Objective: To determine the accuracy of death certificates in the Australian Capital Territory (ACT) by comparison with autopsy reports.
Design: A retrospective study of 495 deaths occurring from 1979 to 1987, excluding coronial deaths and deaths of infants under one year of age. The main cause of death on the death certificate was compared with the main cause of death recorded on an autopsy-supported death certificate created for the study.
Setting: The deaths occurred in both major institutional hospitals in the ACT. These hospitals are government-funded and administered, catering for both private and public patients.
Patients: There were 495 autopsies recorded in the ACT over the study period. The age data were lost in two cases.
Main outcome measures: To find a simple measure of death certificate accuracy and compare the results with previous work.
Results: The accuracy of death certificates was 77%. Age, sex and length of hospital stay made little difference to the accuracy. Neoplastic diseases were accurately reported in 90% of cases, digestive and cardiovascular diseases in 81%.
Conclusions: There is a need for practical methods and standard criteria to evaluate accuracy of death certificates.