Background: A Cochrane Systematic Review of randomised controlled trials of epidural analgesia compared with other or no analgesia in labour reported no overall increased risk of caesarean delivery. However, many trials were affected by substantial non-compliance, and there are concerns about the external validity of some trials for contemporary maternity populations. We aimed to explore the association between epidural analgesia in labour and caesarean delivery in clinical practice and compare with findings from randomised controlled trials.
Methods: Population-based cohort of pregnant women (n = 210 708) without major obstetrical complications who delivered a singleton live infant in hospitals in New South Wales, Australia, 2007-10. Data were obtained from linked, validated population-based data collections. Propensity score matching was used to examine the association between epidural analgesia in labour and caesarean delivery.
Results: Epidural analgesia in labour was used by a third (31.5%, n = 66 317) of the women, and 9.8% (n = 20 531) had a caesarean delivery. Epidural analgesia in labour was associated with increased risk of caesarean delivery {risk ratio [RR] 2.5, [95% confidence interval (CI) 2.5, 2.6]}. The association with epidural analgesia in labour was higher for caesarean delivery for failure to progress {RR 3.0, [95% CI 2.9, 3.0]} than for caesarean delivery for fetal distress {RR 1.9, [95% CI 1.8, 2.0]}.
Conclusions: Epidural analgesia in labour is associated with caesarean delivery in a large maternity population. Population-based studies contribute important data about obstetrical care, when research settings and participants may not represent the clinical settings or broader population in which obstetrical interventions in labour are applied.
Keywords: Epidural analgesia; bias analysis; caesarean delivery; population-based cohort; propensity score matching.
© 2014 John Wiley & Sons Ltd.