Computerized decision support for antimicrobial prescribing: what every antibiotic steward should know.

Area of Science:

• Health Informatics
• Clinical Pharmacy
• Infectious Disease Management

Background:

• Computerized clinical decision support systems (CDSS) offer potential for optimizing antimicrobial prescribing.
• Integration into electronic health records (EHR) and use of machine learning (ML) are advancing CDSS capabilities.
• Current CDSS address prescribing guidance, alerts, resistance prediction, and de-escalation protocols.

Purpose of the Study:

• To evaluate the role of CDSS in antimicrobial stewardship (AMS).
• To identify challenges impacting CDSS effectiveness and implementation in healthcare.

Main Methods:

• A narrative review of existing literature on CDSS in AMS.
• Evaluation of systems in diverse healthcare settings (hospitals, primary care).
• Analysis of stand-alone tools and EHR-integrated systems, including rule-based and ML models.

Main Results:

• CDSS effectiveness is often limited by poor integration into clinical practice, low user engagement, and inadequate system design.
• Challenges include alert fatigue, workflow disruption, and resistance to change.
• Evaluating CDSS is complex due to varied outcomes and methodological limitations.

Conclusions:

• CDSS show promise for improving antimicrobial use but face significant evaluation and implementation hurdles.
• Enhanced integration, usability, and user-centered design are crucial for realizing CDSS potential.
• Rigorous research frameworks are needed for complex healthcare environments to maximize CDSS impact.