24h-payday

The Wax Home Birth Meta-Analysis: An Outsider’s Critique

October 23rd, 2012 by avatar

Today’s post is a fascinating interview that took place between Rebecca Dekker, PhD, RN, APRN of Evidence Based Birth and Kyoung Suk Lee, PhD, MPH, RN, APRN. Rebecca asked Dr. Lee to provide a review of the Wax Home Birth Meta-Analysis, as an “unbiased outsider”, but highly skilled researcher.  Dr. Lee’s comments and critique are fascinating and provided me with many further thoughts.  Please enjoy Rebecca’s interview and share your comments. – SM

__________________

http://www.flickr.com/photos/eyeliam/7353095052/

Shortly after starting my website, www.evidencebasedbirth.com, I had several people ask me if I could write an article about the research evidence on home birth. However, I was hesitant to do so for several reasons. Mainly, I was worried that I could not look at the evidence in an objective manner. My husband and I had recently chosen a home birth for our second child. I was worried that it would be difficult to objectively examine the research evidence on home birth, given my personal experience. The blogosphere is full of people who are strongly pro-home birth or anti-home birth, and their evaluations of the evidence are usually written through the lens of their own biases. I didn’t want to add to the plethora of biased articles already out there.

Then I had a sudden burst of inspiration. What if I asked one of my colleagues—who has no biases about childbirth—to review the home birth literature for me? In particular, I wanted to find someone who could review the Wax home birth meta-analysis (Wax, Lucas et al. 2010) and give me a fair assessment of its scientific value.

I chose the Wax meta-analysis for this review because in 2011, the American Congress of Obstetricians and Gynecologists emphasized the results of the Wax study in its official statement on home birth. Their statement said: “Women inquiring about planned home birth should be informed of its risks and benefits based on recent evidence. Specifically, they should be informed that although the absolute risk may be low, planned home birth is associated with a twofold to threefold increased risk of neonatal death when compared with planned hospital birth.”(ACOG, 2011)

Dr. Kyoung Suk Lee, PhD, MPH, RN, APRN

It did not take me long to figure out who I would ask to review the Wax study. Dr. Kyoung Suk Lee is considered by her colleagues to be a rising star in the field of cardiovascular research. She recently graduated with a PhD in Nursing, and she just accepted a job at a research university. People who work with Dr. Lee say that she is extremely intelligent, hard-working, and a future leader in her field. Dr. Lee’s expertise has been recognized with research awards from the Heart Failure Society of America, the Society for Heart-Brain Medicine, and the Cleveland Clinic Heart-Brain Institute, among others. She has published her work in nursing and cardiology journals. Furthermore, I knew that Dr. Lee did not have any biases about childbirth, home birth, or hospital birth. I asked Dr. Lee if she would be willing to review the Wax meta-analysis for me, and she kindly agreed.

What follows is my interview of her about the study and its results (RD in bold, KSL unbolded).

Do you have any biases or conflicts of interest related to home or hospital birth?

I do not have any biases related to home or hospital birth.

Could you summarize the methods and results of the Wax study?

The purpose of this meta-analysis was to compare maternal and neonatal outcomes between planned home-and hospital-births.

Using an electronic database search and bibliography search, the authors retrieved 237 articles and included 12 articles in their meta-analyses. Of 12 articles included, 3 were conducted after 2000 while 9 were conducted before 2000. Of 12 articles, 2 were conducted in the US (one was a retrospective design) while 10 were conducted outside US.

Women in the planned home birth group had better maternal outcomes than women in the planned hospital group. They had fewer interventions such as epidurals and episiotomies, and lower morbidity (infection, 3rd or 4th degree lacerations, hemorrhages, and retained placenta). There were no differences in cord prolapse between the two groups.

For neonatal outcomes, babies born to women in the planned home birth group were less likely to experience prematurity and low birth weight. However, babies born to women in the planned home birth group were more likely to experience neonatal death compared to women in hospital birth.

What is the difference between neonatal and perinatal mortality? What does this have to do with the results?

Based on the definitions given by the authors, neonatal mortality was defined as “death of live born child within 28 days of birth.” This is a subset of an overall outcome– perinatal mortality, which was defined as “stillbirth (of at least 20 weeks or 500g) or death of live born child within 28 days of birth.”

According to the authors, there were no differences in perinatal death (the overall outcome) between planned home birth and hospital birth groups. However, homebirth was associated with 2 times higher risk for neonatal death (the subset of deaths occurring 28 days after birth) in all infants and 3 times higher risk for neonatal death in infants who did not have any congenital birth defects.

