Antibiotic exposure and the risk of hospital-acquired diarrhoea and Clostridioides difficile infection: a cohort study.

Background Hospital-acquired diarrhoea (HAD) and Clostridioides difficile infection (CDI) may be triggered by antibiotic use. Objectives To determine the effect of specific antibiotic agents and duration of therapy on the risk of HAD and CDI. Patients and methods A single-centre retrospective cohort study was conducted between May 2012 and December 2014 in the internal medicine division. HAD was defined based on documentation of diarrhoea in the medical record or an uncancelled C. difficile test in the laboratory database. CDI was diagnosed using a two-step test (initial glutamate dehydrogenase and toxin A/B EIA, with PCR for discrepant results). Outcomes first occurred on hospital Day 4 or later. Treatment with antibiotics and days of therapy were modelled. Results In 29 063 hospitalizations there were 970 HAD events [incidence rate per 10 000 patient days (IR) = 38.5] and 105 CDI events (IR = 3.9). Any antibiotic treatment increased the risk of HAD [adjusted relative risk (aRR) 2.79; 95% CI 2.27-3.43] and CDI (aRR 5.31; 95% CI 2.23-12.69). Each day of β-lactam/β-lactamase inhibitors (βL/βLIs), carbapenems, IV glycopeptides and metronidazole increased the risk of HAD. Each day of βL/βLIs, third- and fourth-generation cephalosporins and carbapenems increased the risk of CDI by over 2%. Conclusions Preventing HAD and CDI should focus on reducing the overall use of antibiotics and shortening antibiotic exposure, rather than focusing on specific agents.