Improving decisionmaking processes with the fuzzy logic approach in the epidemiology of sleep disorders

Epidemiological studies can provide information not only on specific diagno stic entities but also on their underlying symptomatic constellations. For this purpose, an expert system was developed for the assessment of sleep di sorders and endowed with the fuzzy logic capabilities necessary to determin e the degree to which a given symptom corresponds to a specific diagnosis. Uncertainty is inherent in fields such as sleep medicine and psychiatry, an d becomes evident in clinical practice at the stages of data collection and diagnostic formulation, when the clinician must determine whether a sympto m is present and must choose from several diagnostic possibilities. The pro cess involves a considerable degree of subjectivity on the part of the pati ent in trying to describe his or her symptoms; and of the clinician whose f inal diagnosis will depend on his or her clinical experience and interpreta tion of what is normal and what is pathological. Inferential models of the probabilistic or fuzzy logic type take into account such uncertainty. The S leep-Eval system has been used in epidemiological and clinical studies invo lving 34,044 interviews collected by close to 300 interviewers. The diagnos tic potential of these models is illustrated using data collected in an epi demiological study of the: noninstitutionalized general population of Italy and underlines the advantages and limits of the binary, bayesian, and fuzz y logic methods and analyses. (C) 1999 Elsevier Science Inc.