Efficacy and safety of a digital clinical decision support system using a C-reactive protein-based algorithm and...

Antibiotic stewardship is crucial to reduce antimicrobial overuse and multidrug-resistant microorganisms. Biomarker-guided strategies, especially with procalcitonin, safely reduce antibiotic exposure, but their use in routine practice remains limited. C-reactive protein (CRP) may offer a similarly effective, more accessible alternative. We evaluated whether integrating CRP into a multifaceted stewardship strategy delivered through a mobile clinical decision support system (CDSS) could safely reduce antibiotic use.

In this single-center, open-label randomized controlled trial, adults with suspected or confirmed bacterial infections who initiated antibiotics within 72 hours were enrolled. The intervention group followed an algorithm combining CRP values with clinical and microbiological criteria, while the control group received evidence-based recommendations emphasizing fixed antibiotic durations. Recommendations were delivered via a smartphone application to the research team. The primary outcome was antibiotic duration for the index infection.

Among 110 patients (median age 67 years; 62% male; median peak CRP 117 mg/L), baseline characteristics were balanced. Protocol adherence was 67%. Antibiotic duration was shorter in the CRP group (median 6.0 vs 7.0 days; p=0.015), with no significant differences in safety outcomes.

A CRP-based strategy implemented through a digital CDSS safely reduced antibiotic duration. Multicenter validation is warranted.