Disease Classification/Diagnosis Criteria

Abstract

Well formulated diagnostic/classification criteria are important when one deals with diseases of unknown origin especially if specific histologic, radiologic or laboratory findings are not available. On the other hand, the usual separation of classification and diagnostic criteria is ill advised. When one critically analyses the current classification criteria for rheumatic diseases one sees that they always promise diagnostic criteria to come. This, however, almost never happens. The mental/arithmetic process in either case, with its pretest probability, sensitivity and specificity, is identical. A diagnosis, in essence, is nothing more than classification of the individual patient. Another common thought barrier in formulating classification/diagnostic criteria is trying to avoid circularity while criteria making is, by definition, circular. By preparing a criteria set, we carefully build up a construct and then decide whether what we observe fits that construct. Good examples for these thought barriers can be found in the discussions about the preparation and the implementation of classification/diagnostic criteria for Behçet’s disease and rheumatoid arthritis.