The mortality patterns of coronary heart disease (CHD) vary among countries. Although the mortality rates from this disease are decreasing in the majority of developed nations, most developing countries are experiencing upward trends, especially among men. The mortality trends of acute myocardial infarction (AMI) in Taiwan have not been reported. The purpose of this study was to investigate the effects of age, chronicle period, and birth cohort on the AMI mortality rates using computerized vital statistics data from 1974 through 1993. Compared with the age-adjusted AMI mortality rates in 1974, the mortality rates in 1993 were 2.4 times higher for men and 2.2 times higher for women. When the log-linear Poisson model was used to assess age, birth cohort, and chronicle period simultaneously, women had no elevated relative risk of dying from AMI in different periods. The relative risk of dying from AMI increased 2.5 times for men in the 1955-1959 birth cohort compared with those in the 1890-1894 birth cohort, while the relative risk for females declined by 3.7 times for the corresponding birth cohorts. Based on birth cohort, the trends in mortality differ substantially between sexes in Taiwan. These results may help to generate hypotheses regarding the possible factors, prevention, and control of mortality from AMI in Taiwan.