JAMA Netw Open

AI tool fails to cut C. diff infection rates, but shows promise for antimicrobial stewardship

June 16, 2025

While AI didn’t reduce Clostridioides difficile infection (CDI) rates in this low-incidence hospital setting, it effectively supported antimicrobial stewardship. Future success may depend on improving adherence to infection control protocols and integrating AI tools more seamlessly into clinical workflows.

Study details: This prospective, single-center quality improvement study evaluated the impact of an AI-driven infection prevention bundle on CDI rates. The intervention used a validated, institution-specific AI model to identify high-risk inpatients and guide enhanced hand hygiene and antimicrobial stewardship practices. Outcomes were compared across pre- and post-implementation periods spanning 2021 to 2023.

Results: Among over 79,000 hospitalizations, no significant change in CDI incidence was observed post-AI implementation (5.76 vs. 5.65 per 10,000 patient-days; 95% confidence interval [CI] -1.43 to 1.18; P = .85). However, notable reductions in antimicrobial use were seen, particularly for piperacillin-tazobactam (-9.64; 95% CI -12.93 to -6.28; P < .001) and clindamycin (-1.04; 95% CI -1.60 to -0.47; P = .03), with the greatest impact among AI-flagged high-risk patients (relative reduction for piperacillin-tazobactam, 16.8%; 95% CI, 8.0%-24.6%). Staff engagement varied, with pharmacists more responsive to AI alerts than other personnel.

Source:

Tang S, et al. (2025, June 2). JAMA Netw Open. Guiding Clostridioides difficile Infection Prevention Efforts in a Hospital Setting With AI. https://pubmed.ncbi.nlm.nih.gov/40504526/