# Factors Influencing Antimicrobial Choice and Duration During the Last Month of Life in Hospitalized Patients

## TL;DR

This study examines antimicrobial use in hospitalized patients during their last month of life and finds high usage, often without clear infection or specialist guidance.

## Contribution

The study identifies factors influencing antimicrobial use and duration in end-of-life care, emphasizing the role of palliative care consultation.

## Key findings

- 91% of patients received antimicrobials in the last 4 weeks of life, often with broad-spectrum agents.
- Palliative care consultation was linked to longer antimicrobial-free intervals before death.
- Antimicrobial use was not guided by microbiology results or infection specialist input.

## Abstract

Antimicrobial prescribing in patients approaching end of life is complex. We aimed to describe antimicrobial use in the final 4 weeks of life among hospitalized patients and identify factors associated with antimicrobial consumption.

We conducted a retrospective review of antimicrobial use among inpatients during their last 4 weeks of life in a tertiary Irish hospital. Data collected included antimicrobial agents and duration of therapy, microbiological sampling, infection services input, palliative care services input, patient demographics, frailty scores, and markers of inflammation. Univariate analysis and multivariable linear regression were used to assess factors associated with antimicrobial-free time before death.

Two hundred and fifty-eight patients were included, with 91% receiving antimicrobials during the study period. Only 36% had a presumed infection at the time of admission. Antimicrobial use was characterized by broad-spectrum agents, did not correlate with culture results, and had a median duration (interquartile range) of 10.5 (5–18) days. Palliative care consultation (P = .004) and longer length of stay (P < .0001) were associated with a longer antimicrobial-free interval before death. Within the antimicrobial cohort, 26% developed acute kidney injury. Cessation of antimicrobial therapy often occurred late, with 40% of patients receiving antimicrobial therapy within 24 hours of death.

There was a high burden of antimicrobial use in patients nearing the end of life, characterized by broad-spectrum, empiric therapy, often continued until hours before death. We recommend expansion of antimicrobial surveillance and collaboration between infection and palliative care services in order to optimize and rationalize antimicrobial prescribing in patients approaching end of life.

Antimicrobials were prescribed in the four weeks preceding death in 91% of included patients. Antimicrobial usage was not guided by microbiology results nor by input from infection specialists. Palliative care input was associated with increased antimicrobial-free time prior to death.