Experienced lay women, or doulas, who provide continuous emotional support to women in labor have been credited with improving many aspects of childbirth. The latter include reducing the number of operative deliveries and the use of analgesia and anesthesia, the length of labor and negative psychosocial outcomes. Most studies of the effect of doulas in the United States have focused on the support of young, nulliparous mothers from low-income groups. Little is known about their effect on other populations of women in the United States, particularly middle-class, married women whose spouses or other companions stay with them throughout labor and delivery. Gordon and colleagues studied the effectiveness of doulas in hospital-based births for women enrolled in a large Northern California health maintenance organization.

To be eligible for the study, women had to be nulliparous and in spontaneous labor after an uncomplicated pregnancy. Patients who met the criteria after initial assessment were randomized to receive usual care or doula-assisted delivery. All doulas had completed a community training program and had served as a doula at two or more deliveries. After every delivery, each doula was evaluated by the attending physician, a nurse and the patient. Patients in the doula group were interviewed by telephone four to six weeks after delivery.

A total of 314 women were enrolled in the study. Of these, 149 were in the doula group, and 165 received usual care. More than one half of the patients were white college graduates. Mean patient age was 29 years, and more than 80 percent of patients had attended childbirth preparation classes. Demographic information, stage of labor and types of support persons at the hospital were similar between groups. Rates of uncomplicated vaginal delivery, cesarean delivery, operative delivery and use of oxytocin or analgesics were also similar between groups. However, patients in the doula group were less likely to have epidural anesthesia (54 percent compared with 66 percent), even though slightly more women in this group had planned to request it compared with those in the usual care group. Women in the doula group were also more likely to rate the birth experience as good and to have positive perceptions about themselves and their bodies. No differences were documented in patients' descriptions of their labor or in postpartum measures of self-esteem, including depression. More than 95 percent of women in each group breast-fed initially and both groups had very high rates of breast feeding after four weeks.

The authors conclude that doulas contributed positively to labor outcomes, even in this highly supported population, but the measurable benefits may be limited to a reduction in the use of epidural anesthesia and positive outcomes related to maternal perceptions of well-being. In all other areas, there were no substantial differences. Women who had doulas were very enthusiastic about their experience with them, and those who did not reported high interest in using doulas in future deliveries. The knowledge, support and reassurance that doulas provide during labor and delivery might restore the human aspect to childbirth.