BACKGROUND: We aimed to evaluate the trend of post-term births over time and their association with perinatal mortality based on prospective pregnancy cohorts in a rural area in Bangladesh. METHODS: This cohort study included 72 373 singleton births with gestational ages ≥28 weeks recorded by a health and demographic surveillance system from 1990 to 2019 in Matlab, Bangladesh. We expressed the gestational age as X (weeks) + Y (days)/7 weeks, where X indicated complete weeks, and Y presented the number of completed days out of seven days or a week. Using Poisson regression with robust variances, we estimated the population-based proportion of post-term births and assessed the association between gestational age categories and perinatal mortality. We presented results by adjusted relative risk (aRR) with 95% confidence intervals (CIs). RESULTS: Post-term births declined from 5.8% in 1990 to 2.8% in 2019. Perinatal mortality declined from 58 to 27 per 1000 births from 1990 to 2019. Compared to full-term births (39 + 0/7 to 40 + 6/7 weeks), the aRRs of perinatal mortality were 1.39 for late-term (41 + 0/7 to 41 + 6/7 weeks) and 1.93 for post-term (≥42 + 0/7 weeks) births. The population-attributable fraction of perinatal mortality was 15% for births at ≥41 + 0/7 weeks out of the total perinatal deaths occurring at ≥39 gestation weeks. CONCLUSIONS: In this rural setting in Bangladesh, we observed a decline in post-term birth proportions from 1990 to 2019. We found increased perinatal mortality when pregnancy continued beyond 40 + 6/7 weeks of gestation. This implies that planning the management of pregnant women approaching the post-term period may be needed to further improve perinatal health outcomes.