The aim of the study was (1) to assess the incidence of uterine rupture in West Africa; (2) to identify its risk factors there; (3) to assess their predictiveness. The study (MOMA study) was prospective and population based. Data on a large cohort of pregnant women were collected. Univariable and multivariable analysis was used including stepwise logistic regression. We identified 25 cases of clinically symptomatic uterine rupture in a population of 20 326 pregnant women giving an incidence rate of 1.2 uterine ruptures per 1000 deliveries. Five variables were significantly associated with uterine rupture (in both the univariable and multivariable analyses): uterine scars, malpresentation, limping, cephalopelvic disproportion and high parity (≥7). In conclusion, the incidence of uterine rupture is high in West Africa, even in large cities where essential obstetric care is available and despite the low prevalence of uterine scars. A uterine scar multiplies the risk of uterine rupture by 11. Uterine rupture cannot be predicted from currently known risk factors, including uterine scars. The high case fatality rate (33.3%) and the associated perinatal mortality (52%) bear witness to the absence or inadequacy of health facilities in providing essential obstetric care and to the poor quality of maternal health care, even in major cities.