New analysis from Arizona shows — yet again — that homebirth triples the neonatal death rate

iStock_000001980983XSmall

It’s a remarkably robust finding, repeated in a wide variety of scientific papers and both national and state statistics: homebirth increases the risk of neonatal death by a factor of 3 or more.

The latest example is an analysis prepared by faculty at the College of Public Health of the University of Arizona, Tucson and the Arizona Public Health Training Center for the Arizona Department of Health Services entitled Outcomes of Home vs. Hospital Births Attended by Midwives: A Systematic Review and Meta-analysis.

The authors, 5 professors of public health and 1 doctor, explain why the analysis was commissioned:

Most recently, the licensed midwife community has utilized the democratic process to their advantage to pass legislation to allow for an overhaul of the regulations overseeing homebirths and their profession in the state. Pursuant to HB 2247, AzDHS has formed a Midwife Scope of Practice Advisory Committee, which will evaluate evidence based literature and data to make informed decisions regarding regulation over licensing procedures, scope of practice, and education requirements for licensed midwives in Arizona by July 2013. Of particular salience will be changes in regulations overseeing licensed midwife attendance at births for mothers undergoing a vaginal birth after cesarean (VBAC), breech birth presentation, and multiple fetuses…

In light of Arizona homebirth practitioners’ and clients’ interest in midwives’ scope of practice in the US and elsewhere, we seek in this meta-analysis to compare and contrast direct entry midwives’ outcomes for homebirths with their outcomes in hospital or health care facility settings. The current
limited and conflicting evidence on the outcomes of homebirths versus hospital births with midwives in attendance generates both a need and justification for a review of the available evidenced-based literature.

What did they find?

Nine studies were included in the meta-analysis of child health outcome of births attended by midwives in homes or in hospitals. We analyzed 8 outcomes of child health (neonatal deaths, prenatal deaths, Apgar

Pooled results indicated that homebirths attended by midwives were associated with increased risks for neonatal deaths [pooled OR (95%CI): 3.11 (2.49, 3.89)]. There were no significant differences in outcome of home or hospital births attended by midwives for the other child health measures.

How about maternal outcomes?

… [W]omen who delivered at home with midwives were more likely to have spontaneous delivery and intact lacerations/perineal tear [pooled ORs (95%CIs): 1.64 (1.35, 2.00) and 1.94 (1.25, 3.01) respectively.

Women who delivered in hospitals under the supervision of midwives were more likely to experience assist ed delivery, caesarean sections, forceps, episiotomy, and lacerations/perineal tear (3-4 degrees) [pooled ORs (95%CI s): 0.58 (0.40, 0.84); 0.55 (0.49, 0.60); 0.54 (0.33, 0.9 0); 0.56 (0.41, 0.77) and 0.48 (0.32, 0.72) respectively. Results of the meta-analysis also revealed that homebirths attended by midwives were associated with decreased risk for postpartum hemorrhage >500ml and retained placenta [pooled ORs (95%CI s 0.60 (0.44, 0.81) and 0.58 (0.40, 0.86) respectively.
Homebirths were also not associated increased risk for vacuum extraction, cervical tear, blood transfusion and prolapsed cord.

The authors conclude:

These results suggest that homebirth is a suitable alternative to the traditional hospital setting, as it reduces medical interventions and has been found to have positive maternal health outcomes. However, homebirths should only be recommended to women who are classified as low-risk, as this data demonstrates an increased risk of neonatal mortality among homebirths

For reasons that are unclear to me, the authors state:

The findings suggest that homebirths attended by midwives may be equally safe if not safer for women with low-risk pregnancies.

I find that statement surprising for two reasons. First, that is not what their own data showed. Second, claiming that is “may be” equally safe acknowledges that possibility that it may NOT be equally safe.

Regardless, there is one inescapable conclusion of the analysis; homebirth increases the risk of neonatal death by a factor of 3 or more.

For mothers, homebirth poses a much lower risk of interventions and the complications that may arise from those interventions. But that advantage is purchased at the price of increased risk of neonatal death, demonstrating yet again that much of obstetrics is preventive medicine, designed to prevent neonatal deaths … and that’s exactly how it works. Give birth at home and you are twice as likely to avoid interventions, but three times as likely to end up with a dead baby as the result.