OntoPharma: ontology based clinical decision support system to reduce medication prescribing errors.

Area of Science:

• Medical Informatics
• Clinical Pharmacy
• Ontology Engineering

Background:

• Clinical decision support systems (CDSS) can reduce medication errors but face underutilization challenges.
• Ontology-based systems offer an alternative to traditional relational databases for improving CDSS functionality.
• OntoPharma was designed to address medication prescribing errors using an ontology approach.

Purpose of the Study:

• To design and develop OntoPharma, an ontology-based CDSS, aimed at reducing medication prescribing errors.
• To implement and evaluate OntoPharma in a real-world hospital setting.

Main Methods:

• Developed three ontologies covering medicinal product data, drug appropriateness, and local CPOE concepts.
• Implemented an ontology-driven alert module integrated with Computerised Physician Order Entry (CPOE) via a REST API.
• Utilized Protégé-OWL for ontology implementation and SPARQL for querying, with a four-step development process.

Main Results:

• The developed ontologies comprised 34,938 classes, 16,672 individuals, and 82 properties.
• OntoPharma generated 823 alerts across 1046 patients, with 48.7% being accepted.
• Alerts addressed medication dosage, drug-drug interactions, renal adjustments, and allergies.

Conclusions:

• OntoPharma is an effective ontology-based CDSS implemented in clinical practice for identifying medication prescribing errors.
• Multidisciplinary development enhanced user acceptance.
• OntoPharma offers a more intuitive, extensible, and maintainable approach to representing medication knowledge compared to relational database CDSS.