Regular contributor Henci Goer examines the most recent study on the safety of home birth in the United States. When taking a closer look at the data analysis done by the authors, there are concerns not addressed in the study, that raise issues that cause the study’s conclusions to be questioned. Henci shares some other studies that do not reach the same results about the safety of home birth. Have you read this study? If you had read this study too, did you find more questions than answers when you were done? – Sharon Muza, Community Manager, Science & Sensibility.
“Researchers have already cast much darkness on the subject, and if they continue their investigation, we shall soon know nothing at all.” – Mark Twain
The latest contender in the long list of studies attempting to compare the safety of home and hospital birth, “Selected perinatal outcomes associated with planned home births in the United States,” was published last month (Cheng 2013). Let’s start by summarizing the study:
Using data compiled from the U.S. birth certificate, Cheng and colleagues compared outcomes between 12,039 women “planning” home births with 2,081,753 women having hospital births. All women were at term (between 37 and 43 weeks) and carrying one head-down baby. Women with prior cesarean were not excluded. After adjustment for numerous factors including number of prior births, medical conditions (hypertension, diabetes), risk factors (smoking), and social and demographic factors (race/ethnicity, age, marital status), women having home births were much less likely to have an instrumental vaginal delivery (0.1% vs. 6.2%; odds ratio 0.1), induced labor (1.4% vs. 25.7%; odds ratio 0.2), or labor augmentation (2.1% vs. 22.2%; odds ratio 0.3). They were also, however, twice as likely to have a baby with a 5-minute Apgar less than 4 (0.24% vs. 0.37%; odds ratio 1.9), three times as likely to have a baby experience neonatal seizure (0.06% vs. 0.02%; odds ratio 3.1), and more than twice as likely to have a baby with 5-minute Apgar less than 7 (2.42% vs. 1.17%; odds ratio 2.4). On the other hand, similar percentages of babies needed more than 6 hours of ventilator support, and babies born at home were much less likely to be admitted to intensive care (0.57% vs. 3.03%; odds ratio 0.2). In the discussion, the investigators note that removing the 489 women with previous cesareans who had planned home birth and women with medical or obstetric conditions did not alter that infants of women with prior births who planned home birth were more likely to have a low Apgar score. They don’t specify whether this was 5-minute Apgar less than 4 or less than 7 nor do they report the occurrence rate in this higher-risk subgroup.
There is more. To evaluate the effect of birth attendant qualifications, the investigators excluded births attended by doctors or unknown birth attendant and stratified the remaining home birth population into those attended by professional midwives and those attended by “other midwives.” (Confusingly, study authors state that Certified Professional Midwives [CPMs] were categorized as Certified Nurse-Midwives in the birth certificate data yet go on to refer solely to “CNMs” in the rest of the analysis.) In the subset attended by professional midwives, newborn outcomes were similar except that hospital-born infants were more likely to be admitted to intensive care (0.37% vs. 3.03%; odds ratio 0.1).
Cheng and colleagues conclude that while women planning home births are less likely to experience obstetric intervention, their babies are more likely to be born in poor condition. Do their data warrant that conclusion?
To begin with, the relevant question isn’t the tradeoffs between planned home birth per se and hospital birth. It is: “What are the excess risks for healthy women at low risk of urgent complications who plan home birth with qualified home birth attendants compared with similar women planning hospital birth?” This study can’t answer that question. Here’s why:
The study only includes women actually delivering at home, but you can’t make a meaningful comparison unless you have the outcomes of women transferred to hospital. “Planning” in this study meant only that birth at home wasn’t accidental, not the more usual meaning that birth may be planned at home but problems during labor may alter that plan. I discovered this when I wrote the lead author to request cesarean rates, which, oddly, to me, were not reported in the study. She responded that this was because cesareans aren’t performed at home. Puzzled by this explanation, I wrote back that neither are instrumental vaginal delivery, induction, nor labor augmentation, which were reported. She responded that birth certificate data don’t state how labor was induced or augmented but that perhaps at home births it was by rupturing membranes and that “apparently some midwives or birth attendants do perform vacuum extraction at home,” but it is rare since only 10 were reported.
