Guest PostPregnancy & Childbirth

Guest Post: One More Homebirth Risk Analysis

Last week I wrote a guest post here at Grounded Parents looking at what we can learn from a recently published study in the Journal of Midwifery and Women’s Health on the risks and benefits of homebirths in the U.S. In the days between when I wrote the post and when it was actually posted, Dr. Amy Tuteur asked a statistics professor to write an analysis at Skeptical OB comparing the results in the study to data on U.S. hospital births from the CDC. I wasn’t going to comment on it because I felt like I’d already stepped on a hornets’ nest with my first post, but then i09 picked up the story in a way that suggested that they read neither the original study they were criticizing or even the post at Skeptical OB that they were linking to. I feel like it’s my duty to step in and provide a little more thoughtful review.

So, how do I know that i09 did not bother to read the study or the Skeptical OB post? Their title was “Statistics Professor Challenges Midwives’ Math on Home Birth Safety.” There are so many things wrong just in this title that it’s hard to even know where to begin. First of all, let’s take the term “midwives’ math.” i09 also doubled down on this in the first sentence of their post which started out with “Recently a group of midwives published a paper about birth outcomes for home births…” If i09 had actually bothered to google the authors of the supposedly overturned study, they would have seen that the authors are medical anthropology researchers at Oregon State University. The head author is also a registered midwife in addition to being a medical anthropology professor, though that’s hardly “a group of midwives.”  (Note: I’m not linking to the authors’ bios for their privacy and protection).

Also, if they’d read either the “Midwives’” study or the analysis that they wrote their entire post on, they may have realized that no one is actually challenging anything. The Journal of Midwifery and Women’s Health study did not attempt to make any relative risk comparisons between homebirths and hospital births. All it did was look at data from about 17,000 planned homebirths in the US and report on outcomes for those births. The analysis posted on Skeptical OB then takes the numbers from that study and compares them to CDC risk numbers on hospital births to see if they can tease out whether the outcomes of particular risks are much higher in homebirths than hospital births. No one is overturning anything or “challenging” anything. It’s just one analysis building on another.

If I were to edit i09’s title to make it more accurate, I would write something more along the lines of “Statistics Professor Writes a Blog Post that Builds on Math Regarding Homebirth Risks from a Peer Reviewed Study in a Well-Respected Journal by Researchers from Oregon State University.” But this is probably why I’m not a writer at i09.

Phew. Now that I’ve got that out of the way, I’m just going to jump right into the “statistics professor” analysis at Skeptical OB. (Note: quotes are because I think it’s weird that Prof. Orosz keeps being referred to as “a statistics professor” instead of by her name, not because I’m doubting that she’s really a statistics professor.)

Remember from my previous post, the study in the Journal of Midwifery and Women’s Health only looked at a cohort of women who planned homebirths. It did not have a comparison hospital birth cohort. Brooke Orosz (the previously mentioned statistics professor) then attempts to put together a comparable cohort of hospital births using the CDC’s Wonder Database, which collects data on birth-infant death certificates. The homebirth cohort had very few preterm or low-weight babies so Prof. Orosz removed preterm and low-weight babies from the hospital cohort. She also limited the cohort to only mothers who received some prenatal care, which also makes it a more comparable cohort to the homebirth mothers who also likely received prenatal care.

After a couple sentences about how she chose what filters to use on the CDC data, Prof. Orosz jumps right in to giving us her results for neonatal mortality:

For babies born in a hospital during the period 2007-2010, weighing at least 2500 grams, whose mothers received some prenatal care, the neonatal death rate was 0.71 per thousand. When deaths due to congenital anomalies are excluded, the rate drops to 0.40. The neonatal death rate for MANA STATS babies was 0.77 excluding congenital anomalies (13 deaths out of 16,950 births) and 1.30 including congenital anomalies (22 total neonatal deaths). The neonatal death rate excluding anomalies was significantly higher in the MANA STATS group (p=0.01). In addition, the number of neonatal deaths attributed to congenital anomalies was higher than expected, although the statistical significance of this difference was marginal (p=0.04).

Now, it’s all well and good to calculate a simple p-value to compare the two cohorts, but I’m not even sure why we are at the point of comparing them when I’m not even yet convinced that the cohorts are similar enough to be compared. If this were an actual scientific paper, the author would have given me a whole bunch of data about each cohort so that I can see exactly the ways in which they are extremely similar and the ways in which they differ and how that might affect the analysis, not to mention that this data is coming from two completely different sources. I want to know that all the definitions between the two datasets are the same. I want to know how problematic it may be that the hospital cohort defines “hospital births” as any birth taking place in a hospital while the homebirth cohort defines “homebirths” as a birth that was planned to be at home regardless of where it actually took place. In fact, over 10% of the homebirth cohort gave birth in a hospital. How might this affect the comparison? Was ethnicity or class taken into consideration? Urban vs rural? Were the homebirth mothers more likely to use alternative medicines and therapies in place of science-based interventions? In fact, the p-value part is the easiest part of the analysis, but it is meaningless if the cohorts aren’t similar enough to be compared. I want to see pages convincing me the hospital cohort is a good comparable control cohort and what weaknesses there may be in using that cohort before I see any p-values.

Prof. Orosz then immediately moves on to the comparison of intrapartum stillbirth rates:

The WHO  estimates the intrapartum stillbirth rate across North America as 0.3 per thousand. It is likely that the true number of intrapartum stillbirths among low-risk infants delivered in the hospital is considerably lower. However, in the absence of any other solid data, I will use the value 0.3 per thousand as a maximum reasonable estimate. The intrapartum death rate for MANA STATS sample was 1.30, which is significantly higher. (p<.0001, highly significant.)

Somehow this is an even worse comparison. In the absence of any good control cohort, Prof. Orosz has just taken a big estimated stat that refers to a very large and diverse group of people (all US births) and compared it to a very small, very unique subset (17,000 planned homebirths) and then calculated a p-value and declared it is statistically significant. Of course it is statistically significant. This is a comparison between two completely different groups of individuals and is not controlled in any way.  Not to mention that she makes this comparison without any link to where we can see the WHO’s estimates and evaluate them for ourselves. (EDIT: Commenters pointed out that if you click the link at the introduction to the Skeptical OB analysis where it says “you can find the complete analysis here” it brings you to a copy-paste of the blog post but with a couple sources at the end included the source of the WHO stat. So, when Dr. Tuteur converted it into a blog post she left out the sources at the end, not Prof Orosz)

Professor Orosz concludes the section on neonatal mortality rates with the following statement:

In other words, the expected number of deaths from causes other than congenital anomalies was at most 12, and the actual number of deaths was 35 (44 with anomalies included). It is clear that home birth substantially increases the risk of neonatal death and of intrapartum death.

 

Ariel Facepalm

Look, here’s the thing to remember here. We currently do not have a comparable hospital birth cohort to use to calculate relative risk between homebirths and hospital births. However, the absence of a good control doesn’t help parents who are trying to asses risk factors in order to make a decision on whether to have a home or hospital birth. Therefore, it’s ok to attempt to find a kinda, sorta, maybe, similarish cohort and do some back-of-the-envelope calculations.

The analysis that Dr. Orosz did is perfectly ok to do in the absence of better data. In the real world we don’t always have good data so we have to use whatever we can get. However, if you are going to do these fuzzy analyses, it is imperative you make the weaknesses very clear and use qualifying language. Just because your p-value is statistically significant doesn’t mean there’s actually an effect there if you are using a biased control group (Note: I’m using “biased” in the statistical sense – not as a knock against the author – frankly, I don’t care how biased a person is as long as their data is unbiased).

In this case the cohort groups are very different. They contain completely different types of people and do not even use the same definitions for very simple things like what counts as a hospital birth. In the case of intrapartum death, we don’t even have any cohort comparison group at all. Even so, this is the best data we’ve got. We can compare the numbers to get some rough estimates on potential risk increases, but to claim that “home birth substantially increases the risk of neonatal death” is absolutely not supported by the data. Instead, I would sum up Prof. Orosz’s results for this section as “risk of neonatal death and intrapartum death in the homebirth cohort seem slightly higher than what we generally see in hospital births. Therefore, there may be increased risk of neonatal and intrapartum death associated with homebirths although further research using a comparable hospital cohort will have to confirm.”

See what I did there?  There’s no need to mislead. It’s ok to not have all the answers in the absence of clean data. Attempting to find correlation in real world data is a difficult and messy process. The messier and more unalike your cohorts, the more qualifiers you need when reporting your results. Doing anything less amounts to misrepresenting the results.

I also want to make it clear that it is ok to give your opinion after your results. This is a blog post and not a research study, after all. It’s ok to give your personal opinion that even if you can’t say for sure that there is a risk increase for having a homebirth based on the data, you believe there is a large risk increase. Just don’t try to pass off your personal beliefs about the data as if they were actual results.

The next section of Prof. Orosz’s post is far more egregious. She attempts to compare mortality rates for high-risk subsets in the homebirth cohort to…..well, she doesn’t really say what she’s comparing to. In some cases she does mention that she is comparing to the CDC hospital birth data again but in others she just throws out stats without mention of any source.

At one point she discusses breech births. In the homebirth cohort there were 5 deaths in 222 breech births. There were also a handful of breech births in which there was missing data. Prof. Orosz explains:

Disturbingly, the breech data were incomplete, as the authors explained: “There are 4 singleton pregnancies, 3 of which were breech presentations, for which all birth outcomes data are unavailable. These women began labor at home and then transferred to the hospital prior to birth. The midwives of record were contacted, and in each case the midwife did not accompany the mother, nor did the mother return to the midwife for postpartum care.” This missing data means that the breech death rate could in fact be as high as 36.0 per thousand, or 3.6%, which is similar to the breech birth death rate in the USA circa 1950.

In other words, Prof. Orosz is assuming that in all cases in which outcomes were missing for mothers who transferred to a hospital with a breech birth, the newborn died, though she gives no evidence for why this may be true. She then adds in these supposed deaths in order to come to her conclusion that 3.6% of the breech homebirths resulted in the death of the infant.

I don’t think I can possibly impress on you how not ok it is to just assume all missing data must mean dead babies. In the earlier section, the only thing I criticized Prof. Orosz for was not using enough qualifiers in her results statement to account for all the weaknesses in the data. In this case she’s going far beyond that to a point where it seems purposefully misleading.

Even if Prof. Orosz is trying to prove the point that homebirths are a bad idea in the case of breech births, there is no reason to manipulate the data. The 5 deaths in 222 breech births is already worryingly high. You don’t need to artificially inflate that number to get your point across.

Later, she does the same “assume all missing data is the worse possible outcome” trick when discussing the APGAR scores for the homebirth newborns. APGAR scores are commonly used immediately after birth to determine the health of the baby. The higher the score, the more signs of health the newborn is showing.

Only 245 infants (1.5%) had a recorded 5-minute APGAR of less than seven. Nationwide, 1.1% of babies above 2500 grams had a low APGAR as per CDC birth data, so the rate among the MANA STATS babies is significantly higher (p<0.0001 per binomial test) but not drastically higher.
However, APGARs were unavailable for 401 newborns. This missing data is concerning, as newborn health is a key outcome, and the only recorded variable with a greater number of missing data points was maternal education. If all 401 of those newborns in fact had low APGARs, the true rate of low-APGAR births would be 3.8%, which is three and a half times the national average.

No. Just no. Just because a baby’s APGAR score is missing in the dataset doesn’t mean it was a low score. Maybe some of them were, but we don’t know because they are missing. That’s kind of what “missing” means. Unless you have a really good reason to assume that all missing data = worst case scenario, you cannot just make that assumption.

In the piece I wrote last week I did not provide my personal opinion but there seems to be a lot of assumptions about it, pretty much all of them wrong. I’m going to give my opinion here if only so everyone will stop making assumptions. My personal opinion is that from what I’ve seen of the data reported on in the Journal of Midwifery and Women’s Health along with the CDC data as reported by Prof. Orosz and reviews of other studies I have read in various places (such as Emily Oster’s book Expecting Better, which I highly recommend) it seems as though there is either no increased risk or a very small increased risk to having a homebirth in a low-risk birth situation. High-risk births seem to have a possibly much higher risk of neonatal death in homebirths relative to hospital births. I think that almost no one should be having a high-risk birth at home except in exceptional cases. As for low-risk births, the situation is muddier but I would recommend not going with a homebirth. This is because I feel that in the case of uncertainty over the true risk of neonatal death, we should err on the side of assuming higher risk. I also don’t believe that the potential benefits of homebirths, such as less interventions or increased comfort, are in most cases worth any possibility of increased risk of neonatal death. I also understand that some women may be in a situation in which they value the benefits of homebirths more than I do or are not as worried about a possibility of increased risk, especially if they are minimizing those risks by using a qualified science-based midwife. I also think that it is problematic that some women who choose to have a homebirth do so for reasons that are not necessarily science-based or they don’t fully understand that there could be an increase in risk.

Again, this is my personal opinion and based partially in data, partially on guessing and assumptions regarding data I don’t have or doesn’t exist, and partially based on my personal valuation. Some of you may feel that benefits like comfort and lower rates of interventions are worth more than I am valuing them at. Some of you may feel that the chance that there could be even one death means that no one should ever be having a homebirth. Some of you may think that I’m not interpreting the data correctly or was too harsh on Prof. Orosz’s analysis and i09 for promoting it. Go in the comments and tell me why I’m wrong! Like everything in life, homebirths have some risks and some benefits. Let’s have an honest discussion about what these may be and how these should be weighted.

However, there is no reason to mislead on the real risks. If you are against homebirths and you think that any possibility of a risk is not worth taking, then argue that without resorting to inflating the numbers. It is my opinion as a skeptic that we should never, ever, ever misrepresent information in order to encourage others to take our side. We should never be personally attacking someone who has a differing opinion (and conversely not accusing others of personally attacking us when they merely disagreed with our views). We should never be threatening researchers or their families for publishing data we disagree with.

And i09, next time you come across a blog post that claims that they were able to do a complex cohort comparison study in just a couple days that completely overturns an established research paper done at a University and published in a respected peer reviewed journal, at least take the time to read the study first.

With that said, let’s have an honest discussion about homebirths.

Thank you Grounded Parents for letting me write one last guest post. <3

Jamie Bernstein

Jamie Bernstein is a data, stats, policy and economics nerd who sometimes pretends she is a photographer. She is @uajamie on Twitter and Instagram. If you like my work here at Skepchick & Mad Art Lab, consider sending me a little sumthin' in my TipJar: @uajamie

Related Articles

152 Comments

  1. “I want to know that all the definitions between the two datasets are the same. I want to know how problematic it may be that the hospital cohort defines “hospital births” as any birth taking place in a hospital while the homebirth cohort defines “homebirths” as a birth that was planned to be at home regardless of where it actually took place.”

    Exactly.

    “However, there is no reason to mislead on the real risks.”

    Exactly(squared).

    1. How problematic is it that the Cornell study defines hospital births as those taking place in a hospital, while MANA defines home births as those planned to be at home, regardless of where the baby ended up being born?

