Population and setting
Data collection, data quality assurance procedures and standard operating procedures used for the individual patient database are reported elseswhere.24Briefly, 150 variables (i.e., maternal sociodemographic characteristics, risk factors, process indicators, maternal and neonatal outcomes) were collected for each birth on two wards of the University Obstetric Unit at De Soysa Teaching Hospital for Women, using a standardised two-page form, and entered in real time in an electronic database. De Soysa is the largest referral hospital for maternity care in Sri Lanka and all deliveries occurring in these two wards from May 2015 to May 2019 were entered in the database and considered for inclusion. Overall data quality was routinely monitored with external independent random review of 5% of forms and 5% of entered births to maintain an error rate in data collection below 0.02%.24 Data were also externally monitored for completeness and internal consistency at roughly 4-month intervals.24
We included “low risk women” with singleton pregnancies and a foetus in cephalic presentation whose delivery occurred between 40+0 and 41+6 GW. We excluded all cases with any maternal or foetal characteristics which may have affected outcomes, such as: maternal obesity (Asian criteria-based body mass index -BMI- more than 27.525), previous CS, macrosomia at ultrasonography (defined as estimated birthweight exceeding the 90thcentile for gestational age), hypertension disorders during pregnancy (e.g. pregestational or gestational hypertension, preeclampsia, eclampsia, HELLP syndrome), chorioamnionitis, major foetal malformations, intrauterine growth restriction at ultrasonography (IUGR), small for gestational age (SGA), pre-gestational diabetes, gestational diabetes with the need of drug therapy, maternal cardiac disease, maternal hypothyroidism, polyhydramnios, oligohydramnios, antepartum haemorrhage (APH), major placenta praevia, placental accretism, severe anaemia (haemoglobin <7.0 g/dl) and other foetal and maternal pathological conditions, i.e. systemic lupus erythematosus, pre-pregnancy deep venous thrombosis, epilepsy, suspected cephalo pelvic disproportion, recurrent infection, pancreatitis or glomerulonephritis in pregnancy, chickenpox disease, chronic disease, signs of potentially impaired foetal wellbeing (non-reassuring or pathological cardiotocography, reduced foetal movement, meconium stained amniotic fluid). We also excluded macerated stillbirth from the IOL40 group, as these births are always induced. All women with a reported indication for IOL suggesting the presence of maternal or foetal characteristics described above, such as diabetes, macrosomia at ultrasound, IUGR/SGA, were excluded from the analysis.