Formalizing clinical practice guideline for clinical decision support systems

Area of Science:

• Health Informatics
• Clinical Decision Support
• Knowledge Engineering

Background:

• Clinical practice guidelines (CPGs) are vital for healthcare professionals but passive dissemination is ineffective for behavior change.
• Traditional methods of guideline implementation often fail to translate into improved clinical practice.
• A gap exists in efficiently converting CPG knowledge into actionable formats for healthcare systems.

Purpose of the Study:

• To propose and evaluate a framework for active clinical practice guideline dissemination.
• To automate the extraction of clinical knowledge from CPGs into a format suitable for clinical decision support systems (CDSS).
• To reduce the manual effort required by human modelers in formalizing guideline recommendations.

Main Methods:

• Developed a framework for automated knowledge extraction from CPGs.
• The framework converts CPG recommendations into a CDSS-friendly format.
• Evaluated the framework using recommendations from two Scottish Intercollegiate Guidelines Network (SIGN) CPGs: lung cancer and chronic pain.

Main Results:

• The framework successfully automates the extraction of clinical knowledge from CPGs.
• Demonstrated effectiveness in formalizing drug and procedure recommendations within clinical contexts.
• The system aids human modelers by automating parts of the knowledge formalization process.

Conclusions:

• The proposed framework offers an effective solution for active CPG dissemination.
• Automated knowledge extraction can significantly improve the adoption of clinical guidelines.
• This approach facilitates the integration of evidence-based practices into routine healthcare via CDSS.