Interestingly, if you look at page 243.e3, the authors did a sensitivity analysis. In this analysis, they excluded the studies that had home births that were not attended by certified midwives or certified nurse midwives. In this analysis, they found that there were no differences in neonatal deaths between the home birth and hospital birth groups. This means that in the studies in which midwives with certification of some kind attended home births, the outcomes were the same except there was no increase in the neonatal death rate. In my opinion, we have to pay attention to results of sensitivity analyses because this allows us to see the results based on studies which were definitely known to be eligible or clearly described their methods and outcomes.

What is your opinion on the scientific rigor of this meta-analysis?

One thing that was strange to me is the odds ratios (ORs) in the tables. For example, in table 2, under morbidity, the percentages of infection between home births and hospital births were 0.7 vs. 2.6 (its OR was 0.27) while percentages of perineal laceration were 42.7 vs. 37.1 (its OR was 0.66). To a researcher, these numbers don’t make sense.

Many of the studies included were older (half of the studies were conducted more than 20 years ago) so results may not reflect the current practice at home births or hospital births.

The authors did not provide detailed information on how they evaluated the quality of studies included, although they cited a paper describing the method of study evaluation. This makes it difficult if not impossible to determine whether the studies they included were of good or poor quality.

The authors mentioned that women with high risks would prefer hospital births so that it would expect that home births have better outcomes than hospital births in some maternal and neonatal outcomes. If this was a concern, I wonder why the authors didn’t just focus on only the studies that used matching methods, in order to minimize confounding factors.

What is the difference between relative risk and absolute risk, and how does that apply to women who want to have a home birth?

Absolute risk is the probability of something occurring. They may be expressed as percentages or ratios. For example, neonatal mortality rate in the United States is 2.01 per 1,000 live births. This is .201 percent (2.01/1000 = .201/100).

http://www.flickr.com/photos/mikeporcenaluk/3789756395/

Relative risk is a comparison between different risk levels, such as the neonatal mortality rate of home birth compared to the neonatal mortality rate of hospital birth. The researchers found that there was a higher relative risk in neonatal mortality at home births compared to hospital births, but the overall absolute risk for both was small.

How can women know whether the Wax study results would be applicable to their own individual situation?

Meta analysis is one way to generalize findings from different studies. However, women and clinicians should interpret these results cautiously because the studies included were very different from one another and some of the studies included may not have been of good quality. Also, it would be important to note that the overall neonatal death rate that they report reflects home births that were attended by midwives as well as those that may not have had any kind of certified midwife present.

Because this study seems to have some flaws, the conclusion is tentative. I do not know if this article has any implications for pregnant women.

What do you think is the value of asking someone with no conflicts of interest to evaluate controversial research? Does Dr. Lee’s even-handed critique make you view the results of this study any differently? How do you feel about Dr Lee’s conclusion that the study’s results are tentative, and that the Wax study might not have any implications for pregnant women? Please share your thoughts and comments with other readers.

References

(2011). “ACOG Committee Opinion No. 476: Planned home birth.” Obstetrics and gynecology 117(2 Pt 1): 425-428.

Wax, J. R., F. L. Lucas, et al. (2010). “Maternal and newborn outcomes in planned home birth vs planned hospital births: a metaanalysis.” Am J Obstet Gynecol 203(3): 243 e241-248.

About Rebecca Dekker

Rebecca Dekker, PhD, RN, APRN, is an Assistant Professor of Nursing at a research-intensive university and author of www.evidencebasedbirth.com. Rebecca’s vision is to promote evidence-based birth practices among consumers and clinicians worldwide. She publishes summaries of birth evidence using a Question and Answer style.

Babies, Childbirth Education, Evidence Based Medicine, Guest Posts, Home Birth, Metaanalyses, Midwifery, New Research, NICU, Research , , , , , , , , , , , ,

  1. October 23rd, 2012 at 10:57 | #1

    More info on the Wax study from Science & Sensibility:

    Jumping to conclusions: Popular media spins an abstract into headlines. http://www.scienceandsensibility.org/?p=4118

    When Scientific Methods Fail: New Criticisms Over the Wax et al Homebirth vs. Hospital Birth Study http://www.scienceandsensibility.org/?p=2551

    Becoming a Critical Reader: Questions to ask about systemic reviews and meta-analyses http://www.scienceandsensibility.org/?p=1610

    Planned home birth and neonatal death: Who do we believe? http://www.scienceandsensibility.org/?p=1422

    Meta-analysis: the wrong tool (wielded improperly) http://www.scienceandsensibility.org/?p=1349

  2. October 23rd, 2012 at 12:46 | #2

    Thank you, Deena! I know Science and Sensibility has written on these topics before, and I’m grateful to Sharon for letting Dr. Lee and I look at the evidence from an outsider’s perspective.

