A new era of home birth research

In preparing the Home Birth chapter for the forthcoming second edition of Obstetric Myths versus Research Realities, I have literally just finished reading the entire body of literature on planned home birth. Just last week, I said to my co-author, Henci Goer, “frankly, I’m pretty underwhelmed by the quality of most of the studies.” (Though, don’t get me wrong, I still believe that the preponderance of the evidence strongly favors the choice of planned home birth.) But for the second time this year, an exemplary study on planned home birth has been released. Together with the Dutch study released in April, the current study ushers in a new era of home birth evidence that addresses many of the methodological limitations of previous home birth research. Seriously, folks, these two studies raise the bar.

Researchers in Canada analyzed the outcomes of all women who were intending to give birth at home at the onset of labor in British Columbia between 2001-2004 (n=2899 women). Data were obtained from the provincial database that collects information on all births and is cross-referenced with the national vital statistics (birth/death certificates) database. The researchers compared outcomes in the planned home birth group with those of two groups of women who met eligibility requirements for home birth but planned to give birth in hospitals instead. One of the two comparison cohorts had planned hospital births with midwives (n=4752); the other with physicians (n=5331).

Consistent with many other studies comparing planned home with planned hospital birth, the results showed comparable perinatal mortality rates, less serious morbidity for both women and infants, and lower use of obstetric technology in planned home births. Here are the results, as presented in the study’s abstract:

The rate of perinatal death per 1000 births was 0.35 (95% confidence interval [CI] 0.00–1.03) in the group of planned home births; the rate in the group of planned hospital births was 0.57 (95% CI 0.00–1.43) among women attended by a midwife and 0.64 (95% CI 0.00–1.56) among those attended by a physician. Women in the planned home-birth group were significantly less likely than those who planned a midwife-attended hospital birth to have obstetric interventions (e.g., electronic fetal monitoring, relative risk [RR] 0.32, 95% CI 0.29–0.36; assisted vaginal delivery, RR 0.41, 95% 0.33–0.52) or adverse maternal outcomes (e.g., third- or fourth-degree perineal tear, RR 0.41, 95% CI 0.28–0.59; postpartum hemorrhage, RR 0.62, 95% CI 0.49–0.77). The findings were similar in the comparison with physician-assisted hospital births. Newborns in the home-birth group were less likely than those in the midwife-attended hospital-birth group to require resuscitation at birth (RR 0.23, 95% CI 0.14–0.37) or oxygen therapy beyond 24 hours (RR 0.37, 95% CI 0.24–0.59). The findings were similar in the comparison with newborns in the physician-assisted hospital births; in addition, newborns in the home-birth group were less likely to have meconium aspiration (RR 0.45, 95% CI 0.21–0.93) and more likely to be admitted to hospital or readmitted if born in hospital (RR 1.39, 95% CI 1.09–1.85).

What makes this study stand out from most  of the rest:

1. Planned home births occurred in a context with relatively rigid guidelines for eligibility (see the full-text of the article to view the guidelines). These guidelines, determined by the Ministry of Health, were applied to women planning hospital births in order to construct the comparison cohorts. This increases the likelihood that, medically at least, the populations were similar. In addition, participants in the home birth group were matched with participants in the physician-attended hospital birth for the following parameters: year of birth, parity, single parent (yes or no), maternal age, and the hospital where the midwife conducting the index home birth had hospital privileges.

2. The authors made the cohorts more similar with statistical analysis – first, controlling for confounding variables and second, testing and retesting their data with different assumptions and exclusions. (Referred to as “sensitivity analysis,” this is a mechanism used by researchers to make sure their findings hold up under various circumstances and are unlikely therefore to be biased.)

3. The researchers isolated the effect of the birth setting itself by comparing midwife-attended home birth with midwife-attended hospital birth. In fact, the same group of midwives cared for women in both settings, so differences are likely to be related to the setting and its protocols and technological accoutrements (or lack thereof) rather than differences in the providers who actually provide the care.

4. According to the study authors, midwives in British Columbia are required to offer medically eligible women a choice of planned home or hospital birth. While women still must self-select to one choice or another, this certainly mitigates some bias. Self-selection bias refers to the possibility that individuals who, in this case, select different birth settings or providers, may be different in ways that are not measured but that nonetheless affect the likelihood of important outcomes. For example, women who choose home birth may have better nutrition, stronger family support systems, or a more positive outlook on labor and birth. But these differences may be less pronounced when the group selecting home birth showed up at the same prenatal clinic as all of the rest of the women and were (perhaps enthusiastically) presented the option of birthing at home. Contrast this with the population of women in parts of the United States who must actively seek out home birth, pay out of pocket, and be told that no physician will willingly consult if medical problems arise so they must settle for the emergency room, whether or not the transfer to the hospital is urgent. (The vast majority of transfers from home to hospital are not.) Only the most dedicated are likely to choose such an option.

