Background:  Increasing rates of induction have been reported in the UK, the USA, Canada and Australia since the early 1990s; however, there is a lack of population‐based studies on trends and pharmacological management of induction of labour.
Aims:  To determine population trends in induction of labour and predictors of failed induction (in caesarean section, specifically for failure to progress with cervix dilation ≤3 cm).
Methods:  Trends in induction were determined for women in NSW who laboured at ≥ 32 weeks from 1998 to 2007 (N = 739 904). To determine the predictors of failed induction, 92 359 deliveries of live singletons for whom linked birth and hospital data were available (2001–2005) were examined using logistic regression analysis.
Results:  The rate of induction increased over the decade from 25.3 to 29.1%; however, among those induced with prostaglandin alone, it decreased from 33.5 to 23.8%. Oxytocin alone was the most commonly used labour induction agent overall (51%) and in most population subgroups. The predictors of failed induction in both nullipara and multipara included increasing maternal age, pre‐term and post‐term birth and the use of prostaglandin or mechanical methods of induction (neither oxytocin nor prostaglandin).
Conclusions:  The pharmacological agents used for induction of labour have changed over the past decade. An important area for future research is to investigate how the dosage of oxytocin and prostaglandin affects pregnancy outcomes.