  3. October 23rd, 2012 at 13:25 | #3

    I’m grateful too! It’s always good to hear a fresh perspective :-) @Rebecca Dekker, PhD, RN, APRN

  4. avatar
    Denise Hynd
    October 23rd, 2012 at 16:11 | #4

    Can’t wait for your response to the UK Birth Place study?

  5. October 23rd, 2012 at 19:54 | #5

    Thanks for this great approach to looking at the Wax research. Hearing from unbiased yet well versed in research sources is a great way to help us all look at the evidence free of biases. Really appreciate this and all your work!

  6. avatar
    Ariann
    October 23rd, 2012 at 20:01 | #6

    Am I correct in combining the answers to the second and third questions and concluding that for women having a planned home birth AND using a certified midwife, that both maternal and neonatal incomes are improved in multiple ways in general and perinatal/neonatal death rates are identical to planned hospital births?

  7. avatar
    Wendy Gordon, LM, CPM, MPH
    October 23rd, 2012 at 20:04 | #7

    Thank you for exploring this approach to critical analysis of the research! Dr. Lee made a very, very important observation about the sensitivity analysis that ACOG and others conveniently ignore — in calculating rates of neonatal mortality, Wax & colleagues relied heavily on the infamous Pang study from WA State that included *unplanned* and *unattended* home births in their analysis. Pang’s research was problematic for this reason (and many others) when it was published in 2002, and it’s still problematic in Wax’s meta-analysis. Wax appropriately did that sensitivity analysis, but incorrectly concluded in their bottom line that “planned” home birth resulted in a threefold increase in neonatal mortality. Convenient for ACOG and others who are politically opposed to home birth, but not an ethical or accurate conclusion. For people who don’t look at or understand the fine print in the sensitivity analysis (like the media), this important point went unchallenged outside of homebirth circles. Thank you, Drs. Dekker & Lee!

  8. October 24th, 2012 at 00:00 | #8

    @Denise
    I did an analysis of the U.K. study for Science and Sensibility: http://www.scienceandsensibility.org/?p=3787.

  9. October 24th, 2012 at 06:22 | #9

    There have been so many critiques of the wax study that have pointed out the flaws that invalidate it. While it is nice to have an independent opinion, the one thing she would miss are all the studies that were not included. It would be known, for instance, by someone in the business, that there were 3 major, large recent studies purposely left out of the Wax meta-analysis. I would be curious as to Dr. Lee’s opinions if this were brought to her attention. It is very significant since all 3 of theses large, recent studies clearly showed that for low risk women, homebirth with a certified Midwife was as safe if not safer than hospital birth. She does hint at this element, but does not realize that this important data was left out of the analysis.
    References:
    Johnson, KC, Daviss, BA; “Outcomes of planned home births with certified professional midwives: large prospective study in North America”; British Medical Journal, 2005;330:1416. http://bmj.bmjjournals.com/cgi/content/abridged/330/7505/1416

    Jassen, et al; “Outcomes of planned homebirth with registered Midwife versus planned hospital birth with midwife or physician”. CMAJ 2009. DOI:10.1503/cmaj.081869

    de Jonge A, van der Goes B, Ravelli A, Amelink-Verburg M, Mol B, Nijhuis J, Bennebroek Gravenhorst J, Buitendijk S. Perinatal mortality and morbidity in a nationwide cohort of 529 688 low-risk planned home and hospital births. BJOG 2009; DOI: 10.1111/j.1471-0528.2009.02175.x.

  10. October 24th, 2012 at 06:25 | #10

    @Ariann
    Yes.

  11. October 24th, 2012 at 06:28 | #11

    @Wendy Gordon, LM, CPM, MPH
    Well put, Wendy! The Pang study is so flawed as to be useless for purposes of determining safety in homebirth. I am surprised anyone reputable would quote it other than for critical reasons!

