Path analysis model of epidural/spinal anesthesia on breastfeeding among healthy nulliparous women: Secondary analysis of the United States Certificate of Live Births 2016.

Abstract

The effect of epidural/spinal anesthesia during labor on breastfeeding is unclear. Few studies had assessed whether or how medically assisted delivery (operative vaginal delivery or unscheduled cesarean birth) plays a mediating role. We aimed to examine whether the relationship between using epidural/spinal anesthesia and breastfeeding is mediated by increased medically assisted delivery among healthy nulliparous women. A secondary, cross-sectional analysis was conducted using US birth certificate data from 2016 (n=381199). Logistic regression was used to examine associations between factors. Structural equation modeling (SEM) was used to analyze the model fit of the path models and to quantify the direct, indirect, and total effect of anesthesia on breastfeeding at discharge, considering medically assisted delivery as a mediator. Women who were administered epidural/spinal anesthesia were more likely to experience medically assisted delivery (adjusted odds ratio [AOR]: 95% confidence interval [CI] 3.01 (2.91-3.12)) and less likely to be breastfeeding at discharge (0.95 [0.92-0.98]). Operative vaginal and unscheduled cesarean deliveries were significantly associated with nonbreastfeeding at discharge (0.81 [0.77-0.84] and 0.81 [0.79-0.84], respectively). SEM revealed excellent model fit for our model. The indirect effect was significant (β=-0.038; 95% CI, -0.043 to -0.033), as was the total effect (β=-0.038; 95% CI, -0.043 to -0.033). Epidural/spinal anesthesia is associated with nonbreastfeeding at discharge, mediated through medically assisted delivery. Health care providers should consider these risks and provide adequate support to help all parents attain their breastfeeding goals.