      Uh… NOT problematic at all, for two reasons:

      (1) By including babies whose moms tried to have them at home but had to be transferred to the hospital in the “hospital births” category, the Cornell study *takes those births OUT OF the home births category*, making home births look slightly safer than they are (because many of the home births that developed complications were not counted as home births). So if anything, this study’s conclusion that home births with midwives are 4.25 times more likely to kill the baby than hospital births with midwives (i.e., low-risk hospital births), tells us that home births are even MORE dangerous than this study says.

      (2) The MANA study includes a chart specifically stating how many planned home birth babies ended up being born at the hospital (see table 4 on p.22 of the study: http://onlinelibrary.wiley.com/doi/10.1111/jmwh.12172/pdf). It also shows how many were born through caesarean (table 3), which obviously means they were born in the hospital. And again, since 100% of the women in the MANA study started labor at home with the intention of giving birth there, it shows us the safety levels of PLANNED home birth. Some planned home births will end up in the hospital–just under 11%, according to the MANA study–and the death rate of babies whose mothers PLANNED and TRIED to have them at home is reflected in the MANA stats. This is the only useful information we can have; it’s not actually useful, from the perspective of a woman trying to decide where to birth, to only know the death rate of babies who actually WERE born at home, because you cannot know in advance whether YOUR planned home birth will have complications that require a hospital transfer. In other words you can plan a home birth all you want, but you can’t know in advance which group you’ll fall into–the 89%+ who are able to deliver at home, or the 10.9% who had to be transferred to the hospital. Thus, when planning where to give birth, you need to know the risks of all PLANNED home births–not just all planned home births that actually result in a home birth.

  2. What a pity. This is just depressing that Jamie Bernstein has no journalistic integrity. I’m not even through reading and the first two things I’ve noticed so far in another lazy journalism attempt by Jamie Bernstein:

    1. The authors from the study are not anonymous so you don’t need to “protect their privacy.” They are listed for all to see on the MANA study. You don’t want to link to them because everyone would be able to see they are NOT UNBIASED RESEARCHERS. 2 of the six are CPMs themselves and five of the six are on the MANA Division of Research Coordinating Council

    2. Dr. Orosz DID link to the World Health Organization. Did you bother reading her full analysis?? Or just the summarized portion from Skeptical OB? In her analysis, this is what she said : “The WHO (4) estimates the intrapartum stillbirth rate across North America as 0.3 per thousand. It is likely that the true number of intrapartum stillbirths among low-risk infants delivered in the hospital is considerably lower. However, in the absence of any other solid data, I will use the value 0.3 per thousand as a maximum reasonable estimate. The intrapartum death rate for MANA STATS sample was 1.30, which is significantly higher. (p<.0001, highly significant.)" So then you go to the final page of her analysis and there you have her references. #4 is this link: http://whqlibdoc.who.int/publications/2006/9241563206_eng.pdf

    Alright…. going to attempt to continue reading now.

    1. Thanks, I added an edit with clarification. I didn’t realize that the link which downloads an almost exact copy-paste of the blog post in word doc form had the small difference of adding sources at the end. I don’t understand why Dr. Tuteur would have left out the sources when she converted it into a blog post, but it’s certainly not Prof. Orosz’ fault. Thanks for pointing that out, but next time please do it in a more respectful manner.

      1. Jamie, your tone in your writing is not respectful. I am sorry for the personal attack. I am shocked at what I have read here though. In your first piece, your errors were pointed out. You made no changes (except the math error). Now you are saying Dr. Orosz manipulated data because you read what she wrote incorrectly. (which I posted here in another comment)

        Another note. For the hospital births. Any planned out-of-hospital births that transferred would, yes, be attributed to the hospital numbers. However, given the extremely large number of babies born in hospitals versus those born out of hospitals – and the even smaller number that transfer – do you think that could substantially change the hospital numbers reflected in the CDC?

    2. Absolutely. I was going to point out the same thing you did about the MANA study: three of the six authors are midwives (two CPM’s, one CNM). And of the three who aren’t, one (Dr. Bovbjerg) is a consultant to the Oregon Board of Direct Entry Midwifery (here’s her CV, it’s on page 2: http://health.oregonstate.edu/sites/default/files/people/vita-short/cv-bovbjerg-jan-2012-short.pdf), one (Everson) is on the faculty of the Midwives College of Utah (https://www.midwifery.edu/staff-faculty/), and the only one who apparently has no professional experience in midwifery has what to me look like questionable credentials for this type of research: she’s the lab manager in a primatology lab, focusing on the social behavior of macaques (http://www.vetmed.ucdavis.edu/abl/personnel/index.cfm).

      So, five midwives (all of whom are associated with MANA: http://www.mana.org/research/division-of-research-coordinating-council) and a primatologist are basically telling us, “Ignore the numbers–despite the numbers, home birth is actually really safe!” Here’s a contrary point of view: http://www.skepticalob.com/2014/01/homebirth-midwives-reveal-death-rate-450-higher-than-hospital-birth-announce-that-it-shows-homebirth-is-safe.html

      And here’s a link to the MANA study, which, as you noted, displays the authors’ names prominently at the top–you’re absolutely right, what privacy is there to protect? http://onlinelibrary.wiley.com/doi/10.1111/jmwh.12172/pdf

  3. Again, I’m pretty sure Jamie, the reliable writer that she is, didn’t do all the research needed. Instead of reading the full analysis, it looks like she just read the short, summarized version.

    “Disturbingly, the breech data were incomplete, as the authors explained: “There are 4 singleton pregnancies, 3 of which were breech presentations, for which all birth outcomes data are unavailable. These women began labor at home and then transferred to the hospital prior to birth. The midwives of record were contacted, and in each case the midwife did not accompany the mother, nor did the mother return to the midwife for postpartum care.” This missing data means that the breech death rate could in fact be as high as 36.0 per thousand, or 3.6%, which is similar to the breech birth death rate in the USA circa 1950.

    Had those breech infants been born in the hospital, there is at least an 86% chance that all of them would have survived, probably higher, and a 99% chance that no more than one would have died. Instead, at least five and possibly eight babies died.”

    Then you wrote “In other words, Prof. Orosz is assuming that in all cases in which outcomes were missing for mothers who transferred to a hospital with a breech birth, the newborn died, though she gives no evidence for why this may be true. ” NO, she did NOT make that assumption. She said “COULD in fact be as high as 36 per thousand” she did NOT say “IS in fact as high as 36 per thousand.”

    That is a TRUE STATEMENT that she said. It is not a manipulation. She then went on to say, as I copied above, “at least five and possibly eight babies died.”

    She did not assume they died. She said they possibly died.

    She did not manipulate anything, Jamie. She pointed out that data is missing and that the missing data could in fact make the numbers for the MANA study look even worse. This is TRUE. It is not a manipulation.

    1. Actually it looks like you didn’t read the “full analysis” because it is merely copy-paste of the blog post except with sources at the end! (which I did miss the first time around, because I didn’t realize I’d have to go on a scavenger hunt to find the post sources)

  4. First personal attack (“Jamie Bernstein has no journalistic integrity”) in the second comment!

    We are off to a fabulous start!

  5. Not midwives? Three of the 6 are midwives. Vedam is a CNM (RM is the Canadian licensing designation). Cheyney and Gordon are homebirth midwives (CPM). One is a doula. 5 of the 6 are associated with the Midwives Alliance of North America (MANA), the organization that represents homebirth midwives (CPMs).

    One of the key problems with both your analyses is that you don’t seem to understand that there are two types of midwives in the US, certified nurse midwives (CNM) who are real midwives with a master’s degree in midwifery, and CPMs (aka LMs, DEMs, lay midwives) who are really just lay people who made up a credential and awarded it to themselves.To give you some idea of their lack of basic education: in 2012, the requirements for the CPM were “strengthened” to mandate a high school diploma.

    CPMs do not meet the basic educational and training requirements for midwives in the UK, the Netherlands, Canada, Australia and anywhere else in the developed world. Indeed, they could not be licensed in any of these countries, which require a university degree in midwifery.

    The MANA study involves these self-proclaimed “midwives,” not real midwives (CNMs). Other research has show that CNMs have better outcomes at homebirth than CPMs, not surprising since they have much more education and much more training.

    1. I do know about the different types of midwife certifications and also that the lack of regulation in the U.S. means that in many cases some people working as midwives have little to no actual training or education. I didn’t specifically mention it in this post because that’s not what this post was about, but it is a serious issue that more people should be aware of.

      The MANA data did give stats for how many midwives in the study were CPMs vs CNMs (not to mention a couple people who were acting as midwives without any certification at all). I was a little disappointed that the study did not give outcomes for births by the different midwife certifications. As you mentioned, I would not be surprised to find that that women have better outcomes with CNMs. It’s an important point and I really hope they release data in the future that looks into this.

  6. Grounded Parents, I am just appalled that you are allowing this piece and the other Jamie Bernstein piece to stand. Her attempts to write about these studies/perspectives is so sloppy. She has demonstrated twice now that she does not fully read what she is critiquing. She likes shortcuts, assumptions, likes twisting things around, trying to say that Dr. Orosz manipulated the data just because Jamie didn’t understand what she was reading. This reflects on you. I give up on you, Grounded Parents. 🙁

    Jamie, I cannot wait until Dr. Orosz reads this.

    1. You’re appalled because you have an unhealthy fetish with homebirth. Here, finally, an unbiased TRUE skeptic states the unbiased truth and you simply cannot handle it because it’s not a “rah rah homebirth kills!” post. Every single bit of data from all over the world supports the theory that the mortality rates are the same or only slightly different for low risk women who give birth at home and hospital.

      1. No, I’m appalled because of the reasons I listed below in another comment. This piece is not skepticism. It’s sloppy and lazy, as I and several others have pointed out the issues with what was written here. And I don’t have a “fetish” with home birth. I’m an advocate for safer out-of-hospital births. There are many studies that show home birth can be as safe as hospital birth. Those studies have been done in other countries where the midwives are integrated into the health care and hospital system, those midwives have hospital privileges, those midwives have higher training and education requirements than CPMs and LMs and LDMs, etc etc. Home birth in the USA is not like that. Not a single study shows home birth is as safe as hospital birth in the USA, not even the MANA study.

  7. I am a little resentful that there are so many people would kill for the chance to write on a blog that has so much traffic, and would do a thorough job of it, but when an author here makes some pretty egregious mistakes and fails to post a proper retraction they get yet *another* article to dig themselves in deeper.

    1. I know the feeling. I’d love to see something that pulled together into the recent preponderance of studies* all showing the same thing–that by every metric of neonatal outcome home birth in the US is far riskier than hospital.

      It would also be helpful to describe some of the utter pseudoscience, quackery, and outright fraud that makes up the bulk of the average home birth advocate’s “evidence”, and the crowing about “informed” choices while simultaneously denying or suppressing the information.

      What a waste.

      * I’d link them, but when I tried doing so on the other post I lost a detailed and annotated comment that took a long time to research and write and never left moderation limbo (and then disappeared).

        1. Oh, thank you! I know comments with links sometimes get more scrutiny, but I’d seen others with multiple links so I was starting to feel a bit gun-shy, like was I too new to have links yet or something.

          1. It could have been the combo of being new and having a lot of links. It could have also been a word that tends to get flagged along with the links. We have people who aren’t new not get through because of the word thing. The algorithm or whatever it is behind the spam filter is still a bit of a mystery to me.

    2. There will be no retraction because there is nothing to retract. She fixed the errors in the post.

      I looked through the other accusations of mistakes, and they don’t hold up. If I missed any, let me know, but these are the issues, correct? (My numbers don’t match up with the original list in those comments because that list overlaps issues.)

      1. Jamie couldn’t find the quote at the link Dr. Tuteur provided. This link: Citizens for Midwifery: http://cfmidwifery.org/Resources/Item.aspx?ID=169

      The quote is not there. Rather than continue to click the links on that page to the press release, she went to the original source, the study itself.

      So if the egregious error she supposedly made was in going to the original source instead of the press release, then skeptics are making that particular error constantly and even advising others to do it.

      2. Jamie is supposedly accusing Dr. Tuteur of fabricating the quote, but she says specifically in her post that Dr. Tuteur probably got it from a press release. Which is where she did in fact get it. You can read into her post that she thinks Dr. Tuteur fabricated the quote, but that does require ignoring what Jamie actually said directly. The error is in your misreading. You can fix that. Jamie can’t.

      3. Jamie didn’t know where the numbers came from in Tuteur’s comparison of low- and high-risk homebirth to low-risk hospital birth because she was assuming good faith on Tuteur’s part that she wasn’t actually adding both risk categories for the homebirth sample and making a misleading comparison. So yes, Jamie did not know how Dr. Tuteur was getting the conclusions she was from a low-risk to low-risk comparison. She was also wrong in assuming good faith and that Tuteur wasn’t misleading with her numbers when she actually was. Should she fix that error in her post and point out just how misleading Tuteur was being? Is that what you’re asking? Has Tuteur retracted that post? Are you asking her to?

      4. Jamie used the parameters Dr. Tuteur mentioned in her post for her own CDC Wonder database comparison. If Dr. Tuteur included more parameters than what she wrote in that post that would have affected the results, then the error is in Tuteur’s post. She shouldn’t have omitted information that affects results. If, as she says, omitting this information doesn’t matter because the risk is still higher, then we are at the crux of the disagreement here. Jamie is writing about being accurate with numbers and honest about what they represent.

      Jamie’s post was about how two different people could reach such different conclusions about a study, and she looked at the study to find that both were being misleading in their interpretations.

      She didn’t even get into how Dr. Tuteur’s use of the percentage with small numbers was itself misleading (aside from the percentage being based on a misleading use of those small numbers in the first place). She didn’t point out that her chart looked like “how charts can be misleading” example from a statistics 101 course.

      So no, I do not see a need for a retraction. I see a need for being more careful about how we present the case against homebirth to people who are likely to already mistrust medicine by not giving them more reasons for that mistrust. I also see a need for several posts on how not to be misled by or mislead unintentionally with statistics.

      1. Melanie, I totally agree that I should have added some more commentary at the end of my first post to make it extra clear that things aren’t all positive in the case for homebirths. In fact, my personal opinion leans against homebirths for many reasons. I tried to fix that by making those things clearer in this post.

  8. Jamie, for what it’s worth, I think that one could get a good feel for relative home birth safety not just by looking at it in comparison to US hospital birth. This study found a perinatal mortality rate of less than 0.4/1000 for planned home birth. MANA’s low risk numbers were four times as high. There can be safer home birth, but it’s availability in the US is very hit or miss.

    http://www.ncbi.nlm.nih.gov/pubmed/19720688

  9. “I want to see pages convincing me the hospital cohort is a good comparable control cohort and what weaknesses there may be in using that cohort before I see any p-values.”

    My hospital cohort ISN’T a good comparison. I did not attempt to construct a matched cohort, because, as you noted, that would be very complex and subtle and open to interpretation and misinterpretation. Instead, I selected a cohort that could not possibly be LOWER in risk.

    My hospital cohort consists of all babies born in hospitals during 2007-2010 who weighed over 2500 grams and received at least a little prenatal care. This cohort (most of the USA) is much higher risk on average than the entire MANA database pool, as I said, “The MANA STATS mothers had fewer preexisting health problems, such as chronic hypertension. 66.9% of MANA STATS mothers had a normal (18.5-25) prepregnancy BMI, compared to 47.3% across the 36 states and DC for which data is available. (2) The racial makeup of the MANA STATS mothers was quite different from that of the USA as a whole, 92% White and only 3.1% Black or Native American women, who are at higher risk. Very few MANA STATS mothers carried twins, and none carried higher-order multiples. ”

    (Yes, I just quoted myself.)