5. Although the study was retrospective (meaning data were collected after the fact), data were obtained from province-wide databases to which care providers are obligated to enter data on each and every birth. In addition, these databases have been tested for the reliability of the data (to detect the possibility that certain outcomes are systematically underreported or overreported). Reliability was above 97% for all outcomes. Fewer than 1 in 10,000 records were missing. *

The only thing I did not see in the report that I would have liked to is a detailed description of the circumstances of each fetal or newborn death. These descriptions often provide clues as to whether small differences in perinatal mortality could have been attributed to the planned place of birth and whether and how they might have been averted.

No study of home birth will be perfect, but large perinatal databases and systems that integrate rather than marginalize home-birth midwifery have helped to achieve the “scientific rigor” that the American College of Obstetricians and Gynecologists has called for. And the results are looking very good indeed.

* denotes edited text. Click “read more” to view the original version. Edited September 2 at 10:57pm due to error in original (see comment by @desiree and my response.) Deleted text: And although data, including the intended place of birth, are submitted after the baby is born, the researchers present compelling evidence that perinatal mortality rates are reliable regardless. They note that even had all of the perinatal deaths occurring in the midwife-attended hospital cohort actually occurred in women intending to birth at home (but reported in the database to have planned hospital births), the perinatal mortality rate among women planning home birth including these deaths would still have been lower than the rate in the physician-attended hospital cohort.

Lamaze News, New Research, Practice Guidelines, Science & Sensibility, Systematic Review, Uncategorized , , , ,

  1. | #1

    Thanks for this write-up–it’s really useful. (And just plain neat.)

  2. | #2

    I am so glad you jumped on this. Thanks.

    I had no idea that BC midwives were required to offer home birth or hospital birth. Amazing.

  3. avatar
    | #3

    I’ve been hoping you’d talk about this! Thanks.

  4. | #4

    I was so jazzed to see this study!!! One of my email lists I’m on had emailed out the link to a news article, but not the study, so I looked up the where the study was printed to see if I could find the abstract — totally not expecting to see the actual article — and was thrilled to find the full study! How cool is that?!

    Plus, I’ve already blogged about it, and I’m pretty sure Dr. Amy reads my blog regularly — I can only imagine the frantic cogitations she’s doing trying to ferret out some problem or inconsistency in this study. But so far, she hasn’t gone viral with comments about how this study is “fatally flawed, and besides, hospital birth is safer — trust me!”

    Oh, yeah, I’m having a blast!! :-) So excited!!

  5. avatar
    | #5

    This is such a great post.

    I do have a question though. I’m sure you’re familiar with “dr. amy”. She’s a train wreck. The kind you can’t look away from or you might miss something gorey.
    Ms Romano.. what’s your opinion of her website, and the statistics she puts up. I’m not exactly sure what her problem is, or if she’s just a bored house wife. But I’d like to hear your thoughts on the statistics she drudges up.

  6. | #6

    Oh I was excited to hear there is a second editin of Obstetric Myths vs. Research Realities coming out. Do you know what the expected publication date is?

  7. avatar
    | #7

    Thanks so much for this thoughtful analysis. It’s wonderful to see more and more “airtight” home birth research emerging.

  8. | #8

    @Jill–UnnecesareanAnother reason to move back to Canada! Thanks Amy for this excellent summary and analysis. I’m quoting you in a piece that I’m going to post shortly on Huff Po.

  9. avatar
    | #9

    hm, you wrote:
    They note that even had all of the perinatal deaths occurring in the midwife-attended hospital cohort actually occurred in women intending to birth at home (but reported in the database to have planned hospital births), the perinatal mortality rate among women planning home birth including these deaths would still have been lower than the rate in the physician-attended hospital cohort.

    but that’s not what the study says. it says that:

    In the worst-case scenario,
    if all perinatal deaths attributed to planned hospital birth in
    the midwifery comparison group had actually been planned
    home births, our perinatal death rate of 4 per 2882 live and
    stillbirths would have been 1.4 per 1000 in the home-birth

    since the study quotes a 0.64/1,000 mortality rate for physician-assisted hospital birth, the homebirth mortality rate in this event would be more than twice as high. where did you get that putting all the deaths in the homebirth group would give a mortality rate still lower than the physician group?

  10. | #10

    You’re correct. I had trouble sorting out the calculations the authors were reporting in this paragraph because of their awkward wording of “4 per 2882 live and stillbirths.” Basically, I got mentally thrown off and misinterpreted what they were saying. The situation was hampered by the fact that the authors do not report the actual number of deaths in each cohort, another problem I had with the paper. They do, however, state that no perinatal deaths occurred in planned home births that actually occurred at home. In other words, all perinatal deaths that occurred in the planned home birth group occurred en route to or after arrival at the hospital. This is a sincere error on my part, and I sincerely thank you for bringing it to my attention. With that said, I don’t think any of this (their reported findings or my misinterpretation of them) should affect readers’ overall assessment of the study, because I don’t think it is a critical point to the authors’ findings or conclusions. This was a form of “sensitivity analysis” (which I refer to earlier in the post), and what the authors were trying to demonstrate here was, indeed, the “worst case scenario” (data-wise) – the possibility that all of the instances of the outcome of interest (perinatal death) were miscoded in the database, which is not a strong possibility in the first place.

