In 2019, a meta-analysis examined perinatal and
neonatal mortality of planned home birth among low-risk women in well-resourced countries, with research eligible for inclusion encompassing approximately 500,000 intended home births. The study concluded that the risk of perinatal or neonatal mortality was not different whether birth was intended at home or in hospital. In 2014, a comprehensive review in the
Journal of Medical Ethics of 12 previously published studies encompassing 500,000 planned home births in low-risk women found that neonatal mortality rates for home births were triple those of hospital births. Due to a greater risk of perinatal death, the college advises women who are post-term (greater than 42 weeks
gestation), carrying twins, or have a
breech presentation, not to attempt home birth. The
Journal of Medical Ethics review additionally found that several studies concluded that home births had a higher likelihood of failing
Apgar scores in newborns, as well as a delay in diagnosing
hypoxia,
acidosis and
asphyxia. In North America, a 2005 study found that about 12 percent of women intending to give birth at home needed to be transferred to the hospital for reasons such as a difficult labor or pain relief. A 2014 survey of American home births between 2004 and 2010 found the percent of women transferred to a hospital from a planned home birth after beginning labor to be 10.9%. Both the
Journal of Medical Ethics and the NICE report noted that the use of
caesarean sections was lower for women who give birth at home, and both noted a prior study which determined that women who had a planned home birth had greater satisfaction from the experience when compared with women who had a planned birth in a hospital. In 2009 a study of 500,000
low-risk planned home and hospital births in the Netherlands, where midwives have a strong licensing requirement, was reported in the
British Journal of Obstetrics and Gynaecology. The study concluded that for
low-risk women there was no increase in perinatal mortality, provided that the midwives were well-trained and that there was easy and quick access to hospitals. Further, the study noted that there was evidence that "low risk women with a planned home birth are less likely to experience referral to secondary care and subsequent obstetric interventions than those with a planned hospital birth." The study has been criticised on several grounds, including that some data might be missing and that the findings may not be representative of other populations. In 2012, Oregon performed a study of all births in the state during the year as a part of discussing a bill regarding licensing requirements for midwives in the state. They found that the rate of intrapartum infant mortality was 0.6 deaths per thousand births for planned hospital births, and 4.8 deaths per thousand for planned home births. They further found that the death rate for planned home births attended by direct-entry midwives was 5.6 per thousand. The study noted that the statistics for Oregon were different for other areas, such as British Columbia, which had different licensing requirements. Oregon was noted by the
Centers for Disease Control and Prevention as having the second-highest rate of home births in the nation in 2009, at 1.96% compared to the national average of 0.72%. A 2014 survey of nearly 17,000 voluntarily reported home births in the United States between 2004 and 2010 found an intrapartum infant mortality rate of 1.30 per thousand; early neonatal and late neonatal mortality rates were a further 0.41 and 0.35 per thousand. The survey excluded congenital anomaly-related deaths, as well as births where the mother was transferred to a hospital prior to beginning labor. A 2022 study, which examined the introduction of maternity wards in Sweden, found that the wards substantially reduced home deliveries and early neonatal mortality, as well as positive long-term effects on labour income, unemployment, health-related disability and schooling for individuals born in maternity wards.
Study design Randomized
controlled trials are the "gold standard" of research methodology with respect to applying findings to populations; however, such a study design is not feasible or ethical for location of birth. The studies that do exist, therefore, are
cohort studies conducted retrospectively by selecting hospital records and midwife records. by matched pairs (by pairing study participants based on their background characteristics), In February 2011 the
American Congress of Obstetricians and Gynecologists identified several factors that make quality research on home birth difficult. These include "lack of randomization; reliance on birth certificate data with inherent ascertainment problems; ascertainment of relying on voluntary submission of data or self-reporting; a limited ability to distinguish between planned and unplanned birth; variation in the skill, training, and certification of the birth attendant; and an inability to account for and accurately attribute adverse outcomes associated with transfers". Quality studies, therefore, need to take steps in their design to mitigate these problems in order to produce meaningful results. The data available on the safety of home birth in developed countries is often difficult to interpret due to issues such as differing home-birth standards between different countries, and difficult to compare with other studies because of varying definitions of perinatal mortality. There are also unquantifiable differences in home birth patients, such as maternal attitudes towards medical involvement in birth. == Methods of scientific inquiry ==