Accuracy of clinical diagnosis in a Canadian teaching hospital.

Two hundred autopsies were investigated to determine the correlation between the clinical and pathological diagnoses in three categories--major underlying disease, cause of death and significant incidental pulmonary findings. There was concurrence in diagnosis of the major underlying disease in 76% of cases, with 12% of disagreements being considered minor and 12% major. In only three cases might different management have affected the outcome had the correct diagnosis of the major underlying disease been made during life. There was concurrence of the diagnosis of the cause of death (which was often different from the underlying disease) in 64% of cases, and in 10% of cases the outcome might have been different had the clinical diagnosis been accurate. The clinical opinion that lung disease was the cause of death was confirmed at autopsy in 54% of cases, and 45% of the pulmonary causes of death as determined at autopsy had been recognized clinically. Major incidental pulmonary findings diagnosed clinically were confirmed in 76% of cases, and major pulmonary findings diagnosed at autopsy had been recognized clinically in 83%. The major sources of these discrepancies were pulmonary embolism and pneumonia. If autopsies are to play a role in patient management, clinicians will have to be made aware of discrepancies between clinical and autopsy diagnosis. The real test of efficacy would be modification of patient management for the good.