    I did not attempt to calculate the precise odds ratio for neonatal death at homebirth. Instead, I used worst-case assumptions to show that the risk at home birth is substantially higher than at hospital birth, even for “low risk” women.

    1. Dr. Orosz, you made this point very clear (that you were taking pains to make your comparison cohort as favourable to MANA as possible). I cannot help but believe that Jamie is deliberately ignoring this.

      1. YES. Completely agree. Dr. Orosz was very generous in MANAs favor. A hospital cohort more similar to that of the MANA study (white women, full term, 2500 grams or more, etc) would show even greater risk for babies born at home.

    2. This is probably the most damning thing about Jamie’s post: she notices the problems with comparing the data without noticing (which would be bad enough) or stating (which would be worse) that those differences all work IN FAVOUR of MANA.
      The births who were transfered to hospital did not make the hospial stats look better than they atually are.
      THe only reasonable conclusion and caveat is that the numbers are probably higher.
      I’m also wondering how long Jamie would wait until she would deem it justified to make a strong recommendation against a CPM led homebirth. HOw perfect would the data have to be?

      1. She did make such a recommendation:

        “As for low-risk births, the situation is muddier but I would recommend not going with a homebirth. This is because I feel that in the case of uncertainty over the true risk of neonatal death, we should err on the side of assuming higher risk.”

        1. There’s a vast difference between erring on the side of assuming higher risk out of caution and acknowledging that the comparison is so stacked in MANA’s favor that the only real uncertainty is how much higher the risk is. Maybe it’s just semantics, but it puts a different, more accurate perspective on the recommendation.

        2. That’s not what I would call a strong recommendation, but we can disagree on that. The problem is that the whole rest of the post, including the “assuming higher risk” you quoted reads like we have no data at all on which to base conclusions. We have data. It’s not the best possible data, but it is data. And it’s not the only data, it’s part of a damning pattern.

          1. I agree that we have data, which is why it makes me particularly confused that people are acting like the whole argument hinges on this one analysis of this one study. (“Let’s go get better data!” “We already HAVE better data.” “Great….what’s the problem then?”) Why slug it out so vehemently over a study that can’t even convince skeptics who agree with the conclusion?

            I don’t agree that Jamie ever said we have no data. I believe the OP has stayed pretty true to criticizing one part of one study. And if it were framed as “not the best study but part of a damning pattern” that would be fine. And if the language used were the kind of language Jamie pointed out can be used for weaker studies, that would be fine. Triumphantly holding up the data as being irrefutable proof–that’s the issue.

            I personally am not a statistician, so I have to read analysis of these arguments here and elsewhere and try to see if a consensus emerges. I read a couple of other articles on this. I even spent a couple of hours reading a hundreds-of-comments-long Reddit thread because I was worried Jamie might have been wrong–but they brought up all the same points she did.

          2. The biggest problem with this article is not that it’s critical of an analysis, it’s not really that there have been two articles now about a shitty paper when all the good data sits on the sidelines unaddressed (though it is also kinda that), it’s mostly that the criticism is so, I dunno, cartoonishly bad.

            Statement 1: The control group has been purposefully selected to give the largest room for error that is reasonable.
            Criticism 1: You don’t have a good matched control group!

            Statement 2: The missing data is concerning. Here is what the numbers could be if they were all negative.
            Criticism 2: You assumed missing data was all negative!

            Statement 3: Here are the negative outcomes you would expect given the null hypothesis that there is no difference in death rates and between the MANA sample this ridiculously MANA-favorably stacked control group and here is the much larger number of observed negative outcomes.
            Criticism 3: You don’t have a good matched control group!

            It’s either an incredibly disingenuous or an incredibly ignorant critique.

          3. I think your confusion stems from the fact that we’re talking about different things. I’m talking about the gist of the post.
            Jamie talks a lot about the limitations of the data, but does not acknowledge that it is very reasonable to assume that the data is biased in favour of Mana. To point that out is not fearmongering.
            She talks about the limits of the data and then brushes away the massive conflict of interest the original authors have because she only cares about data as if these people took preexistin g data instead of data from a survey they themselves conducted
            And she outright twists other people’s words to make them fit her argument, as douladani has demonstrated (taking conditional phrases and acting as if they were statements of fact? That’s a classical strawman if there ever was one)

          4. This REALLY isn’t the hill I want to die on but you’ve always taken the time to explain things to me, so I figured I’d tell you what the gist of what the reaction feels like to me. (I don’t know if that’s because you care or because you like to have the last word, but I appreciate it regardless.) I read Jamie’s post very honestly–I read her stated thesis about how two groups can use the same statistics to come to different conclusions and I thought THAT was what the article was about. I truly believe that her characterization of Dr Amy’s analysis as “sloppy” ignited primal rage in people and the cognitive dissonance and talking past each other began IMMEDIATELY.

            Honestly what I see is Jamie making a pretty good case for why she thought that the data were good for conditional statements (or inclusion in a pattern) but NOT absolute statements of medical fact, a half a dozen commenters who have equated her trouble with this study as some sort of failure to stay on message (despite her explicitly stating her feelings otherwise several times), and several others who feel the need to bring in talking points that Jamie never mentioned in her OP. (And there are a couple of people basically saying that shame on us for looking at data of someone who’s doing good work; like the ends justify the means and we shouldn’t be skeptical of each other.) It’s all been a horrible wake up to me about ideological battles within skepticism given that basically every single person on both posts has the exact same opinions of current transparency, CPMs, the movement to glorify home birth, and modern medicine. I’ve lost my stomach to keep commenting. This isn’t the hill I want to die on. But I hope the people bringing the most venom and dog-piling know that I was driven off, not convinced.

            I’ve honestly moved over to a Reddit thread that has more than six or seven people doing all the talking and seems to be focused much more on the study and its math rather than home birth itself (which they pretty much calmly all agreed is dangerous and then got back to the study). They’ve brought up many of the same points Jamie did. Take that as you will…or not. 🙂

    3. You are forgetting something very, very important though Dr. Orosz. The homebirth data is not out of 4 million women like the CDC data, it;’s out of a measly 17, 000 women. So 3% of 17,000 women being black women, that matters. And “very few” mothers in the MANA stats carried twins? What is “very few”. “Very few” could mean nothing for women who birth in the hospital, but could mean a whole lot for those who birth at home. Yet again, something that makes the numbers for low risk women look worse. Are you telling us that you through women with multiples in with the low risk women who gave birth at home? Because if you did, that’s not right. Women who carry twins are not low risk.

      1. “So 3% of 17,000 women being black women, that matters.”

        Um, no, it doesn’t. That’s how percents work. A higher percentage of US mothers overall were minorities, or carried twins, or had high blood pressure, than the home birth pool.

        Since both death rates were expressed per thousand, the PERCENT of high-risk women in each pool is the only thing that matters.

        Please do read the entire MANA study. Please do note that the death rate for the low-risk homebirth pool (which was only about 2/3 of the sample) is still twice the death rate for all full-term babies in the USA, low and high risk all mixed together.

    4. Hi Prof Orosz,

      I understand that you specifically chose a hospital cohort that would be more likely to be higher risk (I assume you actually meant to say “higher” in your comment). But that doesn’t change that you are doing merely a rough comparison with a lot of guessing and handwaving, which is required in this case because better data doesn’t exist. That is all fine, but when it comes to stating your result, you did not acknowledge at all that this is merely a rough comparison. You can give your best guess that the relative risk is real and probably even bigger than what you’re showing but you can’t say it with surety. The homebirth cohort is a very small, extremely self-selected group. They may have other risk factors that are not obvious, such as fear of the medical establishment that may cause them to be more likely to turn down medical interventions, medications or screening procedures. Or maybe not. I don’t know because this study provided no evidence one way or the other.

      1. Actually, I said the the hospital group could not be lower in risk than the home birth group.

        I structured it like a math proof. I’m a mathematician, not a doctor. In mathematics, sometimes you can’t find all the numbers, but a “worst-case” assumption still suffices to complete the proof. As I said, I never tried to estimate the true increase in risk at home birth, just to prove mathematically that it existed and was significant.

        That’s an interesting idea, that home birth women may be at higher risk because they turn down medical interventions and screenings. I don’t believe it would compete with the other risk factors involved, but it exists. (Of course, it would STILL represent a flaw in the US system, because countries that do home birth better don’t LET women try it without appropriate screening.) Just checked CDC Wonder, and the neonatal death rate for full-term normal-weight babies with zero prenatal care in hospital was 0.96 after excluding congenital abnormalities, only slightly worse than MANA’s 0.77. (About 20% of those deaths involved SIDS or suffocation.)

        This zero prenatal care group is obviously rife with other risk factors. Most are young and unwed, and presumably most received no care out of poverty. It’s essentially the most challenging group walking into the hospital, and the hospital does almost as well with them as home birth does with women who are far healthier and more advantaged.

  10. I have to add to the chorus of people deeply disappointed that skepchick/grounded parents has lower standards that I had expected. I was surprised by the first piece by this author because it went after Amy Tutuer who is the loudest, strongest voice in the skeptical community speaking out about the very real dangers of homebirths with CPMs. It is truly depressing to see now a follow up piece, clearly motivated by personal vendetta, and which is full of nitpicks while it and its preceding post aren’t exactly impeccable pieces of analysis.

    There is a whole movement with countless spokespeople writing books, blogs, articles, and running whole institutions dedicated to a whole hell of a lot of woo that aims to convince women that hospitals are scary, unsafe places, and since birth is natural it is safer done at home. Women and babies die unnecessarily because of this movement. The skeptical movement has a responsibity to counter all the misinformation out there, not attack the few skeptics who are doing that job.

    1. You and I read this piece very differently.

      To me it seems as though Jamie is arguing for trying to be as exact as possible with the numbers being used to toss out statistics and how those numbers are reached. She also is arguing vehemently that we should not be using language that suggests certainty when we do not have enough data (or the right sets of data to compare) to be certain at all. She is arguing for complete honesty in the face of the group that, as you say, is twisting (or ignoring) the truth to scare women away from hospitals.

      How are we any different from them if we care more about message than fact? I don’t think that is a nitpick.

      Now, if her numbers are wrong, that’s a discussion really worth having – that’s what we should be talking about… instead I see slams on her integrity (though to be fair, that was apologized for) and a characterization of her piece as a personal vendetta for daring to question how Dr. Tuteur and Prof. Orosz came to their final numbers.

      One critical thing missing from this piece (in my opinion) that Dr. Tuteur filled in very well is the difference between CPMs and CNMs. I wish that every conversation about homebirth would draw this distinction so we didn’t seem to be demonizing all midwives across the board. I think we are all in agreement that CPMs, with no recognized medical training, should NOT be allowed to preside over births as the primary caregiver. If the decision of whether or not homebirth is a viable option for an individual mother were left to her and her doctor or CNM it should all but wipe out high risk homebirths, while leaving it as a possible option for some low risk women who are interested in it.

      1. Silver feather, Jamie’s scrutinization would maybe be justified if her own analysis and evaluation were impeccable. But she herself has gotten so much wrong (as pointed out by commenters in both articles) and she keeps missing the big picture which is that this comparison of numbers is skewed in favor of homebirth however you look at it.

        1. Honestly, I see errors being pointed out on both sides of the debate and, being a layperson, I’m really not sure who to believe. Until I have enough time and motivation to sit down and try to sort it out for myself I’m going to have to refrain from pointing fingers about who got what wrong.

      2. I really wanted to talk about CPMs vs CNMs because there was some basic numbers on how many homebirths in the study were presided over by what types of midwives, but both of my posts were so long already and I couldn’t find a good place to fit it in. It’s a huge issue though. In face, a small number of births in the study were actually presided over by “midwives” who had no credentials at all or had “credentials” in things like “naturopathy.” There is a serious issue in the U.S. with lack of regulation around midwifery that likely increases risk.

        I especially wonder about who the women having high-risk homebirths are and who their midwives are. From what I’ve read, most midwives will not do high-risk homebirths, so anyone wanting one would have to shop around to find a midwife willing to do their birth. I’m not sure I trust these midwives that are willing to do births they know are high risks.

        1. The midwives doing high risk homebirths are apparently the ones who will run away if it starts go badly and then not even bother to find out what happened, as suggested by the 3 “missing” outcomes from the breech group.

          Honestly, I find that even worse, in a way, than the possibility that all three died. That the midwives who were supposed to be caring for them didn’t bother to find out whether they lived or died. Horrifying.

        2. I didn’t have to “shop around.” When I was pregnant for a second time and seeking care (with a CNM in a hospital), I called the Birth Center that had done my prenatal care with my first baby and asked about midwives who handle vbacs (which are by definition “high risk”.) I didn’t say anything about wanting a homebirth, and the woman on the phone offered me their list of midwives willing to do homebirth vbacs. On top of that, she followed it up telling me that even though most insurance won’t cover homebirth, there are ways of finding funding! I wasn’t even shopping around to find a midwife willing to do a high risk homebirth, and here I was being offered a list of them just because I called a Birth Center for a referral.

          I don’t wonder about this issue. After the bullshit I witnessed at what I thought was a credible Birth Center, I started looking into this issue deeper, and was pretty mortified by what I found. Except in the states that explicitly prohibit midwives from doing high risk homebirths, it actually isn’t any more difficult than a Google search and maybe a phone call or two to find a midwife willing to do something like a homebirth vbac. And as soon as a woman who might initially just be curious finds one, they will be fed all kinds of bullshit about how uterine ruptures are no more likely that the risks of having another c-section, how hospitals only ban them because they worry about liability, and so on. I do not hesitate to say that any midwife willing to do a high-risk homebirth is not trustworthy.

          1. It is insane that some people think it is a legitimate criticism to say that “hospitals only ban X or only require Y because they worry about liability.” Have these people forgotten what liability means? It means the hospital’s insurance had to pay a bunch of money… BECAUSE A MOTHER AND/OR BABY WAS INJURED OR KILLED. There is no “liability” for anything unless someone gets hurt or killed!

            So let me translate that: “hospitals only ban X or only require Y because they worry that doing X or not doing Y will INJURE OR KILL YOU OR YOUR BABY.” Um… see, call me crazy, but I instinctively tend to prefer medical professionals whose decisions are based on their desire to not kill me or my baby!

    2. Even when those skeptics are using misinformation to do this job? Even when doing so means that a lot of people we need to reach who already have a mistrust of medicine are even less likely to trust us because one of our loudest spokespeople is blatantly misleading with data?

      Criticizing someone’s misuse of data, even when that misuse is in the name of a good cause, is not going after that person. It’s going after the misuse of data, because that does more harm to the cause than good.

      1. It is imperative that you understand that no one has misused the data. Jamie Bernstein has misunderstood the data. And the misunderstandings aren’t subtle, either. She doesn’t realize who the authors of the MANA study are. She doesn’t bother to check quotes before announcing that she can’t find them. She has no understanding of the subject; she appears to be unaware that we are not talking about real midwives like CNMs and the midwives in Holland, Canada, etc. We are talking about a group of women who made up their own midwifery credential and awarded it to themselves. We are talking about an organization (MANA) that has literally NO safety standards: every midwife is supposed to decide for herself what is safe and what is not.

