Can Buy Zovirax Over Counter Buy Lumigan In The Uk Order Oral Finasteride Cialis For Sale In Toronto Ventolin Discount Mail Order

///2013 Abstract Details
2013 Abstract Details2019-08-02T16:57:45-05:00

Novel statistical analysis shows no statistically significant effect of epidural analgesia on the probability of Cesarean section

Abstract Number: S 18
Abstract Type: Original Research

Stuart G Baker ScD1 ; Karen S Lindeman MD2

We revisited the question of whether epidural analgesia has a statistically significant effect on the probability of Cesarean section. Nygen et al (2010, Matern Child Health J) argued the answer was yes based on (1) an observational study involving propensity scores, (2) a belief that randomized trials are not convincing due to dilution from crossovers and (3) a belief that before-and-after studies are not convincing due to bias from temporal changes. We countered these arguments in three ways. First, we noted that the propensity score methods can be substantially biased by unmeasured confounders, particularly intense pain in labor. Second we performed a meta-analysis of 16 randomized trials using principal stratification to remove dilution from crossovers. Third we performed a paired availability design involving 11 before-and-after studies to average random temporal changes while adjusting for different changes in availability of epidural analgesia. For the latter two methods we compared four novel extrapolation methods (denoted FLQS, REF2, REF, and RE). Figure 1 compares point estimates and 95% confidence intervals for the effect of epidural analgesia on the rate of Cesarean-section in the likely biased propensity score analysis (PS), the meta-analysis of randomized controlled trials (RCT) and the paired availability design (PAD). Based on the results in Figure 1, we concluded that epidural analgesia has no statistically significant effect on the probability of Cesarean section with a reasonably narrow confidence interval.



SOAP 2013