A Systematic Review of Ontologies applied in Clinical Decision Support System Rules (Preprint)

Abstract

Clinical decision support systems (CDSS) play a critical role in the quality and safety of healthcare delivery. Although CDSS rules direct CDSS behavior, CDSS rules have not been routinely shared and reused.

Ontology has the potential to promote the reusing of CDSS rules. Therefore, we systematically screened literature to elaborate on the current status of ontology applied in CDSS rules, e.g., rule management (uses the captured CDSS rule usage data and users’ feedback data to customize CDSS services to be more accurate) and maintenance (updates CDSS rules). We aim to delineate the frontlines of ontologies applied in CDSS rules via this systematic literature review.

The literature search was focused on ontology, clinical decision support, and rules in PubMed, the Association for Computing Machinery (ACM) Digital Library, and the Nursing & Allied Health Database. Grounded theory and PRISMA 2020 guidelines were followed. One author started the screening and literature analysis, and two validated the processes and results. Inclusion and exclusion criteria were developed and refined iteratively.

Among the 81 included publications, CDSS were mainly applied to managing chronic conditions, alerts for medication prescriptions, reminders for immunizations and preventive services, diagnoses, and treatment recommendations. The CDSS rules were presented in Semantic Web Rule Language, Jess, or Jena formats. Despite ontologies being used to supply medical knowledge, CDSS rules, and terminologies, ontologies have not been used in CDSS rule management or facilitate the reusing of CDSS rules.

Ontologies have been used to organize and represent medical knowledge and controlled vocabularies that contribute to CDSS and the content of CDSS rules. So far, there is a lack of reusing CDSS rules. More efforts are needed to improve the reusability and interoperability of CDSS rules. This paper delineated the ontologies, enabling Semantic Web technology, despite inadequate, and their applications in CDSS rules.