The value of the hospital autopsy. A study of causes and modes of death estimated before and after autopsy

Abstract

Among 312 consecutive deaths in a Danish Central Hospital autopsy was performed in the pathology department on 266 cases, i.e. 85%. Retrospectively, the underlying causes of death were estimated from the clinical information alone by an experienced clinician and subsequently compared with the autopsy report. The definite cause of death was determined jointly by the clinician and the pathologist. The clinician's diagnosis was thereby confirmed as incorrect in 18% of the cases if small differences in site and type of malignant tumours were not considered. This is less than in many other investigations, but it is stressed that this could partly be because formal errors in completing the death certificate were avoided. The main causes of death were ischaemic heart disease and neoplasia. Clinical diagnosis of malignant diseases was never found to be erroneous. There was a slight tendency to clinically overestimate ischaemic heart disease, but in general the different errors outweighed each other, so that the total number of different causes of death before and after autopsy was nearly the same. The original death certificate was investigated in 12 accidental cases. Hereby it was found that the mode of death was originally stated erroneously as natural in 7 cases, i.e. 4.5%. It is concluded that hospital autopsy is still needed for the control and correction of causes of death, and it is stressed that clinicians as well as pathologists should be more aware of cases with a trauma in the history to avoid errors in the mode of death. Such errors can imply legal as well as insurance problems.