Jan 31 2014

Death Rate for Home Births

A new study publishes data from The Midwives Alliance of North America Statistics Project, 2004 to 2009. It is being presented as evidence for the safety of homebirths. The authors conclude:

“For this large cohort of women who planned midwife-led home births in the United States, outcomes are congruent with the best available data from population-based, observational studies that evaluated outcomes by intended place of birth and perinatal risk factors. Low-risk women in this cohort experienced high rates of physiologic birth and low rates of intervention without an increase in adverse outcomes.”

Unless, of course, you consider death an adverse outcome.

Their data show that the death rate for birth and up to six weeks following birth was 2.06 per 1000 overall (excluding fatal congenital anomalies), and 1.61 for low risk births. Amy Tuteur (the Skeptical OB) calculates from the CDC database that the same statistics for planned hospital births are 0.38 per 1000 for low risk births. That’s 4.2 x higher.

This is consistent with prior data. A 2010 systematic review concluded:

“Less medical intervention during planned home birth is associated with a tripling of the neonatal mortality rate.”

The press release from Citizens for Midwifery claims about this new study:

“New Home Birth Study from the MANA Statistics Dataset Shows That Planned Home Birth with Skilled Midwives is Safe for Low-Risk Pregnancies”

Generally speaking in medicine we don’t consider a more than 4x increase in death rate to be “safe.” Of course, to be fair, I am talking about relative risk. In absolute terms this is about a 0.13% increase in risk. When talking about mortality, however, a 0.1% increase in risk is considered significant. In the study population this amounted to about 20 extra perinatal deaths.

Dr. Tuteur also points out:

“MANA and homebirth midwives have been lobbying extensively for a scope of practice that includes breech, twins, VBAC, etc.”

These are the same high risk patients that had an even higher mortality rate according to MANA’s own data.


The birthing process is a very personal and emotional topic for many people. A discussion about the ethics, risks and benefits of home birth vs hospital birth is a complex topic. However, any such discussion should be informed by a honest treatment of the data.

The MANA data is legitimately concerning. Trying to whitewash such results paints MANA in a very bad light. If they wish to promote midwifery they need to be honest about and own those results, and explain why they believe homebirths are still a reasonable option.

Also, there are certain conditions for which the home birth mortality rate was significantly higher than the average. A reasonable (non-ideological) position, at the very least, would be to specifically recommend that such high risk births occur in a hospital, rather than lobbying for expanded home birth.

14 responses so far

14 thoughts on “Death Rate for Home Births”

  1. Kawarthajon says:

    Sh*t. I had just commented a few days ago on another of your stories about how safe home births are (for uncomplicated pregnancies) and this is why that my wife and I decided to have both of our children at home.

    Is it possible that this differs between countries (i.e. due to differences in midwifery training, or other health care factors)?

    Here’s a story about how safe homebirths are in Canada: http://www.cbc.ca/news/technology/home-birth-with-midwife-safe-as-hospital-1.862485

    And the study from the CMAJ: http://www.cmaj.ca/content/181/6-7/377.full

  2. TheFlyingPig says:

    “and 1.61 for low risk births. Amy Tuteur (the Skeptical OB) calculates from the CDC database that the same statistics for planned hospital births are 0.38 per 1000 for low risk births. That’s 4.2 x higher.”


    “In absolute terms this is about a 1.3% increase in risk.”

    Am I missing something? Isn’t it a 0.123% increase in risk?

    1.61 per 1000 = 0.161% vs. 0.38 per 1000 = 0.038%
    0.161% – 0.038% = 0.123%

  3. sonic says:

    It appears the mortality rate, neonatal (per 1000 live births) is around 4 in both Canada and the US and has been for each year 2009-2012.

    Here is a fuller explanation of the stats (US data)–

    “According to preliminary data, in 2009, 17,298 infants died before reaching 28 days of age, representing a neonatal mortality rate of 4.19 deaths per 1,000 live births. Although this represents a 1.9 percent decrease from 4.27 per 1,000 live births in 2008, this change was not statistically significant.

    Neonatal mortality rates vary by race and ethnicity. Based on preliminary data for 2009, the neonatal mortality rate among infants born to Black women (including Hispanics) was 8.20 per 1,000 live births, more than twice the rate among infants born to non-Hispanic White and Hispanic women (3.43 and 3.62 per 1,000, respectively).”

    I think most home births are done by white women here in the US- so perhaps the comparison is to white only births, which leads me to the hypothesis that the .38 number has a misplaced decimal.

  4. sonic says:

    My hypothesis is false-
    On further inspection the .38 number has to do with “we are looking at presumably low risk white women, ages 20-44, at term, with babies that are not growth restricted”.
    She is dividing the numbers much more finely than those I looked at.
    My hypothesis is false.

  5. Kawarthajon, while home birth is always riskier than hospital birth, in Canada the differences are less significant than in the US.

    In Canada, midwives are trained, supervised and work with hospitals. They are not allowed to take on high-risk pregnancies.

    In the US, certified or direct-entry midwives have only recently been required to have a high school diploma as part of their qualifications. Many homebirth midwives practice without even that minimal level of certification. They would not be allowed to practice in Canada. The vast majority are unaffiliated with hospitals and don’t carry insurance.

    The situations in the two countries are very different.

