Strengths and Limitations
The following are considered limitations of this study. Large clinical databases may be susceptible to data entry errors, however we have demonstrated the accuracy of a portion (2005 to 2009) of this database in the past.17 Some of the fields used in the analysis were derived from multiple other fields, including some which allowed free text entry. This could lead to misclassification of items such as the indication for caesarean delivery or history of previous caesarean delivery which could bias the findings in any direction. Missing data (most commonly for maternal BMI) can lead to bias, but we used multiple imputation to reduce this risk. As the data were sourced from only two institutions in one city our findings may not be generalizable to other settings. Lack of information about the number of previous caesarean deliveries could have led to underestimation of the proportion of caesarean deliveries explained by maternal factors.
We believe this study had several strengths including the use of a large dataset with near complete data for factors such as maternal age and mode of birth which reduces information bias and the use of multiple imputation to control for missing data. Unlike national databases, our institutional database contained sufficient data fields to allow us to subdivide our outcome into indications for caesarean delivery and Robson categories.
Research is needed into the best way to manage labour as demographic profiles of populations change. For example, it is not known if older women or those with higher BMI require longer labours. There is evidence that strategies such as induction of labour in selected women18–22 or ceasing oxytocin in active labour23 can increase vaginal birth rates. The possibility that younger women are more susceptible to caesarean delivery for fetal concerns due to uterine hyperstimulation could lead to exploring intrapartum management options based on maternal age.24 Studies are needed to establish the real value of planned caesarean delivery for previous obstetric anal sphincter injury or fetal concerns identified on antenatal ultrasound.