BACKGROUND: Growing awareness of risk factors for myocardial infarction (MI), continuous mass campaigns on lifestyle factors, increasing use of heart rehabilitation and improved treatment should imply decreases in the incidence, case-fatality and recurrence rates of MI. The purpose of the study was to assess trends in the incidence, case-fatality and recurrence rate of MI and to analyse whether any changes seen were due to a period or a cohort effect. METHODS: The Danish MONICA population comprises all men and women aged 25-74 years in the period 1982-1991 living in a suburban area of Copenhagen, Denmark. Fatal and non-fatal attacks classified as definite MI and non-fatal attacks classified as possible MI were included. The incidence rate was analysed by Poisson regression, the case-fatality rate by logistic regression, and the rate of recurrence by Cox regression. Age-period-cohort analyses were carried out according to a method described by Clayton and Schifflers. RESULTS: During the 10-year period a significant decrease in the incidence rate of MI was seen for men and women and for the rate of recurrent MI. The decrease in incidence and recurrence could not unambiguously be ascribed to a period or cohort effect. The rate of case-fatality after a first MI was not changed significantly during the period, whereas men and women had different trends in case-fatality after recurrent MI. CONCLUSIONS: In accordance with results from other Western countries we found a decline in the incidence and recurrence rate of MI. Contrary, the lack of a decrease in the case-fatality rate after a first MI was unexpected and difficult to explain.