Association Between Time of Day and the Decision for an Intrapartum Cesarean Delivery

Abstract

Objective: To examine whether the decision and indications for performing intrapartum cesarean delivery vary by time of day.

Methods: We conducted a secondary analysis of a multicenter observational cohort of 115,502 deliveries (2008-2011), including nulliparous women with term, singleton, nonanomalous live gestations in vertex presentation who were attempting labor. Those who attempted home birth, or underwent cesarean delivery scheduled or decided less than 30 minutes after admission were excluded. Time of day was defined as cesarean delivery decision time among those who delivered by cesarean and delivery time among those who delivered vaginally, categorized by each hour of a 24-hour day. Primary outcomes were decision to perform cesarean delivery and the indications for cesarean delivery (labor dystocia, nonreassuring fetal status, or other indications). Secondary outcomes included whether a dystocia indication adhered to standards promoted to reduce cesarean delivery rates. Bivariate analyses were performed using χ and Kruskal-Wallis tests for categorical and continuous outcomes, respectively, and generalized additive models with smoothing splines explored nonlinear associations without adjustment for other factors.

Results: Seven thousand nine hundred fifty-six (22.1%) of 36,014 eligible women underwent cesarean delivery. Decision for cesarean delivery (P<.001) decreased from midnight (21.2%) to morning, reaching a nadir at 10:00 (17.9%) and subsequently rising to peak at 21:00 (26.2%). The frequency of cesarean delivery for dystocia also was significantly associated with time of day (P<.001) in a pattern mirroring overall cesarean delivery. Among cesarean deliveries for dystocia (n=5,274), decision for cesarean delivery at less than 5 cm dilation (P<.001), median duration from 5 cm dilation to cesarean delivery decision (P=.003), and median duration from complete dilation to cesarean delivery decision (P=.014) all significantly differed with time of day. The frequency of nonreassuring fetal status and "other" indications were not significantly associated with time of day (P>.05).

Conclusion: Among nulliparous women who were attempting labor at term, the decision to perform cesarean delivery, particularly for dystocia, varied with time of day. Some of these differences correlate with labor management differences, given the changing frequency of latent phase cesarean delivery and median time in active phase.

Fig. 1.

Flowchart illustrating composition of the sample of nulliparous women with term, singleton, nonanomalous gestations in vertex presentation who attempted labor on the labor and delivery unit. Son. Time of Day and Intrapartum Cesarean Delivery. Obstet Gynecol 2020.

Fig. 2.

Associations between time of day (using decision time to perform cesarean delivery) and decision to perform cesarean delivery/total birth ratio (A), decision to perform cesarean delivery for the indication of dystocia/total birth ratio (B), decision to perform cesarean delivery for the indication of nonreassuring fetal status/total birth ratio (C), and decision to perform cesarean delivery for other indications/total birth ratio (D). The dark lines indicate the estimated probability of outcome by time of day (0–24 hours), and the shaded regions indicate the 95% confidence limits. The probabilities and confidence limits were calculated by generalized additive models. Son. Time of Day and Intrapartum Cesarean Delivery. Obstet Gynecol 2020.