NSW public hospitals were included. Odds ratios (OR) for emergency hospitalisation within 6months following chemotherapy initiation were estimated using logistic regression and survival using Kaplan-Meier and Cox proportional hazards methods. A total of 3,950 women were included and 30.6% were hospitalised. The most common principal diagnosis at admission was neutropenia (30.8%). Women receiving docetaxel/carboplatin/trastuzumab (TCH) and docetaxel/cyclophosphamide (TC) were the most frequently hospitalised. After adjustment for demographic and clinical factors, the increased risk of hospitalisation for TCH and TC remained compared with doxorubicin/cyclophosphamide 3-weekly (OR 1.71, 95% confidence interval [CI] 1.24-2.37 and OR 1.47, 95% CI 1.17-1.85 respectively). Five-year overall survival was similar for women who were (92.2%, 95% CI 90.7-93.8) and were not hospitalised (93.1%, 95% CI 92.1-94.1). Emergency hospitalisations following chemotherapy for early breast cancer were relatively common, especially following docetaxel-containing protocols. Further examination of reasons for admission is needed to inform actions to improve patient safety.