Is the American College of Nurse Midwives lying about the Cornell homebirth study, or merely ignorant of how it was done?

the truth or your story

In the past month, a variety of studies have been published that have shown unequivocally that homebirth increases the risk of perinatal and neonatal death. Even the data from homebirth midwives themselves shows that homebirth has a 450% higher death rate. This is bad news for homebirth midwives and supporters so they’ve embarked on an orgy of lying about what the various studies show, how they were done, and what they mean.

Perhaps I am naive, but I expected better from the American College of Nurse Midwives (ACNM). Unfortunately, they’ve jumped right into the muck. They are actively misleading both laypeople and reporters about the latest Cornell study, which showed that the risk of neonatal death at homebirth is 4 times higher than comparable risk hospital birth. Either they are deliberately lying or they actually don’t understand what they have read, even though it is quite straightforward.

A piece on about the new Cornell study includes the following claim:

In a response sent to TIME, the American College of Nurse-Midwives (ACNM) raises skepticism over determining the accuracy of the data that the study was based on. “Birth certificates are not always completely filled out, nor are they always filled out by the provider attending the birth. They are not always accurate, when compared to medical charts filled out by the attending provider,” they write. “The birth certificate data do not allow researchers to accurately separate out planned versus unplanned home births. We know that mortality rates are higher for unplanned home births, as they are more likely to involve emergency/urgent situations. The inability to separate these out casts doubt on the findings.”

But the study is NOT based on birth certificates, as the ACNM should know if they read the study, Term neonatal deaths resulting from home births: an increasing trend:

A retrospective cohort study using the CDC linked birth/infant death data set for term (>=37 weeks), >=2500 grams, singleton live births, excluding congenital anomalies from 2007 to 2009. Deliveries were categorized by setting: hospitals, birthing centers, and home as well as providers (midwives, doctors, and “others” for home births). Neonatal mortality (NNM) was defined as neonatal deaths up to 28 days after delivery. Hospital midwives served as reference. (my emphasis)

The CDC linked birth-infant death data set is NOT birth certificates, and is far more accurate than birth certificates alone. The birth and death certificates of each infant was linked to provide the maximum amount of information and to cross check the reliability of the data. So the ACNM is either lying about or ignorant of the source of the data and, therefore, the reliability of the data.

Moreover, since 2003 birth certificates have included place of birth, so no one is guessing where the babies were born. In addition, it is a legal requirement that the person who delivered the baby must sign the birth certificate. Any birth certificates signed by homebirth midwives must have been PLANNED, ATTENDED homebirths. While studies of birth certificates have shown that there can be errors in certain forms of data (e.g. maternal risk factors), they have never shown that the signature of the attendant is anything other than highly accurate.

The chief limitation of the CDC linked birth-infant data set is that it UNDERCOUNTS homebirth deaths because any babies who were transferred during a homebirth and were born and died in the hospital are erroneously removed from the homebirth group.

The bottom line is that the Cornell study unequivocally demonstrates that homebirth increases the risk of neonatal death by at least a factor of 4 and probably much more. The ACNM should apologize for making erroneous claims about the study and about the data … and they might consider studying up on the various forms of natality and infant death data since they appear, at a minimum, to be ignorant of the differences between them.