A TEMPORAL ANALYSIS OF QMR

Objective: To understand better the trade-offs of not incorporating ex plicit time in Quick Medical Reference (QMR), a diagnostic system in t he domain of general internal medicine along the dimensions of express ive power and diagnostic accuracy. Design: The study was conducted in two phases. Phase I was a descriptive analysis of the temporal abstrac tions incorporated in QMR's terms. Phase II was a pseudo-prospective c ontrolled experiment, measuring the effect of history and physical exa mination temporal content on the diagnostic accuracy of QMR. Measureme nts: For each QMR finding that would fit our operational definition of temporal finding, several parameters describing the temporal nature o f the finding were assessed, the most important ones being: temporal p rimitives, time units, temporal uncertainty, processes, and patterns T he history, physical examination, and initial laboratory results of 10 5 consecutive patients admitted to the Pittsburgh University Presbyter ian Hospital were analyzed for temporal content and factors that could potentially influence diagnostic accuracy (these included: rareness o f primary diagnosis, case length, uncertainty, spatial/causal informat ion, and multiple diseases). Results: 776 findings were identified as temporal. The authors developed an ontology describing the terms utili zed by QMR developers to express temporal knowledge. The authors class ified the temporal abstractions found in QMR in 116 temporal types, 11 temporal templates, and a temporal hierarchy. The odds of QMR's makin g a correct diagnosis in high temporal complexity cases is 0.7 the odd s when the temporal complexity is lower, but this result is not statis tically significant (95% confidence interval = 0.27-1.83). Conclusions : QMR contains extensive implicit time modeling. These results support the conclusion that the abstracted encoding of time in the medical kn owledge of QMR does not induce a diagnostic performance penalty.