  12. October 24th, 2012 at 06:34 | #12

    I posted a link to the Homebirth annotated guide to the literature on the Rochester Area Birth Network website’s homebirth page:
    http://rabn.org/wordpress/wp-content/uploads/2012/09/Home-Birth-Annotated-guide-to-the-literature-May-2011.pdf

  13. October 24th, 2012 at 11:57 | #13

    I guess I have to say I agree with all the comments so far! @Ariann, yes, if you use the results of the sensitivity analysis, then home birth had better maternal outcomes and equivalent neonatal outcomes to hospital birth. @Amy Haas, yes, that is one of the problems of being an outsider– you might not know the context of the situation. Dr. Lee did not know about the other studies, and I purposefully did not tell her ahead of time about studies that were excluded, or about the problems with the Pang study that was included in this meta-analysis. I did not want to influence her opinion, and I simply wanted her to simply evaluate the merits of this particular study alone.

    There are some benefits to being an insider when evaluating research– but you don’t have the true objectivity of an outsider. There are pros and cons to both approaches. I felt like there had been hundreds of “insider” opinions on the Wax study and that it would be unique to get that objective viewpoint.

  14. avatar
    monica
    October 25th, 2012 at 23:33 | #14

    Thank you for this. I would love to see more unbiased analyses of the other major studies (especially the 2005 BMJ study – this would balance out the “pro” and “anti” studies). I had a home birth for my first, at which time I felt certain it was the best, and safest route. Now, I am just not as sure, and am having my second baby in April. I’m undecided where to give birth, and confused! I wish I had the skills to analyze these studies myself. Looking forward to more!

  15. avatar
    Nora Alterman
    October 26th, 2012 at 12:47 | #15

    Excellent interview. Thank you!
    Regarding the sensitivity analysis of certified midwives – I actually approached Wax to inquire if they had examined all outcomes in this analysis and was answered that indeed all variables from tables 2&3 of the study were examined and findings were similar to the full study, except for the noted result of similar neonatal mortality.
    So if one were to accept the validity of the conclusion of this meta-analysis, one would also have to agree that it shows that planned homebirth with a certified midwife is as safe as hospital birth, while at the same time much healthier for mother and baby.
    The review’s sensitivity analysis shows that homebirth has significantly fewer of the following interventions:
    Epidural analgesia
    Electronic fetal heart rate monitoring
    Episiotomy
    Operative vaginal delivery
    Cesarean delivery
    Significantly fewer of the following adverse outcomes:
    Infections
    ≥3-degree lacerations
    Perineal and vaginal lacerations
    Hemorrhages
    Retained placentas
    And the offspring are less likely to be:
    Born preterm
    Be of low birth-weight
    Require assisted ventilation
    Born before 42 weeks
    While perinatal and neonatal mortality is the same as in hospital births.

  16. avatar
    Nora Alterman
    October 26th, 2012 at 13:22 | #16

    I heard that following the many critiques published (I believe something like 11), the magazine was approached in a request to subtract the paper. Does anyone know what happened with this?

  17. avatar
    samantha
    October 29th, 2012 at 06:54 | #17

    Does the study include figures for maternal mortality? I”m wondering how hospital and home-birth maternal mortality rates differ?

  18. November 3rd, 2012 at 09:41 | #18

    Hi – Ina May Gaskin did a thorough analysis of the biases of the Wax study in her book Birth Matters: A Midwife’s Manifesto. After reading Ina May’s analysis, it is odd to me that the journal could get away with not retracting it. The authors of some of the studies reviewed wrote into the the journal and protested their data was mis-used and mis-represented.

  19. avatar
    Deborah
    March 7th, 2014 at 19:06 | #19

    Your interview choice was a great idea. I too have Ina May’s analysis of the wax study and I can look at the numbers myself, but this interview sold me. Every time I help a mother I have got to know I am doing the absolute safest thing with the most up to date info.

  20. avatar
    Elaine
    April 3rd, 2014 at 23:24 | #20

    Interesting summary! I don’t have a bias either way, really, but it would be interesting to look at the more recent MANA and CDC studies, which were somewhat large in scale (though of course, each were flawed). Both showed an increase in neonatal mortality in homebirths vs. hospital. First time moms and babies in breech were much more likely to be rushed to the hospital and/or have poor outcomes as well, though for non-first time moms and non-breech, the numbers were a little closer.

    Also, I think it’s a little odd to laud the “fewer interventions,” such as constant fetal monitoring, but especially epidural analgesia. Many women choose these interventions, they can’t often be provided during home births, so of course women who want them go to hospitals. I think about 70-90% of women opt for an epidural during birth. This is not necessarily a negative thing–it’s the “customer” wanting a certain experience and getting it.

  1. October 23rd, 2012 at 09:32 | #1
  2. November 12th, 2012 at 19:06 | #2
  3. February 1st, 2013 at 23:19 | #3