EVALUATION OF ATOPY THROUGH AN EXPERT-SYSTEM - DESCRIPTION OF THE DATABASE

Background In order to understand the medical decisions taken during t he initial visit of a new asthmatic patient, a group of experts design ed an expert system which provides conclusions about severity, precipi ting factors and treatment. Rules for atopy and the assessment of alle rgic factors have been discussed and implemented in the expert system. Conclusions about severity have been yet validated using an appropria te methodology. Objective The aim of this study was to investigate a s ample of 471 patients according to conclusions regarding atopy. Method s A total of 471 cases report forms (CRF) was filled in for adult asth matic outpatients, seen for the first time in our clinic without emerg ency situations. Data of each CRF were used by the expert system to dr aw conclusions. The expert system discerns three patterns for atopy, y es, possible or no. The variables known to reflect different features according to the classification of asthma as atopic or not have been s tudied. The variables used in the rules for atopy, obviously linked to the conclusion, were not compared. For many medical problem no unique objective solution exists and this is why a group of patients with po ssible atopy was introduced. Results Patients with atopy had less seve re asthma (P = 0.01), a better FEV(1) value (P = 0.0007) and showed th eir first symptoms of asthma earlier (P = 0.00001) than patients witho ut atopy. Conclusions The characteristics of the group studied here ar e consistent with the literature. This could be considered as an indir ect validation of the expert system. Moreover, patients with possible atopy show intermediate findings for these variables and it is possibl e to suggest a 'dose-effect' relationship.