[The logic of differential diagnosis].

The methodological aspects of making a differential diagnosis have so far received little attention. The aim of the present study is therefore to analyze the logical structure of this phase of clinical procedure. It is commonly believed that making a differential diagnosis involves an inferential process through which the clinician first makes diagnostic hypotheses, rules out all these prospective hypotheses except for one, and then concludes that the one remaining hypothesis identifies the disease affecting the patient. Some authoritative scientific methodologists sustain that the logical mechanism leading to the rejection of a diagnostic hypothesis is the negation of the consequent, and is thus the falsification proposed by Popper as the fundamental inference in experimental science. In the present study it is therefore suggested that in making a differential diagnosis, clinicians utilize disjoined syllogism rather than a negation of the consequent. A description is given of the structures of the two types of disjoined syllogism, which are inclusive and exclusive. An in-depth analysis is made of the entire diagnostic-differential reasoning process, and the authors suggest that, on a logical plane, it consists of two distinct sequential inferences, one being an eliminative induction and the other an inclusive disjoined syllogism. Lastly, it is pointed out that, as in the clinic, perfect pathognomonic signs are rarely encountered, each inference leading to a rejection of a diagnostic hypothesis can never be apodictic, and can only lead to a probable conclusion. The confutation of diagnostic hypotheses thus appears to be based on the rule of the denial of inductive logic.