Abstract
BACKGROUND: Waterbirth is the act of giving birth submerged in a tub of warm water, with the baby born underwater. Support for widespread adoption of waterbirth in the United States (US) has been limited by predominantly retrospective study designs, the underutilization of interdisciplinary citations, and disagreement between professional organizations. No randomized controlled trials of waterbirth have been conducted in the US. OBJECTIVE: The purpose of this randomized controlled trial was to compare pain management, maternal and neonatal outcomes, and maternal satisfaction between low-risk water and land group study participants. STUDY DESIGN: Low-risk pregnant adults were offered study participation between 25 to 34 weeks of gestation during routine prenatal care. Participants who met the inclusion criteria were consented, enrolled, and randomly assigned to waterbirth or land birth in a 2:1 unblinded fashion, to achieve a sample size of n=120 (80 waterbirth: 40 land birth). The sample size was adjusted for attrition and increased to n=186 (124 waterbirth: 62 land birth) based on an interim analysis. Participants in both arms were excluded if they developed intrapartum risk factors prior to delivery using a modified intention-to-treat model. Study participants received routine intrapartum care, including all available pain relief options. Only those assigned to the waterbirth group were given the option of giving birth in water. Prior to discharge, participants completed The US Birth Satisfaction Scale Revised. The primary study outcome was the use of any intravenous narcotics and/or epidural anesthesia. Secondary outcomes included predetermined composite maternal and neonatal outcomes and overall birth satisfaction. Using a modified intention-to-treat approach, outcomes were analyzed using Chi-square, Fisher's Exact, or Wilcoxon-Mann-Whitney test as appropriate. Relative risks (RR) were computed where applicable. RESULTS: From January 2022 to December 2023, 186 participants were enrolled (n=124 waterbirth, n=62 land birth). Following enrollment, 106 participants (n=74 waterbirth, and n=32 land birth) remained eligible for the study. Attrition was 39%. Compared to the land birth group, waterbirth participants were significantly less likely to use intravenous narcotics [RR: 0.36 (95% confidence interval [CI], 0.17-0.75), P<.01] and epidural analgesia [RR: 0.38 (95% CI, 0.24-0.58), P<.01] and had shorter average first stage labor duration [median difference -2.00 hours (95% CI, -3.00 to 1.00), P=.43]. One cord avulsion occurred in the waterbirth group. Additional maternal and neonatal outcomes were not significantly different between groups. Waterbirth participants experienced higher overall birth satisfaction [median difference 3.50 (95% CI, 1.00-5.00), P=.01]. CONCLUSION: This study further contributes to the growing body of evidence that waterbirth reduces intravenous analgesia and anesthesia use and increases maternal satisfaction. The findings of our study provide support for hospitals to offer waterbirth as a pain relief option for low-risk birthing people.