The use of antibiotics in end-of-life patients remains a controversial practice. Despite their limited clinical effectiveness in this context, these medications are frequently prescribed, often in the absence of an identifiable infectious focus, contributing to adverse events, bacterial resistance, and increased healthcare costs. The scarcity of national data on the subject hampers the development of guidelines to support more appropriate clinical conduct. This study aimed to analyze patterns of antibiotic prescription and factors associated with their use during hospitalization of adult patients who died in a teaching hospital in Southern Brazil.

A cross-sectional, analytical, and retrospective study was conducted with patients (aged ≥ 18 years) who died between January 1 and June 15, 2025, in both critical and non-critical care units. Incomplete medical records were excluded. Data were extracted from electronic records and included clinical, epidemiological, and therapeutic variables, with emphasis on antibiotic use.

A total of 192 deaths were analyzed, of which 106 had end-of-life care records. After applying inclusion criteria, 88 patients were included in the evaluation. The median age was 81.5 years, with a predominance of males (56%). Antibiotics were used in 83% of cases, and 27% had no identified infectious focus. The most prescribed antimicrobials were ceftriaxone, piperacillin-tazobactam, and meropenem. The median duration of use was 8.5 days, and 54% of patients continued antibiotic therapy until death.

The high frequency of antibiotic use in end-of-life patients without an evident infectious focus highlights the need for institutional guidelines and physician training in palliative care. Early implementation of comfort-centered strategies can prevent disproportionate interventions, improve the quality of end-of-life care, and promote the rational use of antimicrobials.