Not all women planning home birth were low-risk. For one thing, women with prior cesareans were included. For another, the methods section states that the analysis adjusted for medical risk, and the discussion notes that women with prior children in the home birth group were more likely to have babies with low Apgar scores even after removing women with medical risk, which implies that some of them had medical problems.
Not all women in the home birth group had qualified home birth attendants. Outcome data on the overall population came from women recorded as being attended by MDs, DOs, “other midwife,” “others,” and “unknown/not stated” as well as by professional midwives.
Rates of neonatal seizure and 5-minute Apgar less than 4 were very low, and the study doesn’t report on perinatal death or permanent disability. As concerning as an excess in low Apgar scores and seizures may be, the real question is excess incidence of permanent harm. Even without limiting the population to low-risk women with qualified care providers, only 1 more baby per 1000 born at home experienced very low 5-minute Apgar, and only 4 more babies per 10,000 experienced neonatal seizure, and while babies born in poor condition are more likely to incur permanent neurologic damage or die, most will recover. Also, as we saw, differences in rates of these adverse outcomes disappeared with a qualified provider.
The proof of the pudding lies in studies free of these weaknesses. A study of 530,000 low-risk Dutch women found no difference in deaths during labor or newborn death rates between women planning, but not necessarily having, home birth and those planning hospital birth (de Jonge 2009). A Canadian study comparing outcomes of 2900 women eligible for home birth with women equally eligible but planning hospital birth reported worse newborn outcomes (more required resuscitation at birth or oxygen for more than 24 hrs and more birth injuries), worse maternal outcomes (more anal sphincter tears and postpartum hemorrhage), and more use of instrumental and cesarean delivery in the hospital population (Janssen 2009).
What can we take away from Cheng and colleagues analysis? First, care provider qualifications matter. Women desiring home birth should have access to professional midwifery care, which argues for making CPMs legal in all 50 states. Second, less than optimal candidates are birthing at home, and some women may be continuing labor at home who shouldn’t. Why might that be? Women may choose home birth because they want control over what happens to them, they have had a prior negative hospital experience, or they want to avoid unnecessary medical intervention (Boucher 2009), the last of which will include women denied hospital VBAC. Women may resist hospital transfer for the same reasons or because they know that at best, hospital transfer means losing the care and advice of the care provider they trust and at worst, they will be treated badly by disapproving hospital staff. If we want to reduce their numbers, hospital-based practitioners need to address the behaviors, practices, and policies that drive women away from hospital birth. This would have the added benefit of improving care for the 99% of American women who would never consider birthing at home.
Boucher, D., Bennett, C., McFarlin, B., & Freeze, R. (2009). Staying home to give birth: why women in the United States choose home birth. J Midwifery Womens Health, 54(2), 119-126. http://www.ncbi.nlm.nih.gov/pubmed/?term=boucher+2009+home+birth
Cheng, Y. W., Snowden, J. M., King, T. L., & Caughey, A. B. (2013). Selected perinatal outcomes associated with planned home births in the United States. Am J Obstet Gynecol, 209(4), 325 e321-328. doi: 10.1016/j.ajog.2013.06.022 http://www.ncbi.nlm.nih.gov/pubmed/23791564
de Jonge, A., van der Goes, B. Y., Ravelli, A. C., Amelink-Verburg, M. P., Mol, B. W., Nijhuis, J. G., . . . Buitendijk, S. E. (2009). Perinatal mortality and morbidity in a nationwide cohort of 529,688 low-risk planned home and hospital births. BJOG 116(9), 1177-1184. http://www.ncbi.nlm.nih.gov/pubmed/?term=de+jonge+2009+planned+home
Janssen, P. A., Saxell, L., Page, L. A., Klein, M. C., Liston, R. M., & Lee, S. K. (2009). Outcomes of planned home birth with registered midwife versus planned hospital birth with midwife or physician. CMAJ, 181(6-7), 377-383. http://www.ncbi.nlm.nih.gov/pubmed/19720688