Does Where a Mother Plans to Deliver Matter for the Baby?

William T. Basco, Jr, MD, MS


March 15, 2016

Planned Out-of-Hospital Birth and Birth Outcomes

Snowden JM, Tilden EL, Snyder J, Quigley B, Caughey AB, Cheng YW
N Engl J Med. 2015;373:2642-2653

Study Summary

Over the past two decades, a steady increase has been seen in planned out-of-hospital births, including those taking place at home or in outpatient birthing centers. At present, 0.89% of all births occur in home settings and 0.39% occur in birth centers.[1,2] Past studies have demonstrated a higher rate of perinatal mortality among infants born at home.[3,4]

Data from this study by Snowden and colleagues come from Oregon, which has a home birth rate of 2.4% and a birth-center birth rate of 1.6% (both higher than national averages). Beginning in 2012, birth certificates in Oregon were required to indicate the intended place of birth for all infants. This permitted accurate classification of births that were planned for home or a birth center but ultimately took place in a hospital setting, after mothers were transferred owing to complications or for other reasons.

The births in this study took place between January 1, 2012, and December 31, 2013. Unplanned home births were excluded if the intended birth setting was not known. For analysis purposes, home and birthing center births were aggregated and compared with outcomes of planned hospital births.

An analysis of more than 79,000 births showed that 95.2% of the deliveries were planned for and took place in the hospital, and 4% were planned for and took place out of the hospital. Another 601 births (0.8%) were planned for out of the hospital, but these mothers were transferred to hospitals for delivery.

Racial differences were observed in planned out-of-hospital birth rates. White mothers comprised 67.5% of those who planned hospital births and 88.9% of those who planned and completed out-of-hospital births. Higher maternal education and higher maternal age were associated with a higher frequency of planned out-of-hospital deliveries. Women with chronic medical conditions and those who delivered prematurely were more likely to have planned for hospital deliveries.

The data in this study were analyzed in two ways. First, outcomes were compared according to actual location of birth. Outcomes were recalculated to include the 601 planned out-of-hospital births that occurred in the hospital (classified appropriately as planned out-of-hospital births). Fetal death rates (death before delivery) did not differ by delivery location (0.13% in-hospital vs 0.06% for out-of-hospital births). However, after the appropriate reclassification, fetal death was higher among out-of-hospital births (0.24% vs 0.12% for in-hospital births), but this difference was on the margin of statistical significance (P =.05).

With proper classification of transferred mothers and infants, the rate of perinatal death (the composite of fetal death and neonatal death) was higher among the planned out-of-hospital births (0.39% vs 0.18% among hospital births; P = .003). Similarly, neonatal death (death in the first 28 days) was more common among planned out-of-hospital births (0.16% vs 0.06%).

The actual frequency of infant death (death in the first year of life) was also higher among the out of hospital births (0.21% vs 0.15%), but this difference did not reach statistical significance. Other adverse outcomes, including neonatal seizures and need for ventilator support, were more common among planned out-of-hospital births after the 601 transferred mothers and infants were accounted for.

The adjusted analysis (accounting for maternal race/ethnicity, maternal parity, insurance status, perinatal visits, maternal age, maternal education, history of cesarean delivery, and the presence of chronic medical conditions) showed that the overarching trend was for higher rates of adverse outcomes among planned out-of-hospital births. The adjusted odds ratio (aOR) for fetal death was 2.3; for neonatal death, 2.87; and for perinatal death, 2.43. After full adjustment, the difference in infant death remained insignificant, but seizures (aOR, 3.6) and ventilator support (aOR, 1.36) remained more common among planned out-of-hospital births.

Maternal outcomes generally favored planned home births, in terms of fewer operative interventions. The only exception was that mothers with planned out-of-hospital birth had a higher aOR of requiring a blood transfusion (1.91). Even after propensity-score adjustment (adjusting for factors that predispose women to choose out-of-hospital over hospital births), the association of planned out-of-hospital birth with more adverse outcomes for the infants remained significant.


In 2010, I reviewed a similar study[5] with similar findings: Outcomes are worse for planned out-of-hospital births. This study improves on what is known, owing to the ability to identify planned out-of-hospital births that ultimately delivered in the hospital.

Just as in the older study, I'm struck that the frequency of adverse outcomes is so much higher in the planned outpatient setting, despite all of the factors that bias toward more adverse outcomes among infants born in the hospital, such as known anomalies, premature labor, and mothers with chronic medical conditions.

This is a challenging conversation to have with expectant parents, in part because the absolute differences in poor outcomes are so small—almost all adverse outcomes in this study occurred in < 1% of births—such that even doubling the frequency of adverse outcome (eg, perinatal death rate of 0.39% vs 0.18%) means that adverse outcomes will still occur in fewer than 1 in 1000 deliveries. To really understand the risks, parents would need to understand both the difference in adverse outcome frequency and the absolute risk for adverse outcomes.

However, I suspect that most parents who choose home birth do so for reasons that are very compelling to them, and which are emotional rather than data-driven. That said, the consistent pattern of results of this and other observational studies suggests that there is a real risk worth discussing when it comes to planned out-of-hospital births.



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.