Objective: To study the relationship between epidural analgesia and instrumental and non-elective cesarean delivery in the 1990s in Finland.
Methods: The data was extracted from a nationwide birth register in 1991-1993. To adjust for other factors, hospital-based analyses were made separately for different types of hospitals and in individual-based analyses logistic regression was used.
Results: In the hospital-based analyses there was no systematic correlation between rates of epidural analgesia and operative delivery, but on the individual level there was. Comparison of hospitals with different levels of epidural rates and different types of hospitals suggests that the contradictory findings in the hospital and individual-based analyses are partly due to selection bias, and partly to different hospital policies.
Conclusions: The results in previous studies and our results suggest that epidural analgesia is a risk factor for an operative delivery, but not a sufficient one, and its impact is likely to be modified by other obstetric practices and interventions.