Impact of antibiotic treatment duration on outcomes in older men with suspected urinary tract infection: retrospective cohort study

Ahmed, Haroon, Farewell, Daniel, Francis, Nick A., Paranjothy, Shantini and Butler, Christopher C. (2019) Impact of antibiotic treatment duration on outcomes in older men with suspected urinary tract infection: retrospective cohort study. Pharmacoepidemiology and Drug Safety, 28 (6), 857-866. (doi:10.1002/pds.4791).

Abstract

Purpose: clinical guidelines recommend at least 7 days of antibiotic treatment for older men with urinary tract infection (UTI). There may be potential benefits for patients, health services, and antimicrobial stewardship if shorter antibiotic treatment resulted in similar outcomes. We aimed to determine if treatment duration could be reduced by estimating risk of adverse outcomes according to different prescription durations.

Methods: this retrospective cohort study included men aged greater than or equal to 65 years old with a suspected UTI. We compared outcomes in men prescribed 3, 5, 7, and 8 to 14 days of antibiotic treatment in a multivariable logistic regression analysis and 3 versus 7 days in a propensity?score matched analysis. Our outcomes were reconsultation and represcription (proxy for treatment failure), hospitalisation for UTI, sepsis, or acute kidney injury (AKI), and death.

Results: of 360 640 men aged greater than or equal to 65 years, 33 745 (9.4 had a UTI. Compared with 7 days, men prescribed 3?day treatment had greater odds of reconsultation and represcription (adjusted OR 1.48; 95% CI, 1.25?1.74) but lower odds of AKI hospitalisation (adjusted OR 0.66; 95% CI, 0.45?0.97). We estimated that treating 150 older men with 3 days instead of 7 days of antibiotics could result in four extra reconsultation and represcriptions and one less AKI hospitalisation. We estimated annual prescription cost savings at around pounds2.2 million.

Conclusions: antibiotic treatment for older men with suspected UTI could be reduced to 3 days, albeit with a small increase in risk of treatment failure. A definitive randomised trial is urgently needed.

This record has no associated files available for download.

Contributors

Author: Nick A. Francis

Download statistics