A model for evaluating interface terminologies.

OBJECTIVE: Evaluations of individual terminology systems should be driven in part by the intended usages of such systems. Clinical interface terminologies support interactions between healthcare providers and computer-based applications. They aid practitioners in converting clinical "free text" thoughts into the structured, formal data representations used internally by application programs. Interface terminologies also serve the important role of presenting existing stored, encoded data to end users in human-understandable and actionable formats. The authors present a model for evaluating functional utility of interface terminologies based on these intended uses.
DESIGN: Specific parameters defined in the manuscript comprise the metrics for the evaluation model. MEASUREMENTS: Parameters include concept accuracy, term expressivity, degree of semantic consistency for term construction and selection, adequacy of assertional knowledge supporting concepts, degree of complexity of pre-coordinated concepts, and the "human readability" of the terminology. The fundamental metric is how well the interface terminology performs in supporting correct, complete, and efficient data encoding or review by humans.
RESULTS: Authors provide examples demonstrating performance of the proposed evaluation model in selected instances.
CONCLUSION: A formal evaluation model will permit investigators to evaluate interface terminologies using a consistent and principled approach. Terminology developers and evaluators can apply the proposed model to identify areas for improving interface terminologies.