Data source
We obtained data from the Dutch Perinatal Registry (Perined, formerly known as Perinatale Registratie Nederland [PRN] ). The Registry contains information on pregnancy, birth, and maternal and perinatal characteristics and outcomes (Appendix S1) for 97% of all births in the Netherlands (27-29). All characteristics except birth asphyxia, registration of which started only in 2008, were recorded routinely during the entire study period. Perined combines births in primary and hospital care by using a validated linkage method(30, 31).
The Perined dataset did not contain information on the number of previous caesarean sections in women with a parity of two or more. This hampered assessment of vaginal births after one previous caesarean section in addition to one or more vaginal births. Therefore, in order to ensure a homogeneous group of participants we included only para 1 women with one previous caesarean section (n= 144,166; 67.8% of all women in Robson group 5, meaning all women with one or more previous caesarean sections in addition to any number of vaginal births). We excluded women from analysis in whom mode of birth in the current pregnancy was unknown (n=858; 0.6% of Robson group 5 with a parity of one).