Fuzzy logic for decision support in chronic care

Computerized clinical guidelines can provide significant benefits in terms of health outcomes and costs, however, their effective computer implementat ion presents significant problems. Vagueness and ambiguity inherent in natu ral language (textual) clinical guidelines makes them problematic for formu lating automated alerts or advice. Fuzzy logic allows us to formalize the t reatment of vagueness in a decision support architecture. In care plan on-l ine (CPOL), an intranet-based chronic disease care planning system for gene ral practitioners (GPS) in use in South Australia, we formally treat fuzzin ess in interpretation of quantitative data, formulation of recommendations and unequal importance of clinical indicators. We use expert judgment on ca ses, as well as direct estimates by experts, to optimize aggregation operat ors and treat heterogeneous combinations of conjunction and disjunction tha t are present in the natural language decision rules formulated by speciali st teams. (C) 2001 Elsevier Science B.V All rights reserved.