RESUMO
ABSTRACT: Hospital autopsies frequently reveal errors in diagnosis that could have affected the patient's clinical outcome. The aims of this study were (1) to investigate the ability of autopsy at our institution to elucidate unrecognized antemortem diagnoses and (2) to pilot a method for tabulating diagnostic discrepancies on a prospective basis. The study sample consisted of 296 cases from our hybrid hospital/forensic autopsy service during the period 2016 to 2018. Discrepancies in autopsy and clinical diagnosis were reported by pathologists at the time of autopsy report generation using a standard form. The rates of major discrepancies between autopsy and clinical diagnoses were 37.5% for in-hospital cases and 25% for patients who died outside our hospital (P < 0.05). The most common discrepant category was infection. The overall rates of discrepant causes of death were 14% (in hospital) and 8% (out of hospital) (ns). Overall percentages of cases with major diagnostic discrepancies were higher in our study than have been previously reported. It is possible that the nature of our patient population plays a role in this result. This study describes an important prospective reporting tool that will allow us to track rates of medical errors and improve diagnosis and treatment of the critically ill.