Trends in ischemic heart disease mortality are of current interest because death rates have declined rapidly in the United States, Australia, and some other countries during the last decade, although they have risen elsewhere. There is a need to assess the quality of the data from which such trends are calculated and to consider what information will be required to monitor accurately future patterns of incidence and mortality for this disease. This paper describes two investigations of routinely collected mortality data for ischemic heart disease in Australia. First, information obtained from a heart attack surveillance study was compared with officially recorded causes of death for a defined population during 1979. Second, mortality trends from 1969 to 1978 were compared between the major Australian states. Both studies showed that mortality data were accurate to within about 10%. However, for acute myocardial infarction and other subcategories of ischemic heart disease, the data were neither valid nor reliable. New diagnostic and coding practices are required before it will be possible to monitor accurately long-term trends in subcategories of ischemic heart disease.