        I suggest that you contact a person who has done a tremendous amount of work in this area, Dr. Amos Grunebaum of the Cornell Weill Medical Center. I’m believe that he would be willing to speak with anyone at Grounded Parents to explain the issues.

        Please, do the due diligence that is expected of journalists. Talk to multiple experts in the field. Don’t depend on Bernstein’s analysis, which is ridiculously wrong. No one has misrepresented or inflated the dangers of homebirth with a homebirth midwife. Bernstein has misunderstood what she has read.

        1. You didn’t address the data at all in this comment. That’s my concern. I don’t care who the authors of the MANA study are. Doesn’t change the numbers. I don’t care that she went to the study instead of clicking on the press release. Doesn’t change the numbers. Who the midwives are and where they are does not change the numbers. The safety standards, how midwives practice, the difference between CPM and CNM are absolutely important to me in advocating against homebirth, but they do not change the numbers we have to work with in this particular analysis of the MANA results.

          I’ve stated in another comment that I think the MANA results are almost useless. I do not in fact trust the data in it for multiple reasons. BUT if we are going to make comparisons using that paper, then we need to be careful, transparent, and not mislead with how we present those numbers.

          Jamie’s articles are not about surveying the literature and drawing conclusions about the risks of homebirth. They are about specific analyses of data that are flawed and why. Doing that does not require talking with multiple experts in the field, although that of course makes sense for a post about homebirth risks in general.

          1. If you are confident that Jamie has understood the data, how can it hurt to confirm that with independent experts? Consider approaching skeptic bloggers with actual medical knowledge as well as in depth statistical knowledge, like the bloggers at Science Based Medicine.

            Science is about subjecting your claims and conclusions to public scrutiny, I’m willing to do that, because I don’t fear what others with equivalent or greater knowledge will say. Why are you afraid of independent analysis? That doesn’t sound like skepticism; it sounds like denial.

          2. Why are you afraid of independent analysis?

            Why are you assuming I’m afraid of independent analysis? This is a pattern with you, filling in information you don’t have with the worst possible assumptions. That doesn’t sound like skepticism; it sounds like conspiracy theory.

          3. Great idea Dr. Tuteur. I contacted David Gorski to see what he has to say about this. Thanks for the recommendation.

    3. Martha, I generally agree with you and most of the comments you leave, and even when I disagree on a particular point (like whether home birth risk assessment is something we should respect as a body autonomy issue) we I generally agree on most of the rest (the importance of improving midwife training and combatting the “woo” numbers).

      But I have to tell you that it’s hard to read this particular comment without feeling like you’re essentially saying that the ends justify the means, and that it’s okay not to have a responsibility to the truth in a single study so long as we agree with someone’s overall cause.

      Maybe that’s not what you meant?

      1. That’s not what I meant. I appreciate you giving me some benefit of doubt. What I meant was that both of Jamie’s articles are sloppy. She has no credibility discrediting Dr. Tutuer when the very first objection she makes (that the authors were not midwives) is false. The many problems (some clear falsehoods, and some misleading implications) have been pointed out throughout the comments under both articles. The objections Jamie makes are over such minutiae, which is why I call them non-issues, and why this seems more about personal vendetta than spreading more accurate info for the public regarding homebirth.

        1. Well, see what you’ve said here seems like a shift in what I took umbrage with above. I’m all for criticizing the criticizer. That seems very important in skepticism. As important as looking at studies in the first place. I’m not in a place to call them “sloppy” but that certainly seems to be the word de jour.

          I’ve been following the comments here as best as my non-mathy brain can, so I don’t know what are “gotcha” details and what are actually important to Jamie’s statistical analysis. Jamie has replied several times in good faith and gotten a lot of speculation about her reading comprehension in return, so I can’t say that I think every accusation leveled agains her hasn’t talked past the points she’s made. (Her thread with Dr. Orosz is interesting–they seem to be interacting at least in good faith.)

          And if Dr Amy’s every claim is above reproach, great. Lord knows I won’t mind have more ammo if my partner starts thinking about a VBAC home birth. But I don’t think the fact that we perceive a side as correct should EVER mean that we don’t check each other’s math.

          1. “Sloppy” isn’t just the word de jour. I used it as shorthand to refer to all the damn mistakes that have been mentioned throughout the comments. Do I have to go through the comments and compile a list, or can I just say she was sloppy? I think I can just say she was sloppy.

            Dr Amy’s claims aren’t above reproach. Nobody said that. Nobody even implied that.

            The issue here isn’t the math. Jamie’s criticisms aren’t of the math. Jamie didn’t point out any errors in mathematical calculations. Here criticisms is that she thinks that Tutuer’s and Orosz’s representation of the data is misleading and that the comparison they make is inappropriate. Unlike an argument over what 2+2 equals, these are actually debatable points. Which is why I used the terms “nitpicking” and “minutiae”. Then you throw in that Dr. Tutuer has now posted several articles on her very popular blog viciously criticizing Jamie’s articles and Grounded Parents/Skepchick in general, and that’s why I have concluded that the main motivation for Jamie’s second article is most likely personal vendetta, opposed to cold, unbiased skeptical inquiry. I could be wrong of course. I can’t read peoples’ minds. I can only say what this looks like to me.

          2. “Lord knows I won’t mind have more ammo if my partner starts thinking about a VBAC home birth.”

            Holy shit, I didn’t even make that connection. If you come back to this thread for any reason let it be for this–by MANA’s own numbers, small though they are, the perinatal death rate for VBAC was 5 in 1052, or nearly 1 in 200. This is similar to the rate of uterine rupture in a woman with a previous C-section undergoing a trial of labor for a subsequent birth. Uterine rupture at home, regardless of provider, is a death sentence for the baby and is dangerous for the mother as well.

            Please do not try to wait for more data on this. If stronger data comes out between now and then, great, but VBAC is an absolute contraindication for home birth in every system where home birth can be considered “safe”. When we talk about how home births can be made acceptably safe in this country, VBAC is always considered high risk and not an acceptable condition for home.

        2. Which, in case I failed in my last comment, basically is my windbag way of saying I appreciate the difference that you’ve expressed, and even though I’m not sure I totally agree with you on Jamie’s articles, I am glad I misread your original comment.

        3. You can think anything you want. As can I. You have your reasons for your opinion, as do I. I came here to thank you for clarifying your position regarding your statement that skeptics should “not attack the few skeptics who are doing that job,” not to go down the rabbit hole of who is “right” one more time.

      2. It is much easier to accuse someone of “misrepresenting” info than to accuse someone of making an outright false claim. The former involved a lot of nuance and context. Often a case can be made both ways. The latter is a question of facts which are objective. Dr. Tutuer hasn’t gotten any facts wrong. So the attacks are on how she interprets and presents those facts. Given that she is the loudest and one of the few skeptics fighter against a movements that does spread a lot of false claims as well as misrepresentation of correct info, it disappoints me to see a well known skeptic blog attacking her and doing so in such a sloppy way.

        1. Martha, You think it’s okay for Dr. Tuteur to have compared low- and high-risk homebirths to low-risk hospital births, then presented her conclusions based on that comparison? How about her use of percentages with small numbers, a practice that is recommended against in epidemiology and public health and that has been written about as a problem in medical journals as well because it’s misleading.

          Or how about the blatantly misleading chart with two MANA comparison groups but only one hospital group, and the third bar actually including the second bar to inflate it?

          That’s not even getting into the conspiracy theory meme she sent to Jamie as her first response to the original article (the WHAT ARE THEY HIDING? meme).

          This basically illustrates the many misleading tactics she’s used: http://skepchick.org/2014/03/bad-chart-thursday-shark-attack-risk-increases-1200-in-hawaii/

          Can you really still say that she hasn’t misrepresented the data?

          1. Sigh. I don’t know why I keep trying to explain the basics to you folks, but I’ll give it one more try:

            Melanie’s piece on shark attacks begins with what it apparently meant to be an attack on me:

            “According to International Shark Attack File stats, between 2008 and 2013, shark attacks increased from 1 a year to 13 in Hawaii. I know, I know, that’s still a very small number, but that is a 1200% increase, which is MASSIVE!”

            Hawaii has about 8 million visitors per year and a population of 1.3 million. It is not unreasonable to assume, therefore, that there are at least 10,000,000 episodes of swimming each year (some people swim more than once).

            That means that the rate of shark attacks has gone from 0.1/million to 1.3/million. Although that is indeed an increase of 1200%, the original number is so low that the risk of getting attacked by a shark in Hawaii is still very, very small.

            Melanie, if you are reading this, I want you to concentrate very carefully on what I am about to say next and hopefully you will understand it:

            The numbers we are talking about when we talk about perinatal mortality are ONE THOUSAND TIMES HIGHER. You seem to think that a death rate of 0.4/1000 is a number that is so tiny that no one needs to consider it. There are 2 million term births in the US each year. A death rate of 0.4/1000 means 800 dead infants, hardly a trivial number.

            Your shark example is not as clever as you think. In fact it is not clever at all when you consider that perinatal deaths are 3 orders of magnitude larger.

            Now let’s move on to the issue of high risk and low risk births. Let’s do a thought experiment:

            Imagine that you are thinking of taking a helicopter tour of Hawaii. There are two different companies to choose from and each takes 1,000 helicopter tours per year. You are worried about the risk of a helicopter crash so you ask the folks at Joe’s Helicopter Tours about their safety record. They assure you that they take every possible precaution and have an excellent record with a crash once every 2.5 years (0.4/1000).

            Next you go across the street to Steve’s Helicopter Tours and ask the same question. They tell you that they have 2 crashes every year. In other words, Joe crashes 1 time every 2.5 years, and Steve crashes 5 times every 2.5 years. That probably makes a big difference to you in determining which company to choose, doesn’t it? That’s because you really, really don’t want to be in a helicopter crash and even though both companies can assure that you will survive the vast majority of times, the difference is not trivial, right?

            Now imagine that when you tell Steve that you are going to go with his competitor because of the safety issue, Steve says: “Hey, that’s not fair.” Joe’s safety statistics look better only because he won’t fly during typhoons, tornadoes and lightning storms. Joe thinks those conditions aren’t safe; we think they are very safe. Nonetheless, it is not fair for you to include our crashes that occurred during typhoons, tornadoes and lightning storms. When you subtract them, we only 4 crashes and that’s the comparison that you should make.”

            Do you think that in determining which company is safer, you should subtract the crashes that occur during typhoons, tornadoes and lightning storms? Probably not, since their willingness to take their helicopters up in unsafe conditions tells you that they are not safe pilots, right?

            The same thing goes for homebirths, Melanie. You don’t have to subtract the high risk conditions because the fact that they are willing to attend these homebirths is itself an indication that they are unsafe practitioners.

            Let’s summarize, Melanie:

            1. Your shark attack comparison is inane because the rate of perinatal deaths is one thousand times HIGHER than the rate of shark attack. The risk of death of a baby in childbirth is HIGH, not low.

            2. If a helicopter company has lots of crashes compared to its competitor because they fly in unsafe conditions while the competitor does not, you can’t subtract the crashes in bad weather and declare the first helicopter company “safe.” The mere fact that they are flying in unsafe weather is an important indicator that they are not safe pilots. Similarly, it is quite appropriate to consider high risk homebirths when assessing the safety of American homebirth midwives (CPMs). The mere fact that they are attempting to deliver these patients at home is an important indicator that they are not safe practitioners.

            Do you understand now?

          2. Thank you for finally addressing the issues at hand. This has been incredibly frustrating because your comments and blog posts have been derailing this to be about everything but the actual issue.

            First, it wasn’t an attack on *you*. None of this has been an attack on you. You are the only one engaging in personal attacks here. This has been criticism of how you present numbers. The reason for this criticism is that it is important not to be dishonest in how we present the case against homebirth if we are going to persuade people who are on the fence about it.

            I can’t send people to your blog for information because it is full of sensationalized numbers, like the post in question, and multiple personal attacks. It reads like a tabloid. I’m sure it’s great for “persuading” people who already agree with you, but for the rest of us trying to persuade people who don’t, many of whom already mistrust medicine, it gives them the false impression that the case against homebirth is dishonest. That’s the issue. You’re making it harder for me and others to educate people about homebirth risks by being misleading, sensationalist, and illogical (ad homs, hasty generalizations, jumping to conclusions based on missing information).

            Yes, the shark attack example is deliberately ridiculous to illustrate that the tactics you are using are misleading. It’s easier to see when you apply the same tactics to a ridiculous example like the shark attacks. They are the exact same tactics. They are misleading no matter who is using them, even if for a good purpose. All the numbers I present are true, but how I present them is misleading, in the same way you are doing it with the numbers from the MANA paper.

            Let’s take a look at them one by one.

            1. Using a percentage to describe relative risk based on small numbers. This is not an okay tactic to use. Here’s an example in epidemiology in public health that describes why and even uses infant mortality rates as an example: http://www.in.gov/isdh/23986.htm (search the page for “Rule of Twenty” to get what I’m talking about). If you need more examples, there have been articles published about misleading uses of relative risk in various medical journals. For those following along, this page (http://www.brighthub.com/science/medical/articles/20073.aspx) and the reference it cites (Gigerenzer G. et al., “Helping Doctors and Patients Make Sense of Health Statistics”) are also accessible explanations.

            Using a percentage at all in this case was misleading, even absent the fact that this percentage was based on a faulty comparison.

            You are saying this is not a case of actually small numbers, but that’s not true. This is about the MANA data, which did not show thousands of deaths. That’s the issue here. You’re not presenting the numbers from all the studies against homebirth. Your post was about the MANA data specifically. That is what you are misrepresenting. And you are doing so unnecessarily. There are plenty of arguments to make without misrepresenting the numbers. Hell, there are multiple problems with the study itself before we even get to the numbers. By saying this is about thousands of deaths in this comment thread, you are STILL misrepresenting the numbers we are actually talking about.

            2. Comparing low-risk and high-risk homebirth to low-risk hospital then presenting the result as the increased risk of homebirth. This is a faulty comparison. An obviously faulty comparison. I understand that you think you are justified in making this faulty comparison, but it doesn’t make it any less faulty or misleading.

            No one is saying that you shouldn’t also discuss the multiple additional factors that could affect these numbers. You absolutely should! What we are saying is that you don’t artificially inflate the numbers to do this. You compare low-risk to low-risk to get the increase, then discuss the many ways in which this risk is likely to be even higher because of these other factors you can’t quantify.

            Right now, someone could go to your post, see that you are making a faulty comparison and highlighting it in the title and the bulk of the post and have every reason to believe that the case against homebirth isn’t as strong as it actually is! If the case against homebirth were strong, why would this prominent voice against homebirth have to fudge the numbers like this?

            The thing is, the case against homebirth is strong enough not to do this. Your tactics are weakening it. Please stop. Seriously.

            So looking at your helicopter example, here’s what you do when comparing the numbers. You compare the numbers they actually gave you, then discuss the context of those numbers and reasonably conclude that the numbers don’t tell the whole story and that the company flying in unsafe conditions, Joe’s, is less safe despite the lower crash rate.

            You don’t just decide that because Joe’s is obviously less safe, you’re just going to inflate his numbers based on no quantifiable data for that inflation, compare the two, and call that a factual comparison of numbers. It’s not honest. And it’s not necessary.