    I’ll fix the text. Thanks for reading my blog.

  11. | #11

    Thanks Louise! Everyone should go read Louise Roth’s great piece in The Huffington Post!

  12. | #12

    Henci and I have our fingers crossed for a late 2010 release date. After nearly 3 years of work on the book (rewriting every word and reorganizing the structure), you can bet that we will be more than eager to let folks know when it is ready and available. So stay tuned!

  13. | #13

    As difficult as it is for me to refrain, I don’t think I am going to go on record with my opinion of Dr. Amy Tuteur here. Other than that I think, as Jennifer Block (who has actually me the good doctor) wrote in her piece on Babble last month, “For some reason, which [she] never quite got to the bottom of, [Tuteur] believes in every cell that Home Birth Kills Babies (that’s in fact the title of her most recent post on her new site, The Skeptical OB), and no amount of research evidence will convince her otherwise.” I know she has visited this blog (she has commented in the past), so we’ll see if she decides to engage. Last time she did, I was actually appreciative. The conversation forced me to clarify my points and generated some interesting and passionate discussion and in the end I don’t think it made her look very good.

    OK, so maybe I just gave a bit of my opinion.

  14. | #14

    I got wind of this study by reading your post about it, Kathy – so thanks! I thought I was done for the night until I checked Facebook one last time, only to discover a link to your blog and that the Canadian study I’ve been waiting for had finally been published. I ended up working on it into the wee hours. So thanks for being so timely (I think?)! And thanks for your post which was really thoughtful, especially considering how fast you got it out there. :)

  15. | #15

    @Amy Romano

    Glad I could be of service! :-)

    I’d actually been cogitating for some time on the question, “How can home birth be safer, or even as safe as, or at least not so much less safe than, hospital birth?” Not that I disbelieved it, but as an answer to the “hospital/technology/intervention = safe” mentality that is so pervasive in this society. So, I had pretty well thought out a post already, and was able to just sit and type and “let it flow.” And as I said, I was pretty jazzed after reading the article, so I typed more energetically than usual. 😉

  16. avatar
    | #16

    @Amy Romano

    sure thing, i enjoyed your blog post.

    whenever i read studies like this, i wonder why the authors don’t just go that extra step and actually find out whether the intended birth location of the fatalities were properly recorded. i mean, there were 4. why raise the issue and leave it hanging? and, why not look into cause of death? they might all be congenital abnormalities for all we know… it irks me when studies don’t spend that extra bit of effort to tie up these kinds of loose ends.

    finally, what wonderfully low neonatal mortality rates all around! way to go BC.

  17. avatar
    Lee Saxell
    | #17

    I am one of the Canadian home birth study authors. Thanks for all your thoughts and opinions on our study – and the links to other sites discussing it.

    I just want to clarify that we used the provincial data base for our study information. All identifiers were removed before we received the data and we did not have permission to view the actual charts. We could eliminate mortality caused by preterm birth, congenital anomaly etc, but we could not drill down on the deaths occurring after labour began. And sadly, as we all know, there is often no definitive cause for intrapartum death.

  18. | #18

    @Lee Saxell

    Ok, so all of the deaths included in this study were either intrapartum or neonatal deaths? I ask because the term used was “perinatal mortality,” or deaths from 20 weeks of gestation through 7 days of birth. None were, for example, stillbirths at 28 weeks of gestation? They were all term births, and living at the start of labor?


  19. avatar
    Lee Saxell
    | #19

    In Canada, stillbirth death is defined as intrauterine death after 20 weeks gestation. Early neonatal death is from birth to 7 days. Together this time period is referred to as perinatal death and it is a standard mortality rate in Canada.

    Following this first broad data search from 20 wks to 7 days after birth, the three groups were then matched for comparison. All three groups had to meet the eligibility requirements for home birth, regardless of planned place of birth. In order to qualify for a home birth in BC you must be between 37 and 42 weeks gestation so all births occurring between 20 and 36+6 wks gestation were eliminated from the final data set.

  20. avatar
    | #20

    I love this blog! I do a lot of research (evolutionary pediatrics and ethnopediatrics) and a friend recommended that I check out your blog. It’s great! 😀

    Could I have the citations for both studies? Thank you!


  21. | #21

    Aradia, here are the links for the Dutch study http://www3.interscience.wiley.com/journal/122323202/abstract?CRETRY=1&SRETRY=0

    and the British Columbia study http://www.ncbi.nlm.nih.gov/pubmed/19720688

    Thanks for stopping by the blog!

  1. | #1
  2. | #2
  3. | #3
  4. | #4
  5. | #5
  6. | #6
  7. | #7
  8. | #8
  9. | #9
  10. | #10
  11. | #11
  12. | #12
  13. | #13
  14. | #14
  15. | #15
  16. | #16
  17. | #17
  18. | #18
  19. | #19
  20. | #20
  21. | #21

cheap oakleys fake oakleys cheap jerseys cheap nfl jerseys wholesale jerseys wholesale nfl jerseys