AIMS/BACKGROUND: A major medical role for postmortem examinations is the detection of clinically unexpected disease processes contributing to death. The aim of the present study was to determine whether simple clinical parameters can predict the presence of important unanticipated findings at necropsy. METHODS: Prospective audit of adult necropsies carried out in a single year to assess the extent of unexpected findings at necropsy, to compare these cases with non-necropsied deaths to confirm they are a similar population and to seek features that predict which cases have unexpected necropsy findings. RESULTS: No correlation was found between age, sex, duration of in-hospital treatment, surgical intervention, clinical specialty, or necropsy request rates and incidence of unexpected findings in 187 adult necropsies. CONCLUSIONS: No parameters have been identified for patient selection to permit an increase in the yield of clinically unexpected findings. Until there is clear evidence that the current practice of patient selection is anything more than random, an increase in postmortem examination rates, as proposed by the Joint Working Party of the Royal College of Pathologists, the Royal College of Physicians of London and the Royal College of Surgeons of England in their report The Autospy and Audit, will increase the workload without necessarily producing a commensurate gain in knowledge.