            3. Misleading visual representation of data. Not only did your title and the bulk of your post focus on your misrepresentation of the data, you created a chart that further misrepresents the data by first excluding a comparable hospital group for the MANA all risk bar and by making that bar include the data from the second bar to artificially inflate it.

            I don’t think you understand that I’m not at all arguing that homebirth is safe. I’m not at all arguing about all the studies on homebirth. I’m arguing that it’s not okay to be misleading with how we present the numbers in this case or in any other. I’m arguing that doing so is not only unethical, but it does more harm than good for the cause of educating people about homebirth risks who already have a mistrust of medicine. You are giving them cause for that mistrust.

          3. “…the third bar actually including the second bar to inflate it”

            What exactly does this mean? Because I’ve seen you use the the same wording like three or four times now and I can’t parse it.

            The MANA all risk mortality rate is the *rate* of death–deaths per 1000 births–not the *number* of deaths in the sample. If you took out the low risk cohort and just counted the high risk births (which you can’t, unfortunately, because MANA didn’t break it down that way), the death rate for the “third bar” would be very much higher. I’d guess around five or six per thousand, instead of close to two.

          4. alli, you’re right. And I’m not wording myself well on this. The bars should be low risk and high risk for both groups. Four bars. Doing this with the actual study in question isn’t possible to do honestly. The chart doesn’t do anything but imply that the highest bar is comparable to the lowest, which is misleading. I don’t actually think that this study and the comparisons made even lend themselves to chart form at all, for these reasons and because the numbers are so small. It isn’t possible for a chart on this data to not be misleading.

          5. Can you explain to me why using relative risk with small numbers is misleading? Below, you cite the Rule of Twenty, but since there are more than 20 cases in the data, what is the problem? It is certainly misleading to give only the relative risk without the absolute numbers, but Dr. Amy cited the absolute numbers as well. Standard practice is to give both.

            Just to make it clear that relative risk matters in samples with small numerators, I link to THE primary work on the statistical analysis of rare events, by Gary King, a methodologist from Harvard. http://gking.harvard.edu/files/gking/files/0s.pdf He argues: “It is important to appreciate that what may seem like small values of the probabilities can have overwhelming importance in substantive analyses of genuine rare events data….Relative risks are typically considered important in rare event studies if they are at least 10– 20%” So, in other words, even with rare data, we should consider substantive change to result from far, far smaller levels of change than the 450% that Dr. Amy finds.

            He also notes, by the way, that the problem with rare events is that they UNDERESTIMATE the real probability of rare outcomes (in this case, neonatal death) and, that it can therefore underestimate the effect of the independent variable (in this case, homebirth.) So, once again, you are complaining about a characteristic of the sample that means, not that Dr. Amy, has overstated the risks, but that she has likely UNDERSTATED them.

          6. I’ll take a look at the paper you linked to but haven’t had a chance yet.

            I provided a link to the Rule of Twenty to illustrate that a small change in small numbers can produce a large variance in rate. The Rule of Twenty applies to calculating absolute risk rates. I have not at all taken issue with calculating absolute risk rates. Yes, 35 is higher than 20. Not by much, but it is. It is also 35 out of nearly 17,000, and the page I linked to is talking about numerators with denominators less than 1,000. The absolute risk rates we’re talking about have numerators far lower than 20. Relative risk percentages from these small numbers will vary a great deal with small changes in the numbers, a variance that is not going to match the actual variance in absolute risk. The principal is the same. This is not even getting into the fact that the 450% is based on a misleading and faulty comparison. She not only used a relative risk percentage, she inflated it.

            The companion to the Rule of Twenty link that I provided is to a page talking about comparisons and transparency in medical research statistics (http://www.brighthub.com/science/medical/articles/20073.aspx, a shorter version of the explanations in this paper, which I mentioned but meant for that mention to be a hyperlink: http://www.psychologicalscience.org/journals/pspi/pspi_8_2_article.pdf). The paper talks in part about how relative risk is misleading to physicians as well as to patients and journalists (based on studies demonstrating the lack of statistical understanding among these groups related to relative risk and other presentations) and recommending “transparent risk communication” that includes using “absolute risks instead of relative risks” because relative risks can mislead (emphasis mine).

            tldr; Using relative risk percentages with small numbers is misleading because a small change in the number can lead to a large change in the percentage. Highlighting this relative risk percentage when it is based on faulty comparisons is dishonest.

            Use the numbers for what they are, honestly, and then talk about the factors that make the risk likely to be even higher. That’s all I’m asking. I really don’t understand why there is so much resistance to this idea.

          7. Melanie–I think a low to low risk and all to all risk comparison could be done by using the hospital CNM and MD rates, respectively, for comparison. MD would include things like breech and VBAC that would not be in the CNM group and give a better estimate of relative risk. It would still be skewed in MANA’s favor by the fact that women with breech, VBAC, twins, GD, or pre-eclampsia are less represented in the home birth sample than hospital. And, as you said, there’s no way to accurately get a high risk only rate from the MANA data, and as far as I can tell no way to get a high risk (for those specific conditions) estimate from the public CDC data at all, more’s the pity.

            I’ve been trying to think of how to explain why I didn’t and still don’t find Dr. Tuteur’s comparison misleading, and I think I finally put my finger on it–it’s really a comparison of home versus hospital *midwifery* care. An infant is ~5.5 times more likely to die under home midwifery care than hospital midwifery care, not only because of the lack of training or resources, but because of how home midwifery is done–allowing conditions that would never be acceptable for hospital CNMs.

            The decision to go with a home birth midwife instead of a hospital midwife is more than five times as risky in large part because home birth midwives inappropriately keep patients who would be risked out to MD care by hospital midwives. I don’t think it’s unreasonable to make that kind of comparison, because home birth parents are likely relying on medical advice* from someone who may either not have awareness of risk or seriously downplay it.

            *This isn’t aimed at you necessarily and may be detracting from my point, though I think it’s relevant: it bothers me quite a bit that mothers are expected to become experts in childbirth and know when they have risk factors that would make midwifery care inappropriate, or what type of prenatal testing and care they need, when we don’t expect a similar level of expertise for any other medical thing. Most people really cannot make these assessments themselves, nor should they be asked to, and so need actual expert advise. Not getting that advise raises risk more than having a risk factor in the first place.

          8. Okay, I can’t reply below, but I am responding to your explanation for why the small number of events means that we can’t use relative risk. First of all, your cite says that using relative risk *can* be misleading. A true statement, but not one that means use of relative risk is *inherently* misleading. To make the case that it would be misleading, one has to explain what one was led to believe that is not true. You could argue that it makes it homebirth seem like a big risk when it is small one, but the latter classification would be your *opinion*. There is no one standard for what a “small risk” is and as numerous commenters have made plain, plenty of people think that a increase in absolute risk of 1/1000 is NOT a small risk. It is not unreasonable for Dr. Amy to conclude the impact is a substantively large one, and therefore it is not misleading for her to present it as such.

            Second, to the extent that the small number of events make the relative risk more unstable, a) this is fully accounted when we calculate statistical significance and b) while a single additional death may make the relative risk change a lot, the whole point is that deaths are rare, so extra deaths are not supposed to be just popping in and out of the data by chance. It is of course possible that we could accidentally select a year with an unusually high death rate by chance, but the probability of this is low. It is even less likely that we could pull such unusual data three years in a row; and less likely still that other authors working with other populations could all draw the same atypical sample to arrive at similarly erroneous estimate of relative risk. I know you are only analyzing this one study, but context matters when we are assessing the validity of measures: the consistency across studies means that Dr. Amy is justified in presenting the relative risk as a valid estimate of the true risk.

            Finally, you are basing your claim on one cite about how to present data. These people are not the only people to write about relative and absolute risk and it is *far* from settled in the literature that presenting relative risk is problematic. There are plenty of scholars in the camp that argues that relative risk is absolutely necessary for fully understanding impacts. Which is why, as I noted, the standard practice is to provide both relative and absolute risk, and I would expect to see both in a typical academic article, even one dealing with rare events.

            By the way, though you were going for ridiculousness, it was not ridiculous to present relative risk about shark attacks either. A rate of shark attacks that is 1200% of normal is indeed something that would (and does) catch the attention of both marine biologists and beach authorities.

          9. I’m very interested if Melanie Mallon has read the paper linked to be professor custard. From my reading of this disagreement, the issue is whether using relative risk is misleading for rare events such as neonatal death when comparing rates at homebirth versus hospital birth.

            The accusation from Jamie Bernstein and Melanie Mallon is that it IS misleading, however they haven’t yet looked at their assumption and provided the evidence that backs up this claim. Perhaps they are right and perhaps they are wrong, personally I’d like to see some evidence from them that it IS misleading.

          10. You are basing your claim on one cite about how to present data.

            No, I’m not basing my claim on one cite. I presented one paper that was an accessible answer to the question of why relative risk percentages are misleading. This does not mean there aren’t other papers. There is quite a bit of discussion about this, but if you know about this discussion, then surely you know about the many papers demonstrating that relative risk can be misleading in presenting medical information. These are just a few, chosen because they had easier to quote chunks of information without having to cobble various quotations and points together.

            In Cochrane:

            Using alternative statistical formats for presenting risks and risk reductions (http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD006776.pub2/abstract)

            For communicating risk reductions, relative risk reduction (RRR), compared with absolute risk reduction (ARR) and number needed to treat (NNT), may be perceived to be larger and is more likely to be persuasive. However, it is uncertain whether presenting RRR is likely to help people make decisions most consistent with their own values and, in fact, it could lead to misinterpretation.

            In the Journal of the National Cancer Institute
            Helping Patients Decide: Ten Steps to Better Risk Communication (http://jnci.oxfordjournals.org/content/early/2011/09/19/jnci.djr318.full.pdf)

            Recommendation 2: Present Statistical Information Using Absolute Risk Rather Than Using Relative Risk or Number Needed to Treat Formats

            In the American Journal of Medicine:

            Absolutely relative: how research results are summarized can affect treatment decisions

            Of the 235 physicians who completed the questionnaire, 108 (46%) gave different responses to the same results presented in different ways. Of these, 97 (89.8%) indicated a stronger inclination to treat patients after reading of the relative change in the outcome rate (p less than 0.0001).

            The paper you linked to also shows that relative risk does show a much bigger change when there’s a change in absolute risk. It’s also a paper about rare events in a political context (wars, epidemics, etc.), events that if they occur certainly would affect many people. I think it’s a stretch to compare that to individual health decisions and risks, especially when there are many papers about this subject specific to medicine you could have looked at.

            Perhaps the crux of this is whether we agree on the principal of informed consent or not. I think truly informed consent in making personal medical decisions is based on accurate information. If presenting the case against homebirth in a way that exaggerates it is worth it to persuade people of the dangers of homebirth, then we fundamentally disagree about informed consent. I think we should present the numbers accurately, then make the case that even this risk is not worth it. It is a small risk. The numbers show that. The point is that it’s a small risk not worth taking. It’s a risk that is likely to be even bigger than what the numbers show. Tell people this and why. Don’t inflate the numbers. Not only is the latter an ethical problem, it is a mistake when trying to persuade people who have a mistrust of medicine by giving them reason to mistrust the information about homebirth risks.

            Yes, if something can be misleading and you have a way to present the information that isn’t, that is absolutely reason enough not to use the potentially misleading approach. Do you really disagree with that?

            You haven’t even addressed that her percentage is based on a misleading comparison of groups.

            I feel like the comments have reached the point of being less a good faith discussion about how we present numbers and more a “defend Dr. Amy at all costs with any rationalizations available” discussion. Maybe that’s my misinterpretation. It stems from things like providing a citation that is not in the medical context when there are plenty you could look at in medical journals and such; telling me that there is an ongoing discussion about this issue yet also assuming I had just one cite to back my claim (when if you know of this discussion, you know there are many). Other people are arguing basically that her use of the numbers isn’t misleading because of various factors that do not actually make her use of numbers any less misleading. They only attempt to justify why it’s okay to be misleading.

            This whole thing really has very little to do with Dr. Amy and everything to do with discussing how we present numbers. Consider if you took her completely out of this equation and we were discussing an article written by a journalist in the NY Times or something. How many people would be bending over backwards to rationalize why it’s okay to make this faulty comparison between groups and then use the results of that faulty comparison with a tactic that is straight out of Sensationalist Journalism 101? Please consider this honestly.

            Or don’t. I would have loved a good faith discussion on this, but wading through the bad faith gotchas throughout these comments is exhausting and a waste of my time. I think this discussion has run its course, at least for me.

          11. Thank you for your response. Yes, I am a fan of Dr. Amy’s, but I am aware of cognitive bias (it is in fact a large part of what I study) and I actually spend a good deal of energy trying to be sure her claims are correct before defending them. I will even admit that the 450% is rounded up from 415%, for no apparent reason, and that some of Dr. Amy’s comments on this thread are rather tangential. Nevertheless:

            1) Maybe you do have more than one cite, and you do know about the debate. But how am I, as a reader, supposed to know that? Especially when you present a contentious issue as if it were decided, and cite the same one piece several times?

            2) You are right that I presented a social science article. That is because I am a social scientist. But the laws of statistics don’t change based on the subject we are analyzing. To the extent that the subject matter changes the *interpretation” of those results, say, by changing how bad a “bad outcome” is, then we are at the heart of the matter.

            3) There is no doubt that relative risk and absolute risk, presented alone, produce *different* understandings of risk, and that generally relative risk causes people to believe the impact of independent variable is larger. (see e.g. http://mdm.sagepub.com/content/29/2/167.long) But what you have not explained is why this makes presenting relative risk WRONG. Why is it your argument that presenting the relative risk causes people to OVERestimate the risk, when one could just as easily make the argument that presenting absolute risk causes people to UNDERestimate risk? To distinguish between the two, you have to have some sort of objective measure of whether a risk is big or small. And there is no such thing. Well-educated, well-intentioned and intellectually honest academics can spend ridiculous amounts of time arguing about whether the impact of X in study Z is big or not: I’ve been part of many such debates, and none were every resolved. The norm of providing both relative and absolute risk arose out of an attempt to circumvent these debates and to acknowledge that there is no one true measure of “risk”. Dr. Amy presented both relative and absolute risk and is therefore totally in the clear by normal academic standards, and standards for informed consent.

            This is all to say that no one objects to your interpretation that a) absolute risk is more important than relative risk and that b) the absolute risk is low. That is a reasonable interpretation and entirely up to you. The part that we all keep pushing against is that you have apparently decided “home birth is a small risk” is the ONLY correct interpretation and that presenting information that could possibly lead to a different conclusion is dishonest. Which is, in itself, a strikingly disingenuous conclusion.

  11. “If you are against homebirths and you think that any possibility of a risk is not worth taking, then argue that without resorting to inflating the numbers. It is my opinion as a skeptic that we should never, ever, ever misrepresent information in order to encourage others to take our side.”

    Yes. Thank you. It seems almost as if many people are saying some combination of “they do it too, so it’s okay,” and “the ends justify the means,” and have truly, epically, spectacularly missed the point that criticizing ONE analysis of ONE study is not the same as home birth advocacy nor can we particularly hope to make a compelling case against current practices with data so unsound. Countering misinformation has never meant “except for people we agree with.”

