Validating a guidelines based asthma decision support system: step one.

Purpose. To quantify the accuracy of a computerized decision support system in discerning severe asthma in a clinical setting. Design. A total of 69 consecutive asthmatics examined in an asthma clinic were classified as "severe" or "mild" by the computerized decision support system and expert asthma clinicians. The expert asthma clinicians were the reference standard. Results. The accuracy was 91%, the sensitivity 96%, the specificity 73%, the positive predictive value 93%, and the negative predictive value 85%. Conclusions. The asthma decision support system was able to discern "mild" from "severe" asthma in a similar fashion to expert asthma clinicians.