A method for estimating the incidence of acute myocardial infarctions (AMI) from routinely collected hospital morbidity records is presented. The method is based on a review of a sample of routinely collected statistics on hospital diagnoses and the subsequent estimation of the "sensitivity" and "positive predictive value" of hospital diagnoses of AMI. The incidence estimate of AMI is a function of the total number of hospital diagnoses of AMI, the "sensitivity", the "positive predictive value", the proportion of cases of AMI treated at home, and the proportion dying prior to hospital admission. Such estimates may be a practical alternative to community registers for monitoring the incidence of AMI.