Abstract
BACKGROUND: Antibiotic allergy labels are associated with sub-optimal prescribing patterns and poorer clinical outcomes in non-cancer populations, but the effect of labelling on antimicrobial usage in patients with cancer is unknown. FINDINGS: A retrospective review of hospitalized patients admitted to the Peter MacCallum Cancer Centre (2010-2012) identified 23 % of cancer patients (n = 198) with an antimicrobial allergy label (AA). Comparison of those with an antimicrobial allergy label to those without demonstrated increased antibiotic use per admission (3 vs. 2, p = 0.01), increased fluoroquinolone use (11 % vs. 6 %, p < 0.05), increased antibiotic course duration (15 vs. 13 days, p = 0.09), higher readmission rates (53 % vs. 28 %, p < 0.001) and poorer concordance with prescribing guidelines (47 % vs. 91 %, p < 0.001). Patients in the AA group on multivariate analysis had a higher number of antibiotics employed, longer duration of antibiotic therapy and higher rate of readmission. CONCLUSIONS: Antimicrobial usage, including the use of restricted antibiotics, is higher in patients with cancer. Antibiotic de-labelling strategies in cancer patients must be evaluated to aid antimicrobial stewardship initiatives.