    I know it must have seemed insane to risk clarifying your point given the laws governing the bottom half of the internet, and the contentiousness of the topic involved (and the 200+ comments that were not all productive or kind on your last post), but I hope you keep in perspective that a lot of the voices taking vociferous umbrage (especially the most vitriolic and…um…repetitive) are mostly coming over from a website that has something of a reputation for its readers lancing out and attacking the slightest criticism. Stay strong.

    1. Can it be that there are those of us who are sick of the risks being downplayed? When you know women personally who have lost their babies at home births, these discussions take on a different meaning. When you have seen the lies and censoring and misleading information all touting/protecting this idea that home birth is just as safe as hospital birth — even when there isn’t a single study done on home birth in the USA to support that — it gets exhausting and frustrating. I see these little babies in my Facebook feed – the babies of my friends that died preventable deaths. My friends who were not given informed consent and/or real information regarding the risks. I don’t think it is OK to lie about the rates no matter what side anyone is on. I don’t think it serves anyone at all. But it is so, so frustrating that the risks are completely brushed under the rug. The risks of epidurals and hospital infections and any interventions at all are blown completely out of proportion and yet, the simple truth that home birth increase risk of mortality is brushed off… excuses are made (I have heard too many to count!) and all these official sounding governing bodies are filling women full of just crud.

      It is so frustrating to be blown off…. people will say “oh don’t listen to them, they are trolls / minions / coming over from a website that has something of a reputation / whatever.” No, no. I am someone who is sick of seeing preventable deaths! I am sick of the lack of accountability for home birth midwives! I am sick of the lack of transparency for home birth midwives! I am sick of loss moms being blamed and midwives being protected, no matter the level of negligence.

      I am SICK of the risks of home birth being minimized.

      Acknowledge that it is completely possible that there are people fighting for women to know the truth because we are sick of seeing innocent, wanted little babies die preventable deaths.

      1. If you are genuinely tired of that, I would strongly suggest that when you do go to another website that has been linked (knowing full well that there will be half a dozen or more of you doing the same thing) that you work extra hard to communicate in good faith instead of all coming in with guns blazing. Bringing the dog-pile with you AND being hostile is a pretty good way to get fairly labeled as trolls.

        I’m sick of the risks of home birth being minimized as well. I have a VERY VERY VERY personal stake in hoping that good, solid, strong data exists before my partner wants to get pregnant again. But the thing is, I didn’t see a single person arguing against increased training and transparency of data. In fact, every G.P. writer and everyone I saw on that thread (I may have missed a comment or two) was WILDLY in favor of it. Everyone here AGREES with you.

        If you’d focus on what we (all) AGREE on, you’d see that the only thing that has been done is to analyze the SINGLE conclusion of a SINGLE study for the purposes of examining how the same statistics get read differently by different sides. Everyone just lost their shit and started calling us home birth advocates because Jamie said DrAmy’s analysis was sloppy. Literally some of the comments were “how can you criticize someone who is doing such good work.” (Because that’s what skeptics do. They don’t have a side that is too good for criticism.) None of us wants home births to continue with current levels of training or ignorance surrounding them, so maybe it’s possible to have a conversation that is productive and civil without the usual patterns of people reacting to a post of DrAmy by swarming the link and being nasty. That you’ve done so probably has more to do with being dismissed (and may in other places on the net as well) than anything.

        1. You are right. I should not be offensive in my posts and I apologize for going overboard. It’s an emotional topic. I do not agree with Jamie’s posts though and that is where part of my frustration lies.

        2. Here you are, good strong solid data! http://groundedparents.com/2014/02/26/guest-post-homebirth-safety-and-risk/#comment-1599

          I mentioned in the other post just now, I think this kind of analysis would have gone down a lot better if there had been an earlier treatment on GP of “what we know about home birth risks” to point to first. The topic hasn’t been as rigorously done in skepticism as, say, vaccination; there isn’t as much activism for “holy shit, ban these uneducated midwives who get in over their heads even as they don’t realize they are!” as “dammit, vaccinate your kid!” So something like this comes across as more critical of the entirety of anti-homebirth than it would of pro-vaccinate.

          There’s good cause to step a little more lightly when it hasn’t been established that there’s general consensus (movementwise, medically see above).

          1. You know the thread you’re pointing back to DOES have a couple of prior articles in it that are anti-home birth and hypercritical of the idea of a “right” kind of birth respectively. They were pointed out to one of the people saying they were disappointed in the whole blog and threatening to leave. I believe some of our writers have linked the data in this comment (awesome job,by the way) before. Probably not all of these links though. Kudos for putting it all in one place.

            I think one of the frustrating things from THIS side of these interactions is that certain groups only show up to be furious that we’ve said certain things. It feels like we’re in the Homebirth Debate version of The Invisible Boy, where we can write tons of articles against pseudoscience and even home birth and no one will notice but if one of our writers happens to write about the wrong topic, then suddenly the entire blog is criticized for not giving enough air time to the “right” side. The miasma of hostility gets so hard to see through when nit picking over one study that it may be hard to take a step back and realize that the overall thesis (and the one which most of the GP writers commenting on the threads have tried to reinforce) is that we would win more hearts and minds with good data.

            We already agree. But if you can’t convince people who agree with you that your data is good, then maybe it’s not actually that good. Let’s go find BETTER data. Let’s slam dunk them!

          2. Dude. I’m trying to bring you the data that all say the same thing to various magnitudes of horror and you want to get personally snarky about the fact that I was very upset about a thing in the immediacy but am trying to be calmer and more rational in the hours and days later? Thanks.

            I don’t know what to tell you. The numbers probably aren’t going to get any slam-dunkier. That’s four papers or studies in the last year, on several different metrics, plus this MANA paper which slots right in to them, death rate-wise, even as it tries to claim that home birth is totes safe. If someone needs more than that, they’re probably not going to get it. If a death rate being three times higher, at the best, plus increased risks for long term damage isn’t going to do it, I don’t know what will.

            GP has lots of posts about personal experiences and different philosophies and withholding judgement about what is “best”, and that’s great. But there haven’t been many with numbers. Jenny Splitter’s was good, but an N of 1 is not a consensus. There are a lot of things I take for granted as “solved” within skepticism–vaccinations, evolution, abortion, homeopathy and the vast majority of alt med–but home birth has not struck me as one of them. So to see the first (two, now there’s two!) really meaty posts about statistics be these (and this second one is just a travesty) is off-putting when I have better numbers sitting in front of me, begging to be used to tear some pseudoscience-y shit up.

          3. What was “personally snarky” about that reply? I’m honestly asking you. I read it three times; I have no idea what you’ve picked up on.

            Given that you HAVE better studies than the one being criticized, and that no one is suggesting that better training and transparency aren’t important, I’m not exactly sure why there’s been such personal affront that some data has been examined critically if better data exists and no one is using that data to call home birth perfectly awesome.

            The author said she’s not advocating home birth. Every writer that has commented has said basically the same thing regarding transparency and increased training. What kind of consensus were you looking for?

          4. “They were pointed out to one of the people saying they were disappointed in the whole blog and threatening to leave.”

            That wasn’t snarky? How was it intended?

          5. It wasn’t meant to be snarky. That’s what happened. A commenter said they were disappointed and going to quit reading G.P. Fortunately, some of the writers explained that we were not a united front, including linking some of the articles we’ve done that have been pretty anti-home birth.

            You know we’ve only been at this for three months now. Our magnum opus of anti-home birth statistics post may simply not be here YET. (Also you may like one of our posts going up tomorrow.)

          6. Okay…so I didn’t realize it was you. But even so…isn’t that what happened? You said you were going to leave because you were disappointed in the whole blog. Several of us pointed out that we had many opinions and linked you some other articles.

            I’m not seeing the snark here, and I’m genuinely trying to. Did I mischaracterize your original comment: (“Unfortunately I don’t know that I can stick around anymore.”) as a threat? In that case, I apologize.

          7. Is that what you would have said if you knew it was me? I’d assume not. Imagine writing that knowing it was me, and how much differently you’d have to think to write those same words. That’s where the hurt was–and I was hurt, because I thought I was bringing you the data you’d asked for and you were making fun of me in response.

            I don’t know the reasons your partner opted for a home birth, but I hope those studies and any further ones change her mind in the future because I absolutely believe that home birth is a bad idea. I don’t know how much less-than-scientific ideas were part of the decision, but her reaction you described at having to have a C-section it involved some less-than-scientific pressure to have a baby in a certain way, to avoid “failure”, to do “the best” for her baby and that’s a damned shame. I look forward to the Grounded Parents anti-home birth magnum opus. I just wish these two posts hadn’t come first.

          8. I swear to not-god I proofread. Something about the comment box just throws me off.

            It should be: “but based on her reaction you described at having to have a C-section,”

          9. My partner opted for a home birth because she wanted one. She doesn’t like hospitals and she didn’t want a new person every eight hours feeling her up. She knew the risks, had no complications and made a choice. No one–even the midwives–pressured her to have a vaginal birth. Honestly our midwives were easing her into the idea of the c-section before the OB recommend it. (It’s one of the reasons I liked them.)

            As for the other part, I am honestly, genuinely sorry. I probably *would* have phrased it the same way even if I’d known it was you, so maybe that’s something I need to think about. I DO appreciate the data.

      2. Can you name ONE friend who wasn’t high risk? Post dates, breech births, premature births, infections, red flags after red flags, VBACs, multiple VBACs.. etc. Jamie is stating quite clearly that homebirth is dangerous for high risk women. There simply is no evidence that the same is true for low risk women. The point is, if you claim to be a skeptic, if you claim to want to women to have informed consent, you will not lie and inflate numbers to “prove” homebirth is dangerous for everyone, period. It’s just not true. Can you tell us about one low risk women you know of who lost a baby at home? It almost never happens. I only know of one woman and her baby turned out to have a inoperable heart defect.

        1. There is evidence that even low risk homebirths are less safe, and the analysis being criticized here is part of that evidence.

          As for examples of babies who died because a low risk mother attempted homebirth, Dr. Amy did a rundown of homebirth deaths and near deaths from 2012 and 2013, and if you sift through them you can definitely find ones that were low risk. Here are links:

          http://www.skepticalob.com/2013/12/2013-the-year-in-homebirth-deaths-and-disasters.html
          http://www.skepticalob.com/2012/11/15th-and-16th-homebirth-deaths-i-have-reported-on-this-year.html

          From my personal experience, my cousin’s friend who was low risk attempted a homebirth. There were complications that lead to her being rushed to the hospital to deliver. Because of the delays getting to the hospital she was in bad enough shape that they had to helicopter her to another hospital better equipped to deal with her condition. Both mother and child lived, but just barely, and it is likely the child will suffer long-term effects from the trauma.

        2. mmella –

          3 shoulder dystocias and 1 fetal distress. Low risk women.

          I should let you know, though, that certain women have become my friends *after* I started my blog / advocating for safer home birth midwifery.

          “There simply is no evidence that the same is true for low risk women. ” That is not true. The risk is higher at home even for low risk women, as every single study on home birth in the USA has demonstrated.

          Home birth can be a reasonably safe option! When you have highly trained/educated/skilled midwives who have hospital privileges, who carry insurance, who legally carry the necessary drugs/equipment, who are practicing within their scope, who have a good relationship with an OB/GYN and hospital, who risk mothers out who are not truly low risk, who have a second midwife or birth attendant who is trained in neonatal resuscitation and CPR at each birth with them, and when the home is in close proximity to the hospital where the midwife would transfer the mother. THESE are reasons home birth works well in other countries. This is not the norm for home birth in the USA though. This is the *minority* of home births in the USA. This is why, even for low risk mothers, the mortality rates are too high at home. This should be the MAJORITY of home births! If people wake up and open their eyes to these issues and start demanding better then we will get there some day! Safe options for home birth will be available to more and more women but only if people start demanding changes to the way things are currently being done.

          1. If I could “like” this post I would. These are the standards we should be advocating!

          2. I so agree, Douladani. We need standards like those–standards like every other civilized country on earth.

        3. Hi Mmella, just wanted to clarify: I was not saying that there was no evidence of risk in low-risk homebirths. I was saying that the evidence (from just this one study and the numbers Prof Orosz got from the CDC because I did not do a literature review) suggest that there is probably a small risk even for low-risk births.

          1. Wait, what is your definition of “small”? If you have two alternatives (A and B) and one of them, B, is almost *five times* more likely to kill your baby, I think it’s a little misleading to call B a “small risk.” It may be small in the absolute sense (i.e., your baby probably won’t die). It’s not small in the relative sense, i.e., in terms of helping you personally DECIDE BETWEEN A and B–because both risks are small, but one of them is almost five times bigger (and that’s just the death risk–which is inherently going to be smaller than the risk of a brain-damaged baby).

            And it’s also not small if you’re looking at it from a policy perspective. If you were a legislator, would you really be okay passing a law that’s going to kill–say–1000 babies a year, when your other option would mean only 250 babies a year would die?

            And finally, I don’t see what possible basis you could have for using the word “probably,” as in “probably a small risk.” We’ve got two studies here–the MANA one looking at 17,000 home births, and the Cornell one looking at over 10 million births (including more than 60,000 home births). They both showed almost identical death rates for babies born at home with midwives. There is no “probably” here; there are almost 80,000 home births (of which almost 60,000 were with midwives), which is a far bigger cohort than you’ll typically see in *any* medical study (including probably every study you have ever based any of your medical decisions on). Two different studies by two totally different groups of researchers, tens of thousands of births, and almost identical results: Cornell shows home births as being about 4.25 times more likely to kill the baby, while MANA shows they are about 4.5 times more likely to kill the baby.

            You have no basis–no numbers, no studies, no logic–for dismissing as merely “probable” the death risk that both these studies show.

        4. Two words: shoulder dystocia.
          Two more words: placental abruption.
          Those can happen to women regardless of whether they’re high or low risk. They can kill a baby in minutes, so if you’re not at the hospital, the fact that your house is “only 15 minutes from the hospital” doesn’t help you.

    2. In other words, you still don’t understand, and you appear to think you’ve done your due diligence by asking your friends at Skepchick if they agree with you. Pathetic.

        1. “You still don’t understand my point” is not an ad hominem attack. “You think you’ve done your due diligence by doing X” is not an ad hominem attack. Dr. Tuteur expresses herself very brusquely, sure, and many people dislike that, but she is absolutely engaging with your argument and telling you, over and over, why *your argument* is wrong.
          An ad hominem attack isn’t one that says “here are the specific flaws in your understanding, which lead to flaws in your argument.” It’s not that at all; it’s an attack in which “a claim or argument is rejected on the basis of some irrelevant fact about the author of or the person presenting the claim or argument.” http://www.nizkor.org/features/fallacies/ad-hominem.html
          Random, simplistic example: “You’re a woman, so you couldn’t possibly be right about this.”
          And however much you may disagree with her, I haven’t seen Dr. Tuteur reject your argument on the basis of some irrelevant fact about you. She’s rejecting it because, as she has explained, your posts show that you still do not understand her point. If you don’t understand someone’s point, you can’t really make any effective arguments against it. And I’m not sure what she means by “due diligence” in this case, but generally due diligence means something along the lines of “doing the necessary intellectual work–such as research, analysis, etc.–to figure something out.” Apparently Dr. Tuteur thinks that instead of doing such research, analysis etc., your posts indicate that you just basically checked whether other people agreed with you and decided that since some of them did, you must be right. So whether or not you or anyone else agrees with her, that’s still not an ad hominem attack at all. It’s an attack on your understanding of the issue and your methods of deciding what is correct–and both those things are totally relevant.

