Compared with women who plan to have their baby delivered in a hospital, women who intend to have their baby delivered by a professional provider at home have a greater likelihood of complications during and after labor and delivery, and their newborn has an increased risk of compromised health. In particular, newborns of women who planned a home delivery are twice as likely to have a very low Apgar score or to die as are newborns of women who planned a hospital delivery, according to a retrospective analysis of births in Washington State from 1989 to 1996.1 Moreover, among previously nulliparous women, planned home birth is associated with elevated risks for prolonged labor and postpartum bleeding. As the analysts comment, these findings are important to consider, given the increased popularity of out-of-hospital births in recent years.
In this population-based cohort study, the researchers used information from birth certificates and infant death certificates from 1989 to 1996 in Washington State to assess whether intended place of delivery--home or hospital--affects selected maternal or neonatal health outcomes. They examined 6,133 singleton deliveries by a health care provider among women who planned to deliver at home, including 279 attempted home deliveries that resulted in the woman's transfer to a hospital. The study was limited to births that occurred at 34 or more weeks' gestation after an uncomplicated pregnancy (i.e., prior to onset of labor, the woman was not known to have chronic or pregnancy-induced hypertension, eclampsia, diabetes mellitus, hepatitis B virus infection or any of 13 other conditions).
For the study's primary analysis, outcomes in the planned home births were compared with those in 10,593 randomly selected births involving planned hospital deliveries, which were matched for year of birth to the planned home deliveries. Secondary analyses examined only births at 37 or more weeks' gestation to an infant weighing at least 2,500 g: in all, 6,052 intended home births, including 269 transfers, and 10,347 planned hospital births.
Women who intended to give birth at home were older than women with planned hospital births (96% and 89%, respectively, were aged 20 or older). Higher proportions in the planned home birth group were white (92% vs. 81%), parous (76% vs. 57%) and married (83% vs. 75%). In addition, a higher proportion had more than a high school education (61% vs. 49%), had insurance or paid for their own care (72% vs. 61%), had an infant whose birth weight was at least 2,500 g (99% vs. 98%) and had never smoked (89% vs. 82%). Meanwhile, a larger proportion of women in the planned hospital birth group than in the planned home birth group resided in an urban setting (79% vs. 73%) and received prenatal care in the first trimester (82% vs. 72%).
Adverse events overall were rare among the infants in this study. However, three outcomes seemed more common in newborns of women who planned to deliver at home than in newborns of women who intended to deliver in a hospital. The researchers calculated relative risks to assess the significance of the apparent differences. First, in analyses adjusting for maternal age, they found that compared with infants of women who planned a hospital delivery, newborns of women who planned a home delivery had a significantly elevated risk (relative risk, 2.3) of having a low score (0-3 on a scale of 0-10) on the Apgar test taken five minutes after birth. The results were similar regardless of whether the woman had given birth previously.
Second, the risk of neonatal death for newborns whose mother planned to deliver at home was nearly double that for infants whose mother intended to deliver in a hospital (after controlling for parity, the relative risk was 2.0), and the differential increased slightly when the researchers controlled for the women's insurance status and level of education. The risk of neonatal death in newborns whose mother planned a home delivery remained elevated (relative risk, 2.1 after controlling for maternal age) in the analysis restricted to infants born at 37 weeks' gestation or later and with a minimum birth weight of 2,500 g. The risk was especially elevated for newborns of previously nulliparous women who intended to deliver at home (2.7 in the primary analysis and 3.0 in the more restricted analysis). No differences were found in the groups' risk of postneonatal death.
For the third measure of infant health, respiratory distress (i.e., requirement of ventilation for more than 30 minutes), the newborns of previously nulliparous women who intended to give birth at home appeared to be at increased risk. However, the result was significant (relative risk, 3.2) only in the secondary analysis.
Among previously nulliparous women (but not for those who had given birth before), planned home deliveries were associated with an elevated risk of two maternal outcomes: prolonged labor (relative risk, 1.7) and postpartum bleeding (2.8). The results were similar in the secondary analysis.
According to the researchers, this study suggests that planned home births might be associated with an increased risk of adverse neonatal and maternal outcomes, particularly among women who have not given birth previously. However, the researchers also acknowledge several limitations of their study--notably, the possibility that the true intention of the mothers for delivery location, as well as other factors, was misclassified. Thus, they caution that further studies are needed: "More light needs to be shed on this controversial topic before practitioners and expectant parents can be fairly counseled about the safety of planned home births."--C. Coren
1. Pang JWY et al., Outcomes of planned home births in Washington State: 1989-1996, Obstetrics & Gynecology, 2002, 100(2):253-259.