Our goal was to contrast the influence of intermittent and continuous support provided by doulas during labor and delivery on 5 childbirth outcomes. Data were aggregated across 11 clinical trials by means of meta-analytic techniques. Continuous support, when compared with no doula support, was significantly associated with shorter labors (weighted mean difference -1.64 hours, 95% confidence interval -2.3 to -.96) and decreased need for the use of any analgesia (odds ratio.64, 95% confidence interval.49 to.85), oxytocin (odds ratio.29, 95% confidence interval.20 to.40), forceps (odds ratio.43, 95% confidence interval.28 to.65), and cesarean sections (odds ratio.49, 95% confidence interval.37 to.65). Intermittent support was not significantly associated with any of the outcomes. Odds ratios differed between the 2 groups of studies for each outcome. Continuous support appears to have a greater beneficial impact on the 5 outcomes than intermittent support. Future clinical trials, however, will need to control for possible confounding influences. Implications for labor management are discussed.