This retrospective cohort study evaluated the association between antibiotic prescribing and survival in 6,169 outpatient hospice patients enrolled between June 2017 and January 2023, using time-dependent Cox proportional hazards models adjusted for age, sex, Charlson Comorbidity Index, and hospice-qualifying condition (HQC).
Antibiotic use was associated with **increased hazard of death** overall (aHR 1.47; 95% CI 1.37–1.57). The effect was strongest in patients with cancer HQCs (aHR 1.67; 95% CI 1.50–1.87). A survival benefit was seen only in patients with infectious HQCs (aHR 0.60; 95% CI 0.40–0.89). Median time to first antibiotic prescription was 16 days post-enrollment (IQR 3–83).
- Observational design cannot confirm causality; increased mortality may reflect confounding by indication (sicker patients more likely to receive antibiotics). - No data on symptom burden, functional status, or whether antibiotics were prescribed for comfort vs. life-prolonging intent. - Single health system limits generalizability.
For most hospice patients — especially those with cancer — antibiotics are unlikely to prolong life and the symptom burden prompting their use may reflect the dying process itself. Clinicians should counsel patients and families that antibiotics rarely extend survival in this setting, reserving use for those with a clear infectious HQC where a modest benefit was observed.