Background Southampton University Hospitals Trust is a tertiary centre with over 6000 deliveries annually. Despite a stable caesarean section (CS) rate of 21–22%, we have previously reported an unexplained increasing trend in forceps deliveries.1 Robson groups can be used to compare CS rates in distinct pregnancy categories.2 A modified Robson Group classification has been used in the coding of all deliveries in our unit since 2004, with subdivision of groups to identify induction of labour from prelabour CS.

Design A review of computerised obstetric data from 2004 to 2009 using modified Robson groups to investigate trends in all types of delivery.

Results The distribution between individual groups has remained extremely consistent over the 6 year period with one-third of all deliveries being in Group A (nulliparous, singleton, cephalic, term, spontaneous labour). There has been no change in the proportion of induced labours between categories. The forceps delivery rate in Group A shows an upward trend from 12 to 17% but a more notable rise to 28% is seen in Group B (nulliparous, singleton, cephalic, term, induced). In Group B, which accounts for 10% of all deliveries, there was a simultaneous fall in ventouse deliveries and decreasing trend in normal vaginal delivery to only 29% but the emergency CS rate remained stable (30–35%). There was no increase in forceps rate (3–4%) in Group G (multiparous, singleton, cephalic, term, induced).

Conclusion We found this modified Robson group classification useful in interpreting obstetric data related to instrumental and normal deliveries as well as CS.