          1. Actually, she is in fact arguing against me the person in place of arguing my points, which she didn’t respond to at all. And her argument against me as a person, the point about asking my friends at Skepchick and thinking that’s due diligence, has no basis in reality. See my argument above (the one with the boldface points) that she is responding to here. She is not addressing my points at all but instead making up something about me talking about this with friends. I don’t even know what that means. My arguments are my own, and if they are flawed, please do point out the flaws. She hasn’t, and by responding with this comment undermining me, she is replacing an argument against my points with an argument against me as a person.

          1. The “pathetic” was silly and insulting, I totally agree. But I don’t think saying “you still don’t understand the point I’m making” is the intellectual equivalent of “nuh-uh,” and if the person you’re discussing something with still doesn’t understand the point you’re making, why would you bother responding to *their* points? Their points are not relevant if they don’t understand *your* point. I’ll try to illustrate this:

            Person 1: X is true.
            Person 2: I don’t believe that Y is true because of points A, B and C.
            Person 1: I wasn’t talking about Y. I was talking about X, which is not true.
            Person 2: But Y is not true for these reasons!
            Person 1: You still don’t understand what I’m trying to say…

          2. I totally agree with your example, daleth, except that I’m person 1. That is in fact exactly what she’s doing. She’s done more to respond to anything BUT X is in this entire conversation. X is about whether she used misleading tactics in her post. In the bulk of her many words on these posts in her comments and on her blog, she has raised all the letters of the alphabet to talk about but has continuously avoided the actual subject X, addressing her misleading use of numbers in her post. That is EXACTLY why I’m so frustrated.

          3. ChrisBrecheen, saying “pathetic” is harsh (and pointless), but if it’s referring to the other person’s argument or their intellectual approach to the argument, as opposed to referring to them overall as a person, it’s not ad hominem. “Your argument is pathetic for the reasons I just stated” is not the same as “you are pathetic.” I’m not at all saying it’s a good debate style or something that’s totally fine, but dismissing someone else’s points as ad hominem when they’re actually not ad hominem at all isn’t either. (It may be less offensive than saying “pathetic”, but it’s still not a valid criticism.)

          4. Actually you said “silly and insulting,” but if you think “insulting” in a comment that derisively brings in people’s “friends” isn’t ad hom, that’s cool.

            Actually, you’re right (officially). In order to be an ad hom attack (officially according to debate forensics) the fallacy would have had to have been ensconced in an actual argument. It wasn’t.

            I would take you more seriously (honestly, I’m being genuine here) if you didn’t seem unable or unwilling to point out bad arguments when it’s “your” person making them.

        2. So do you believe that women’s health organizations are fear mongering when they point out that maternal mortality among African American women is 3 times higher than for white women?

          According to your claims enunciated above, we should ignore the difference because the numbers are so small (measured per 100,000). And we should accuse those women’s health groups of misusing data to sow fear, while rolling our eyes at their use of graphs that show “OMG 3 times as many black women die in childbirth as white women!” Is that what you mean Melanie?

          1. You really aren’t going to address my actual points, are you? You’re now bringing up a completely incomparable example as though it’s the same thing. So in response to my telling you you’re making a faulty comparison, you make another one.

            Did you read the links I provided that explain exactly what I mean about small numbers and transparent use of relative risk? It doesn’t apply to the example you’re providing here. The only way any of the points I’m making would apply in that instance would be if they were reporting the rate without giving the numbers the rate is based on. That would also be a misleading use of relative risk, although of a different kind. Again, I provided a link to a study that explains this as well as a link to a very brief explanation of when you don’t report rates with small numbers and why (the Rule of Twenty link). You’re also not addressing the problems I pointed out in your graph, and yes, if I saw a women’s health organization using a graph in the way that you did, excluding one of the comparison groups and including the data of one bar in the second bar to inflate it, I would say something.

            If you can’t or won’t address my points, please just say so or move on, because this topic shifting to avoid doing so has gotten beyond ridiculous.

          2. Why is maternal mortality different from perinatal mortality? The only difference I can see is that maternal mortality involves even smaller numbers than perinatal mortality.

            If it is valid to note the discrepancy in maternal mortality between two groups, how can it be fear mongering to note the much larger difference in perinatal mortality between two groups?

          3. As for the rule of twenty that you cited above, how does that apply when the deaths in the MANA cohort numbered 35?

          4. The maternal mortality example you gave was of numbers for which there are enough people to calculate out of 100,000. This study doesn’t even come close to having 100,000 to pull the numbers from. This study shows 35 total deaths out of under 17,000 births. That is why these are not comparable. The total number of deaths we are talking about with maternal mortality is itself much higher than the 35 we’re talking about because that is a discussion of a much larger pool of people to draw from. Your post was about the MANA data specifically.

          5. So you’re saying that MANA’s claims about the safety of homebirth are meaningless because the are based on 17,000 births? Why didn’t you say that before? Are you suggesting that we should just ignore the MANA paper altogether?

          6. The Rule of Twenty doesn’t apply when calculating an absolute risk rate. I did not point out any problems with calculating the absolute risk rate. It does point out that when the numerator is this small that the rate can change deceptively in comparisons. The other article I linked to then goes into more depth about how misleading it is to use relative risk percentages with small numbers, including examples of this using small numbers that are larger than what we’re talking about with the MANA study.

            The 1200% increase in shark attacks itself illustrates pretty clearly how misleading this tactic is. I used your exact tactics in that post to illustrate why these are misleading ways of presenting numbers. If they were not misleading tactics, I would not have been able to use them to mislead. The difference between my post and yours is one of degree (my examples are ridiculous, even if the numbers are accurate) but not a difference in kind (the misleading tactics are the same).

            And you are not even addressing that the percentage you used was based on a faulty number comparison, that you omitted the all risk hospital bar for comparsion in the chart, and that you created a third bar that includes the numbers in the second bar to inflate it. The entire post was misleading in multiple ways.

          7. So you’re saying that MANA’s claims about the safety of homebirth are meaningless because the are based on 17,000 births? Why didn’t you say that before?

            I did. Repeatedly.

            Are you suggesting that we should just ignore the MANA paper altogether?

            No. I’m suggesting that we address it honestly, not by being misleading in presenting the data that is in it. There are many, many ways to criticize the MANA data and survey itself without artificially inflating the numbers in our comparisons.

          8. Okay, you seem to be suggesting that the increased death rate in the MANA data is not statistically significant because the number of deaths, 35, is too small. Unfortunately, this is not true, and that’s exactly what my article is about.

            Let’s use my comparison rate of 0.7 deaths per thousand. (Remember, this is based on hospital births which are overall probably significantly higher in risk than MANA’s group). Assume that the risk of death at home birth really is 0.7 per thousand. What is the probability of finding 35 or more deaths in 17,000 women through bad luck alone? Because the numbers are pretty small, let’s use the binomial probability formula directly. My computer says: 0.00000004, or 4 out of one hundred million. Coincidence has essentially been eliminated as an explanation for this difference, especially since the alternate explanation, that home birth is dangerous, is quite plausible.

            Going from one shark attack death to a dozen, in a whole state, means nothing. The number of shark attacks certainly does vary from year to year, but overall remains quite low. There’s also no reason to think that sharks have somehow suddenly become more dangerous.

            Getting 35 deaths instead of 12, on a sample of 17,000 (basically a small town) means plenty.

          9. Okay, you seem to be suggesting that the increased death rate in the MANA data is not statistically significant because the number of deaths, 35, is too small.

            No, that really isn’t what I’m saying. What I’m saying is that these numbers are too small to be presented with a relative risk percentage. That presentation of the numbers is misleading. The shark example (1200%) illustrates this tactic. I also provided links to more information about misleading uses of relative risks as a problem among physicians, patients, and journalists that further illustrates why this is a misleading tactic to use even when the numbers are accurate.

            Further, the percentage is based on comparing numbers for low- and high-risk homebirth to low-risk hospital birth, which is a faulty and misleading comparison. An honest approach to the numbers would be to compare low risk to low risk, then discuss other factors outside those numbers that make the risk likely to be even higher with homebirth. You don’t artificially inflate the numbers to make this point.

            My points are ENTIRELY about how we present the information, not what that information actually says. That’s a different discussion. The reason how we present the information is important to me (aside from the straightforward argument for being honest for the sake of being honest) is that we are dealing with many people who already have a mistrust of medicine. We do more harm than good in persuading people of the risks of homebirth when we are presenting our arguments in misleading ways.

            I have commented many times about the study itself being almost useless for many reasons. So no, it absolutely does not tell us that homebirth is safe. We can point this out and talk about the flaws in the study and talk about many other aspects of homebirth that people may not realize (such as low risk not always being low risk during labor and childbirth because many problems don’t show up until that point). What we should not be doing–what we don’t need to do–is present the numbers in misleading ways. I have to admit, I’m kind of exhausted by how difficult it has been to make this point in this discussion, which seems to have derailed into discussions about anything but the topic of how we present numbers.

          10. Melanie, you’ve been told a number of times by a number of people, including a math/statistics professor, that your idiosyncratic “interpretation” of the statistics of small numbers is flat out, embarrassingly WRONG. How long do you plan to continue making a fool of yourself?

            You and your compatriots have done incalculable damage to the reputation of both Grounded Parents and Skepchick. There doesn’t appear to be a single person among you who understands basic statistics. People wonder why women aren’t taken seriously in the skeptic community. Behavior such as this, evincing ignorance, pettiness and a total unwillingness to learn from your mistakes, is part of the reason why.

          11. Wow.

            Dr. Amy, you say that Melanie is “embarrassingly WRONG” in her statistical analysis. Unfortunately, I took only three credits of Probability and Statistics in undergrad, and I’m not versed in the application of statistics to social science studies, so I really don’t know enough to agree or disagree. Maybe Melanie is right and you aren’t, and maybe you’re right and Melanie’s wrong. But that’s not why I say “Wow”.

            No, I say “Wow” because of the sexism: “People wonder why women aren’t taken seriously in the skeptic community. Behavior such as this, evincing ignorance, pettiness and a total unwillingness to learn from your mistakes, is part of the reason why.”

            So, if a woman is wrong in a disagreement with you, the problem is that she’s a _woman_. When you don’t budge an inch, it’s admirable. When you write an article with the expressed intent to sully Jamie’s reputation in Google searches, that’s the non-petty behavior of a skeptic. But if Melanie holds her ground with calmly-expressed citations, it’s an indictment on women overall as ignorant, petty, and obstinate.

            Again, I don’t know if your initial analysis of the study in question was impeccable or deeply flawed. And perhaps GP and Skepchicks damaged their reputation in the eyes of your fans by having temerity to subject your claims to further analysis. Either way, though, your reliance on insults, argument from authority, and blatant sexism in this comment, other comments, and the posts on your blog about this matter have certainly done damage to your reputation outside your bubble.

            And contra your post in which you claimed without knowledge or evidence that your comments were held up or deleted due to a sexist insistence that GP wants you to ‘behave like (a lady) and be “nice”‘, this isn’t about your being a woman. Objecting to your reliance on insult, fallacy, and sexism isn’t an attempt to put you in a box with a pink bow – it’s disappointment that you hurt the cause of skepticism. Your mistakes are your own, not your gender’s, and knowing the difference isn’t censorship.

          12. Clearly Lance; this seems to be the M.O.: If your awesome argument that can’t even convince a bunch of skeptics who already mostly agree with you, change the subject to talking points not in any way related to the thread or the OP. Begin insulting anyone with the temerity to stay on point. Use a rotating melange of fallacies to “win” (appeal to authority, ad populum, ad hominem, strawman, Ignoratio elenchi) rather than communicate with the slightest good faith. In no way address the actual points even when they are repeated MULTIPLE times, but stridently defend *other* points instead. Ignore someone even when they say “no, that’s not what I’m saying” and try to correct course in good faith.

            If all that fails, invoke sexism.

          13. Just curious: is there anyone here defending Jaime and Marlene’s creative “interpretation” of statistics besides other Grounded Parents’ writers? It seems that it is Grounded Parents vs. everyone else, suggesting that the writers on Grounded Parents ought to spend more time learning statistics and less time devoted to supporting their colleagues mistakes.

          14. I think you should spend more time working on effective argument strategies instead of just insulting everyone.

  12. Jamie, you still have yet to respond to what I wrote about the breech births. Same thing goes for the low APGAR comment. Dr. Orosz did not manipulate. You are not reading things thoroughly enough (or at all).

    Here is what Dr. Orosz wrote regarding breech births:

    “Disturbingly, the breech data were incomplete, as the authors explained: “There are 4 singleton pregnancies, 3 of which were breech presentations, for which all birth outcomes data are unavailable. These women began labor at home and then transferred to the hospital prior to birth. The midwives of record were contacted, and in each case the midwife did not accompany the mother, nor did the mother return to the midwife for postpartum care.” This missing data means that the breech death rate could in fact be as high as 36.0 per thousand, or 3.6%, which is similar to the breech birth death rate in the USA circa 1950.
    Had those breech infants been born in the hospital, there is at least an 86% chance that all of them would have survived, probably higher, and a 99% chance that no more than one would have died. Instead, at least five and possibly eight babies died.”

    This was your response, Jamie:

    “In other words, Prof. Orosz is assuming that in all cases in which outcomes were missing for mothers who transferred to a hospital with a breech birth, the newborn died, though she gives no evidence for why this may be true. ”

    “I don’t think I can possibly impress on you how not ok it is to just assume all missing data must mean dead babies.”

    “Even if Prof. Orosz is trying to prove the point that homebirths are a bad idea in the case of breech births, there is no reason to manipulate the data.”

    You accuse her and her analysis here THREE times because you did not read correctly. She did NOT manipulate the data. She brought up a good, valid and true point that there is data missing for breech births so the breech birth mortality rates COULD be even higher.

    You did the same thing again with your point (or lack thereof) of low APGAR scores.

    Dr. Orosz wrote this:

    “If all 401 of those newborns in fact had low APGARs, the true rate of low-APGAR births would be 3.8%, which is three and a half times the national average.”

    You again accused her of manipulating the data: “No. Just no. Just because a baby’s APGAR score is missing in the dataset doesn’t mean it was a low score. Maybe some of them were, but we don’t know because they are missing. That’s kind of what “missing” means. Unless you have a really good reason to assume that all missing data = worst case scenario, you cannot just make that assumption.”

    Again she brought up a valid concern, that APGAR data is missing. She said “IF all 401 of those newborns in fact had low APGARs….” She did not make assumptions and manipulate data. She did not say “those 401 newborns probably had low APGAR scores” or “those 401 newborns most certainly had low APGAR scores” or anything like that. She said IF. It’s a valid point.

