Melissa Cheyney and the Midwives Alliance of North America finally acknowledge their hideous death rates

Grieving family with an infant's coffin --- Image by © Leah Warkentin/Design Pics/Corbis

I told you so!

I’ve been writing about homebirth for more than a decade. For most of that time, the Midwives Alliance of North America (MANA) and Melissa Cheyney, the Director of Research for MANA have insisted that their data show that homebirth is safe.

[pullquote align=”right” cite=”” link=”” color=”” class=”” size=””]In every category — with risks or without — homebirth increases the risk of fetal/neonatal death substantially and often enormously.[/pullquote]

In the meantime, the publicly available data on CDC Wonder has made it possible for me to demonstrate that homebirth deaths rates have been 3-7X higher than comparable risk hospital birth. Amos Grunebaum, MD and colleagues have published several papers using the same data and confirming my analysis. The most comprehensive analysis of homebirth death rates was performed by Judith Rooks, CNM MPH for the state of Oregon. Rooks found that homebirth midwives had a perinatal death rate 800% higher than comparable risk hospital birth!

Now, MANA and Cheyney have finally relented and published their own data that shows that PLANNED birth at home or in a birth center (generally just a rented home without special equipment) in the US has death rates EVEN WORSE than we imagined.

The new paper is Perspectives on risk: Assessment of risk profiles and outcomes among women planning community birth in the United States to be published in Birth: Issues in Perinatal Care, the journal owned by Lamaze International.

The authors used the data generated by their own members:

Data on n=47 394 midwife-attended, planned community births come from the Midwives Alliance of North America Statistics Project. Logistic regression quantified the independent contribution of 10 risk factors to maternal and neonatal outcomes. Risk factors included: primiparity, advanced maternal age, obesity, gesta- tional diabetes, preeclampsia, postterm pregnancy, twins, breech presentation, his- tory of cesarean and vaginal birth, and history of cesarean without history of vaginal birth. Models controlled additionally for Medicaid, race/ethnicity, and education.

Ostensibly, the authors were attempting to determine appropriate criteria for allowing or risking out homebirths and birth center births.

They found:

The independent contributions of maternal age and obesity were quite modest, with adjusted odds ratios (AOR) less than 2.0 for all outcomes: hospital transfer, cesarean, perineal trauma, postpartum hemorrhage, low/very-low Apgar, maternal or neonatal hospitalization, NICU admission, and fetal/neonatal death. Breech was strongly associated with morbidity and fetal/neonatal mortality (AOR 8.2, 95% CI, 3.7-18.4). Women with a history of both cesarean and vaginal birth fared better than primiparas across all outcomes; however, women with a history of cesarean but no prior vaginal births had poor outcomes, most notably fetal/neonatal demise (AOR 10.4, 95% CI, 4.8-22.6).

The author’s definition of “modest” are quite different than mine. They found that for almost all outcomes, homebirth had a nearly 100% increase in fetal/neonatal death. Breech babies had a death rate 700% higher and attempted VBAC had a death rate more than 900% higher than that baseline death rate at homebirth (which is already higher than the hospital death rate).

Two charts provide the most important information.

The first shows absolute death rates:


The death rate for first babies was 3.43/1000 and the death rate for second or subsequent babies was 1.03/1000. Compare that to CDC data that shows a hospital death rate of 0.36/1000 overall and 0.44 for first babies. In other words, homebirth had a neonatal death rate more than triple (200% increase) that of hospital birth for women having second or subsequent babies and 780% higher for first time mothers.

The second chart shows the increased risks of various pregnancy complications compared to the already elevated risk of homebirth demonstrated above:


For example, attempted VBAC increases the risk by more than 10 fold. Twins increases the risk more than 3 fold. Breech increases the risk more than 8 fold. Postdates nearly triples the risk. Pre-eclampsia increases the risk more than 10 fold.

In every category — with risks or without — homebirth increases the risk of fetal/neonatal death substantially and often enormously.

What conclusions do Cheyney and colleagues reach?

The outcomes of labor after cesarean in women with previous vaginal deliveries indicates that guidelines uniformly prohibiting labor after cesarean should be reconsidered for this subgroup. Breech presentation has the highest rate of adverse outcomes supporting management of vaginal breech labor in a hospital setting.

In other words, they attempt to minimize their own findings, despite the fact that they are uniformly terrible and often hideous. And these findings almost certainly UNDERESTIMATE the true death rate at homebirth in two critical ways: first, they compare complicated homebirths with uncomplicated homebirths, undercutting the impact of the fact that even uncomplicated homebirths have higher death rates than comparable risk hospital birth; second, this data is only a subset homebirths attended by members, voluntarily submitted by those members. The real death rates at homebirth are almost certainly even higher.

The bottom line is this: after years of denying that homebirth has a dramatically increased risk of fetal/neonatal death, Melissa Cheyney and MANA have finally admitted the truth. There is not a single category in which homebirth is a safe as hospital birth and in many cases, homebirth increases the risk of fetal/neonatal death by nearly 1000%!

Every American woman has a right to have a homebirth because she has the right to control her own body. But homebirth advocates should stop pretending that homebirth is safe. It is never as safe as hospital birth and generally far more deadly.