  6. thechadwick says:

    There are a couple of issues with this:
    1. The population of people getting home births are already less likely to receive other medical treatments (vaccinations, antibiotics etc…), putting them at a greater risk of dying in the first place. As such the variable of home-birthing isn’t adequately controlled for.
    2. I hear it is difficult to get malpractice insurance to cover homebirths, and in some areas it’s even illegal. As such, the most competent practitioners are less likely to help. If the same study was done in a country with fewer repercussions to the practitioners willing to help, the results will likely be different. The argument is that the homebirthing process is dangerous because society makes it dangerous out of fear, not because it is an inherently dangerous process.

  7. flyingpig – you are right. I meant 1.3 per thousand, which is .13% (corrected in the text)

    The data is definitely complex, and we will never have randomized blinded data, only observational, which is plagued by confounding factors.

    I do have a problem with the way MANA is whitewashing over all this and simply concluding that there is nothing to see here, while pushing for permission to deliver high risk women at home.

    Rather, at the very least, the data should encourage caution and perhaps improved regulations and restrictions.

  8. Kawarthajon says:

    Alison Cummins:
    “The rate of deaths per 1,000 births in the first month of life was 0.35 for the planned home births, 0.57 for the hospital births with a midwife, and 0.64 with a physician, the researchers found.”

    This Canadian study seems to suggest that it was much safer having a planned home birth than a physician assisted birth in hospital, this for non-complicated births. Almost twice as safe at home than in the hospital.

    As for the training, you are correct. In Canada you need an undergraduate degree in a health-related field and then there are 4 extra years of midwifery education in university, as well as a residency. The amount of training they receive is comparable to that of doctors – in fact the Ontario Midwives Association is currently arguing that they should be paid the same as doctors, as they have a similar level of training and responsibility, or at the very least as much as nursepractitioners. If midwives only have high school in the US, that might explain the difference in mortality.

    When we had our home birth, the midwives were excellent and really seemed to know what they were doing. We also had to register at a nearby hospital ahead of time, just in case there were any complications and needed to be rushed for emergency care.

  9. embeetee says:

    > The amount of training they receive is comparable to that of doctors

    I suspect you’ll find a *lot* of argument on that position if it’s not specifically referencing the narrowed scope of study; it could at least be reasonably debated with that limitation.

    > In Canada you need an undergraduate degree in a health-related field and then there are 4 extra years of midwifery education in university

    At UBC at least, you need to meet standard (high school) entry requirements, plus a course on how to write English at a university level and a university level anatomy and physiology course which must be completed within 5 years of beginning the midwifery programme; no previous undergrad.

    http://www.skepticalob.com/2009/09/new-canadian-study-is-bad-news-for.html has a link to the full study in the story.

    As with many complex issues, we must not be quick to jump to conclusions from statistics. There could be many reasons to account for the lower levels of interventions in home births; that a midwife-assisted home birth is inherently better is only one possibility. From the study, for example (and there are others that could be raised):

    “We cannot exclude the possibility that differences in findings between the groups were attributable to unmeasured characteristics of the women who chose home birth. Although our study cohorts were closely matched on prognostic variables, we do not underestimate the degree of self-selection that takes place in a population of women choosing home birth.”

    > it was much safer having a planned home birth than a physician assisted birth in hospital, this for non-complicated births. Almost twice as safe at home than in the hospital.

    The study actually concludes “Our study showed that planned home birth attended by a registered midwife was associated with very low and comparable rates of perinatal death and reduced rates of obstetric interventions and adverse maternal outcomes compared with planned hospital birth attended by a midwife or physician.”

    I don’t dismiss midwifery as a birthing option in Canada (ie. with regulated education, licensing and health/selection conditions which must be met for home birth), but I certainly wouldn’t leap to it as obviously better and safer.

  10. clsh says:

    Rewiew from Cochrane Library, 2012, concludes that for low risk births “assisted by an experienced midwife with collaborative medical back up in case transfer should be necessary” there is “no strong evidence from randomised trials to favour either planned hospital birth or planned home birth for low-risk pregnant women.”

    The review finds that there are less interventions and minor complication for home births.


  11. SimonW says:

    I have a comment somewhere on Amy’s blog pointing out that using the NHS’s (much abused in the US press) use of pricing to ration medical care on the basis of QALYs, it would be cost justified to give all your midwives the longer training your best midwives get.

    It is a small absolute risk, but when babies die you lose a LOT of good years.

    Ironically one of the local maternity units to me does very well with home births, although largely by strict enforcement of “any complications go to hospital”, and a small region with very good Ambulance coverage so if something does go wrong they are in the operating theatre relatively quickly (hardly anywhere is more than 10 miles from the hospital).

  12. BillyJoe7 says:

    I was delivered by home birth and I turned out awe write.

  13. Bruce says:


    You actually bring up a very valid point. I work in local government quite closely with the NHS, and while my job is primarily “older people” (ie 65+), there is a VERY big drive right now to provide as much care at home as possible.

    I would imagine that extends across the NHS (possibly not) but even so I think they would very much prefer people to deliver at home as it would be cheaper and less strain on the hospital services IF it were as safe or relatively as safe and if that lack of safety could be covered in the cost to the emergency services. As things stand, I think having someone in hospital puts less potential strain on emergency services.

    Put bluntly, while delivering at home might be safe, and the cover by emergency services might mitigate some of the risk, is the strain on emergency services more critical than the strain on hospital services? I suspect not an easy thing to answer, but knowing the NHS and their metrics, someone, somewhere has probably crunched the numbers, hence the preference for hospital births.

Leave a Reply