Development and evaluation of an ontology for guiding appropriate antibiotic prescribing

Purpose. Misuse of antibiotics contributes to the significant global problem of antimicrobial resistance. Addressing appropriate antibiotic prescribing practices is part of the solution. Decision support systems (DSSs) can help promote appropriate antibiotic prescribing and ontologies can make the creation and maintenance of such tools more efficient. The overall goal of the research project was to develop and evaluate an antimicrobial-microorganism ontology for guiding appropriate antibiotic prescribing.;Methods. In Phase I, we used focus group methodology to define the functional requirements for an antibiotic DSS and determine the scope of the antimicrobial-microorganism ontology. In Phase II, we developed and evaluated the ontology. The ontology was developed in six steps and implemented using Protege-OWL. The evaluation studies assessed the accuracy of the ontology, usefulness of the ontology for performing knowledge management tasks related to antibiotic prescribing, and usefulness of the ontology for generating alerts to guide antibiotic prescribing.;Results. Phase I analyses identified the general functional requirements of a computerized antibiotic DSS and scope of ontology development. Community-acquired pneumonia and urinary tract infection were selected as targets for ontology development. The resulting ontology included 199 classes, 10 properties, 1,636 description logic restrictions, and Semantic Web Rule Language rules to create three prescribing alerts: (1) antibiotic-microorganism mismatch alert; (2) medication-allergy alert; and (3) non-recommended empiric therapy alert. The evaluation studies confirmed the accuracy of the ontology, usefulness of the ontology for representing and maintaining antimicrobial treatment knowledge rules, and usefulness of the ontology for generating alerts to guide antibiotic prescribing.;Discussion. This study contributed to the understanding of ontology development and evaluation methods and addressed one knowledge gap related to using ontologies as a DSS component a need for formal ontology evaluation methods to measure the quality of ontologies from the perspective of their intrinsic characteristics or usefulness for a specific task. These methods can be reused to support the development and evaluation of additional ontologies.