Antibiotic and nonantibiotic drugs associated with Clostridioides difficile infection risk: a pharmacopoeia-wide case-cohort study

Background — Clostridioides difficile infection (CDI) is principally precipitated by antibiotics, due to their disruption of gut commensal bacteria. The comparative role of nonantibiotic drugs is poorly characterized.

Methods — We examined the contribution of antibiotic and nonantibiotic drugs to CDI risk among residents age >65 years old and not hospitalized in the prior 30 days, between 2018 and 2023. The study used a case-cohort study design, with logistic regression analysis. The case definition consisted of first incident CDI, identified using comprehensive C. difficile testing, hospitalization, and treatment data. Outpatient oral drug exposures were measured in a 1–90-day window prior to case and control days. Adjusted regression models included covariates for age, sex, year and quarter, region, comorbid conditions, healthcare exposures, and drug exposures.

Results — Among 16 196 CDI case patients and 549 831 controls, 335 drugs were included. After adjustment, the antibiotics amoxicillin-clavulanate (odds ratio [OR], 6.05 [95% confidence interval (CI), 5.69–6.43]), clindamycin (16.83 [15.53–18.24]), ciprofloxacin (3.83 [3.59–4.09]), and cephalexin (3.05 [2.86–3.25]), were the largest contributors to CDI risk. Nonantibiotic drugs pantoprazole (OR, 1.33 [95% CI, 1.27–1.39]) and ferrous fumarate (1.71 [1.61–1.82]) were the next largest. Metformin had a protective association (OR, 0.67 [95% CI, .63–.72]). In a meta-regression on a subset of 182 drugs, in vitro anticommensal activity was positively associated with CDI risk (P < .001).

Conclusions — This study provides insights into CDI etiology and avenues for stewardship and drug repurposing to combat CDI and antimicrobial resistance.