This study was published in June and is a welcome addition to the homebirth v. hospitals debate which has been raging in this country for years. Some of the biggest arguments against homebirth studies so far have been that they’ve been predominantly retrospective, which brings up issues of selective reporting and potential bias, and few studies can distinguish between planned and unplanned home births. Well, no longer! This study measures only planned homebirths, and is prospective in nature, which provides a rather conclusive wallop to the ongoing debate.
In a large, prospective cohort study carried out with assitance from the North American Registry of Midwives (NARM), the outcomes of low risk women giving birth in hospitals were compared with those giving birth at home under the care of a CPM. 409 practicing CPMs agreed to take place in this study, enrolling all of their patients with an expectant date of delivery in the year of 2000. Overall, 7623 women were registered, although several women were eliminated from the data pool for multiple reasons (such as medical problems removing them from the low risk category, deciding to deliver in a hospital instead of at home, declined participation, etc. etc.); the study group was eventually nailed down to 5418 women who intended to give birth at home at the start of their labor. This number was then compared to all the women who gave birth to singleton, vertex babies of at least 37 weeks gestation in a hospital in 2000, after determining that they were low risk through a screening tool consisting of 13 personal and behavioral variables associated with perinatal risk. The data for this control group was provided by the National Center for Health Statistics.
It was found that the intrapartum and neonatal mortality rate for homebirths was on par with what other studies have shown for hospital births, while the rate for medical inverventions was much lower for the homebirth group as compared to the hospital birth group. In other words, for low risk women, homebirth is just as safe as hospital birth, and hospital birth is just as safe as homebirth, from a statistical standpoint. Of the 5418 women, 655 (12.1%) were transferred to the hospital, mostly for epidural medication (4.7%) or cesarean delivery (3.7%), indicating medical intervention rates astoundingly lower than the national average (19% in 2000 for low risk women). The intrapartum and neonatal mortality rate was calculated at 1.7 deaths for every 1000 planned home births, after breech and twin deliveries were excluded (since they don’t qualify as low risk). This is consistent with the findings of other studies of planned home births and low risk hospital births. No maternal deaths occurred.
In general, this study is very strong. It’s prospective and limited to planned homebirths, the data is clearly presented and easy to understand, and the results are consistent with other findings. The weaknesses of this study lie in the fact that the control group was not prospective as well, but the authors addressed this in the discussion, pointing out the difficulties in trying to create a workable design for low risk hospital births. Vital statistics forms do not accurately collect information on medical risk factors, either, so it was difficult to create a retrospective low risk hospital group as well. Additionally, there may always be confounding factors, such as some unidentified difference between women who select home birth as their method, when compared to women who choose to give birth in a hospital.
Sadly, I’m not sure how much a study like this is going to change current attitudes towards homebirth, despite it’s strong design and clear findings. On the BMJ website, several practitioners wrote comments in response to this study which pretty much flat-out stated that it wouldn’t change their practice in any way. Another practioner commented about the 1.3% of all babies born in the homebirth group who had five minute Apgar scores below 7; this statistic is very small, and in line with hospital statistics, but the practioner honed in on it as if any home without an on-site NICU is inadequate, and as if the CPMs were not prepared to resuscitate the babies properly (they are, and babies are usually stable before being transferred to the hospital). Even so, the majority of the comments were resoundingly positive, and I should take this opportunity to remind my impatient, idealistic self that change is slow, and is often a long time in coming. It was nearly 20 years, after all, before the medical community came to accept handwashing between patient visits as a common practice. Our country is slowly, slowly moving away from technocratic birth, and coming to a larger acceptance of homebirth and midwifery. We’ll get there eventually.