BACKGROUND: In many low and medium human development index countries, the rate of maternal and neonatal morbidity and mortality is high. One factor which may influence this is the decision-to-delivery interval of emergency cesarean section. We aimed to investigate the maternal risk factors, indications and decision-to-delivery interval of emergency cesarean section in a large, under-resourced obstetric setting in Uganda. METHODS: Records of 344 singleton pregnancies delivered at ≥24 weeks throughout June 2017 at Mulago National Referral Hospital were analysed using Cox proportional hazards models and multivariate logistic regression models. RESULTS: An emergency cesarean section was performed every 104 min and the median decision-to-delivery interval was 5.5 h. Longer interval was associated with preeclampsia and premature rupture of membranes/oligohydramnios. Fetal distress was associated with a shorter interval (p < 0.001). There was no association between decision-to-delivery interval and adverse perinatal outcomes (p > 0.05). Mothers waited on average 6 h longer for deliveries between 00:00-08:00 compared to those between 12:00-20:00 (p < 0.01). The risk of perinatal death was higher in neonates where the decision to deliver was made between 20:00-02:00 compared to 08:00-12:00 (p < 0.01). CONCLUSION: In this setting, the average decision-to-delivery interval is longer than targets adopted in high development index countries. Decision-to-delivery interval varies diurnally, with decisions and deliveries made at night carrying a higher risk of adverse perinatal outcomes. This suggests a need for targeting the improvement of service provision overnight.