BackgroundThe incidence of second stage cesarean delivery has been rising globally due to the failure or the anticipated difficulty in performing instrumental delivery. Yet the best way to interpret the figure, and its optimal rate remain to be determined. This is because that it is strongly influenced by the practice of other two modes of birth, namely the cesarean performed before reaching the second stage, and the assisted vaginal birth during the second stage. In this regard, a bubble chart which can display three-dimensional data through its X-axis, Y-axis, and the size of each plot (presented as a bubble), may be a suitable method to evaluate the relationship between the rates of these three modes of births. ObjectiveTo conduct an epidemiological study on the incidence of second stage cesarean rate of over 300,000 singleton term births in 10 years from eight obstetric units; and to compare their second stage cesarean rate in relation to their pre-second stage cesarean rates and assisted vaginal birth rates using a bubble chart. Study DesignThe territory-wide birth data between 2009-2018 from all eight public obstetric units (labelled as “A” to “H”) was reviewed. Inclusion criteria were all singleton pregnancies with cephalic presentation delivered at term (≥37weeks). Pre-second stage cesarean rate was defined as all elective cesarean births and those emergency cesarean births occurred before full cervical dilatation, per total number of births. Second stage cesarean rate and assisted vaginal birth rate were calculated according to the respective mode of delivery per the number of cases that reached full cervical dilatation. The rates of these three modes of births were compared between the parity groups and between the eight units. Using a bubble chart, each unit’s second stage cesarean rate (Y-axis) was plotted against its pre-second stage cesarean rate (X-axis) as a bubble. Each unit’s second stage cesarean to assisted vaginal birth ratio was represented by the size of the bubble. ResultsDuring the study period, a total of 353,434 singleton cephalic presenting term pregnancies were delivered in the eight units, and 180,496 (51.1%) were from nulliparous mothers. When compared to the multiparous group, the nulliparous group had a significantly a lower pre-second stage cesarean rate (18.58% vs 21.26%; p<0.001), but a higher second stage cesarean rate (0.79% vs 0.22%; p<0.001) and a higher assisted vaginal birth rate (17.61% vs 3.58%; p<0.001). Using the bubble of their averages as a reference point in the bubble chart, the eight units’ bubbles were clustered into five regions indicating their differences in practice: Unit B and Unit H are close to the average in the center. Unit A and Unit F were at the upper right corner with higher pre-second stage cesarean rate and second stage cesarean rate. Unit D and Unit E are at their opposite. Unit C is at the upper left corner with a low pre-second stage cesarean rate but a high second stage cesarean rate, while Unit G is at its opposite. Unit C and Unit G were also in the extreme in term of pre-second stage cesarean to assisted vaginal birth ratio (0.09 and 0.01 respectively). Although some units had apparently very similar second stage cesarean rate, their obstetric practice were differentiated by the bubble chart. ConclusionThe second stage cesarean rate must be evaluated in the context of the rates of pre-second stage cesarean and assisted vaginal birth. A bubble chart is a useful method for analyzing the relationship between these three variables, so as to differentiate the obstetric practice between different units.