    And lastly, your point about the authors, which has already been addressed here in the comments by myself and Amy Tuteur, is not true. Not just one of the authors is a midwife. Three of the six are midwives. Here are the six authors and some things worth noting about them:

    1. Melissa Cheyney – Certified Professional MIDWIFE and Chair of the MANA Division of Research
    2. Marit Bovbjerg – Director of Data Quality for the MANA Division of Research
    3. Courtney Everson – Director of Research Education for the MANA Division of Research
    4. Wendy Gordon – Certified Professional MIDWIFE and Director of Social Media for the MANA Division of Research
    5. Darcy Hannibal – a home birth mother
    6. Saraswathi Vedam – Certified Nurse MIDWIFE and Senior Advisor for the MANA Division of Research

    Aside from the fact that you noted incorrectly how many authors are midwives, I’d just like to ask you, does this look like an unbiased group of researchers?

    Jamie, you are arrogant in your writing and clearly too proud to make changes to your work since you didn’t fix any of the errors (other than the math error) in your last piece. So I am not holding my breath that you’ll do the right thing. You make comments about you being personally attacked when you are doing it to others. First to Dr. Tuteur. Now to Dr. Orosz. All because you lack reading comprehension skills and cannot seem to dig a little deeper in your research. Any time anyone Googles your name they will see these two pieces you wrote, they will see Dr. Tuteur’s response here and on her website (especially on her website), they will see these comments that point out your mistakes and they will see that you made no effort to change what you wrote.

    “The truest characters of ignorance are vanity and pride and arrogance.” – Samuel Butler

  13. While I agree that majority of homebirthing advocates are anti-science and anti-medicine, nothing about Jamie’s piece indicates that she’s giving that type of rationale her blessing.

    Yes, having a medically trained professional (like a licensed midwife or nurse) with you during a problem free, low risk homebirth is probably not any riskier than having a baby at a hospital. Not all hospital births require intervention. The problem is, IF there is a problem, it just seems like such an unnecessary risk to take when we are so fortunate to have highly trained physicians and life saving technology at our disposal, inside our local hospitals. I personally believe all babies should be given that opportunity, even if the chance they would need it (in a low risk birth) is small.

    As someone who believes that babies should have the right to the best medical care available, I think it improves our position when we’re willing to acknowledge that there is indeed a subset of women who can homebirth safely. Why? Because it gives us credibility, that’s why. I think that’s Jamie’s main point here. She can correct me if I’m wrong.

    1. “As someone who believes that babies should have the right to the best medical care available, I think it improves our position when we’re willing to acknowledge that there is indeed a subset of women who can homebirth safely. Why? Because it gives us credibility, that’s why. I think that’s Jamie’s main point here.”

      Yes Jessica, THIS. THANK YOU!!!

    2. Hi Jessicabc, maybe you already know this but just in case, I want to point out that “licensed midwife” doesn’t mean anything. The only midwife qualification that actually means the person has medical training is Certified Nurse Midwife (CNM). A CNM has a nursing degree and a Master’s degree in midwifery. A “licensed midwife” could have nothing more than a high school diploma, CPR certification (just like you could get in 2 hours down at the YMCA), and a letter stating that she has assisted a midwife at 50 or more home births.

      Be very careful when choosing midwives. What you want is a CNM, not a CPM or LDM (“licensed direct entry midwife”) or any other letters after her name. CNM = real midwife. Everything else = god only knows what you’re getting.

  14. I’m wondering WHAT data would satisfy Jamie and others.
    How many babies would have to die an unnecessary death, how many women would have to suffer serious complications until the data was good enough?
    Here’s the problem: obtaining such data would be fundamentally unethical. No respectable scientist would do a large enough cohort study for breech deliveries when there is strong indication that the risk is so high.
    As others have pointed out, it is reasonable to assume that the comparison is not unfair on MANA, but on hospital births.

    1. Hey Gillell, I’m a social scientist. I know what it’s like to work with imperfect data. I think we shouldn’t be dismissing data just because it’s really unclear and messy if its the best we have got. It’s ok to make decisions based on the best data we have, even if it’s not good. I wrote about that a bit in my post. I just don’t think we should be taking analyses from really messy data and presenting it as though the results are perfect without even mentioning the potential weaknesses. That’s just not honest.

      1. The problem with your comment, Jamie, is that this is NOT “really messy data,” nobody is presenting it “as though the results are perfect,” and the weaknesses in the Cornell study actually make home birth look safer by counting home birth hospital transfers (i.e., home births that developed complications serious enough to require a hospital transfer) as hospital births. IOW, the impact of that weakness in the Cornell study means that if anything, home birth is MORE dangerous than the study concludes. So referring to home birth as 4.25 times more likely to result in a dead baby (that’s the bluntest way to summarize the Cornell conclusions) is accurate–that is, it is certainly not an OVER-estimation–because the data and the way it was counted show us that home births are AT LEAST that dangerous.

        1. Daleth, I don’t know what study you are refer in to. I did not do a literature review. I reviewed only 1 study. Every study is going to have different strengths and weaknesses. Anything I said refers to only the one study I reviewed and no others.

  15. Another issue that can skew the numbers is the fact that women who choose homebirth are often more likely to refuse ultrasounds, refuse testing for things such as GBS and GD. They include the Amish, they include religious zealots who believe everything is “God’s Will”. They include women who might refuse prenatal care or have no idea what their true due dates are. The list is endless. But again this is not proof that homebirth is dangerous. I don’t see what is so difficult with advocating that high risk women be given true informed consent and at the same time, accepting that there IS in fact a group of women who can safely give birth at home. The benefits are plenty.

    1. If you’re a midwife, is it ethical to agree to assist at the home birth of someone who hasn’t even gotten a single ultrasound and has no clue of her due date? I would say no, because as a midwife YOU KNOW there is nothing you can do if her baby is very premature (which is very possible since she doesn’t know her due date), and YOU KNOW there’s nothing you can do if she’s carrying twins that share a placenta (in which case placenta abrupta during labor will likely kill Baby B unless you’re in a hospital), YOU KNOW there’s nothing you can do if she’s carrying monochorionic monoamniotic twins whose cords get entangled, YOU KNOW there’s nothing you can do if the baby is born with a congenital anomaly that requires immediate medical care to prevent death, and YOU KNOW that if the baby is breech (especially if it’s twins with the first one breech), although you theoretically may be able to help, there’s a very high risk you won’t be able to and the baby or babies will die…

      The list of things you can learn from an ultrasound, and that it is very important to know about before birth, goes on and on.

      And I can see an argument being made that it’s ethical to help because chances are she won’t have those complications and she’s better off with you than by herself or with some non-midwife trying to catch the baby, but that’s not a valid argument because most women, if the midwives say “Sorry, no, I can’t agree to do that unless you get an ultrasound because it’s too risky for you and the baby,” will probably go get an ultrasound. Problem solved: with the ultrasound info, now you can make an informed decision about assisting at their birth.

  16. mmella said, “Another issue that can skew the numbers is the fact that women who choose homebirth are often more likely to refuse ultrasounds, refuse testing for things such as GBS and GD. They include the Amish, they include religious zealots who believe everything is “God’s Will”. They include women who might refuse prenatal care or have no idea what their true due dates are. The list is endless. But again this is not proof that homebirth is dangerous.”

    Yeah, but those people tend to live off the grid anyway, so I’m not sure that’s affecting the data.

    For me, homebirthing is the antithesis of everything I stand for like science, modern medicine, access, and self awareness. I also have a big problem with much of the over-riding belief system of homebirthing, as a movement, because it seems to try to convince women that modern medicine and hospitals are totally unnecessary.

  17. Also, is there any way to change my password. When I click on my profile, it says I don’t have adequate permissions to do anything and I hate my password! It’s like 37 digits long.

  18. I am the author of the io9 post. While Jamie is correct that the headline doesn’t entirely sum up the group of professionals evaluating the MANA data, given that three of six authors were midwives, analyzing data from a midwives’ organization, and commissioned by that organization, I think the term “midwives” is a fair term for a headline. Jamie does not mention the fact that in the post, I write, “A group of PhDs and midwives analyzed the data and concluded, in a paper published in the Journal of Midwifery and Women’s Health.” In other words, I do bring up their academic credentials just as I brought up the credentials of Dr. Orosz.

    As for the claim that the study, “did not attempt to make any relative risk comparisons between homebirths and hospital births,” that is incorrect. The conclusions of the study clearly state “perinatal outcomes are congruent with the best available data from population-based observational studies that have evaluated outcomes by intended place of birth and by pregnancy risk profiles. Low-risk women in this sample experienced high rates of normal physiologic birth and very low rates of operative birth and interventions, with no concomitant increase in adverse events.” They analyzed the data by “intended place of birth,” and found “no concomitant increase in adverse events.” Comparing “place of birth” and finding “no increase in adverse events” is making a relative risk comparison.

    While I understand Jamie’s criticism of the headline and teaser blurb, and think she has a point, I would like to point out that she left out the fact that I did mention academic credentials. As for claiming that the study “made no comparison,” that is disingenuous, if not outright false.

    1. I read every word of that study. The conclusion claimed in the abstract and the conclusion that there was no increase in adverse events, but nowhere in the body of the study did I see the authors compare their own rate of adverse events to any other sources. It is a long document, if I missed it the comparison, please point it out to me!

      1. There was some stuff in the discussion. I don’t have the document handy, but I do have a couple of quotes I lifted from it (copied from a previous post, so the commentary is not directed at you or estheringlisarkell:

        “They say things like this:

        ‘Rates of spontaneous vaginal birth, cesarean, low 5-minute Apgar score, intact perineum, breastfeeding, and intrapartum and early neonatal mortality are all consistent with reported outcomes from the best available population-based observational studies of planned home and birth center births.’

        Comparing home birth to home birth, but don’t mention rates for hospital birth.

        Oh, but they don’t have a problem showing how “great” their rates of intervention are compared to hospitals:

        ‘Only 4.5% of the total MANA Stats sample required oxytocin augmentation and/or epidural analgesia, which is notably lower than rates of these interventions reported more broadly in the United States (26% for oxytocin augmentation and 67% for epidural analgesia). Rates of operative vaginal birth and cesarean are also substantially lower than those reported for hospital-based US samples (1.2% vs 3.5% and 5.2% vs 32.8%, respectively).’

        Finally this bit:

        ‘Low-risk women in this sample experienced high rates of normal physiologic birth and very low rates of operative birth and interventions, with no concomitant increase in adverse events.’

        Death is an adverse event. They cannot actually say that with a straight face, because they never compared home birth death to hospital death to show that there wasn’t an increase (likely because there *is* an increase).”

    2. So now we have a trifecta:

      Jamie wrote about my interpretation of the MANA data and I pointed out that she was sloppy, didn’t understand the data, and didn’t understand my analysis.

      Jamie wrote about Dr. Orosz’ analysis of the data and Dr. Orosz pointed out that Jamie grossly misrepresented her analysis.

      Jamie wrote about Esther Inglis-Arkell of discussion of the data on ios9 and now she has shown up to point out that Jamie made claims about her piece that are “disingenuous if not outright false.”

      Anyone seeing a theme here? I am. Bernstein has doubled down on her sloppy, misleading and incorrect claims and Grounded Parents is still defending those sloppy, misleading and incorrect claims. Three different professionals, from three different disciplines who all believe that they have been misrepresented and misunderstood. Isn’t it time to stop defending the indefensible and admit to the many mistakes.

      Got integrity, Grounded Parents? Or do you think it is okay to publish bloggers who have no respect for the truth?

    3. Hi Esther,

      I agree that I could have probably been a little more generous to the fact that you did mention in your piece that the researchers had PhD’s and that the article was published in a scientific journal. I mainly just felt like the title and introduction were misleading.

      Of the two sentences you quoted, the first one is comparing the risk rates of homebirths in their study to risk rates for homebirths in other studies, which they did to confirm that their data looks similar. The next line is ambiguous as to what they are comparing to. On my original read-throughs I assumed they were comparing to homebirth data from other studies done in other countries because they mentioned at the beginning of the article that they wanted to see if homebirth rates were higher in the U.S. due to differing regulations than in other studies and because the sentence before it was comparing homebirths to homebirths. However, the authors certainly do not clarify what they are comparing to and two sentences later they specifically say that they cannot compare relative risks for high-risk births between homebirths and hospital births, which implies that the previous sentence may have been making this same comparison. I’m still not sure. The abstract is far clearer that they are only comparing to data from other homebirths. If they are attempting to make a comparison to hospital births, then they are just as guilty as Prof Orasz at making a rough comparison and then presenting the results as fact.

      Even so, that’s clearly not what the study was about. The study was just stating the results they found from the homebirth group. This is pretty clear in the abstract and the study itself and I don’t think should be disregarded because of 5 ambiguous words in the conclusion. Prof Orosz analysis then built on that by comparing it to a data source on hospital births.

      It does seems that midwife organizations have been using the study to claim that homebirths are just as safe as hospital births, which is absolutely not intellectually honest. Just like Prof Orosz, they are trying to do a rough comparison than state the results as perfect fact when they are not.

    4. Oh and Esther, forgot to mention that I really love your articles at i09. I think I was a little too harsh in my criticism, but that probably came out of the fact that I generally love pretty much everything you write on i09 (especially when it’s on cognitive biases, which is my personal specialty). I think you do a great job at bring skeptical viewpoints to a big audience that isn’t necessarily getting it from their other pop culture reads.

  19. Articles like this on “scientific analysis” in blogs by non-experts is dangerous. With all due respect, the author of this article doesn’t even have a PhD in statistics, has no medical degrees, and is certainly not an expert or “guru”. Just because someone may have taken a few courses and is a “stats junkie”, does not make them an expert; and it shows with the numerous math and factual errors pointed out by many.

    Scientific analysis should be done in the context of peer reviewed research, not on a blog post; especially if the post accuses another professionals of making mistakes or manipulating data. It is scary to think that anyone, regardless of their qualifications or level of expertise, can post on a blog accusing other professionals of mistakes or wrongdoing; without peer review. I feel like qualified professionals now have to waste their time debating amateurs in comment threads. This is not how we want science to be done.

    If you are going to post on a particular scientific analysis, make sure you get qualified professionals who only discuss peer reviewed research.

  20. Jamie:
    “Prof. Orosz is assuming that in all cases in which outcomes were missing for mothers who transferred to a hospital with a breech birth, the newborn died, though she gives no evidence for why this may be true.”

    MANA:
    “There are 4 singleton pregnancies, 3 of which were breech presentations, for which all birth outcomes data are unavailable. These women began labor at home and then transferred to the hospital prior to birth. The midwives of record were contacted, and in each case the midwife did not accompany the mother, nor did the mother return to the midwife for postpartum care.”

    Leaving aside that she made no such assumption, as others have pointed out, you may not understand why she thought it was a reasonable possibility. A reason for transfer in the case of a breech birth that is so embarrassing that the midwife will not accompany the labouring woman to the hospital is that the baby’s body is dangling between the mother’s legs and they can’t get the head out. In this circumstance the baby has died of suffocation long before reaching the hospital. Midwives and pregnant women are typically very close and the midwife does postpartum follow-up with both mother and baby. A reason for the family not to want to talk to the midwife after the birth of their child, or for no postpartum care to be done, is that the baby died. Out of all 16,575 vertex presentations, in only one case did the midwife claim not to know whether or not the baby lived or died. Out of 222 breech presentations, four claimed not to know. One thing we do know is that breech presentation is notoriously lethal and that attempting do deliver a breech baby at home is an all-round dangerous idea. It is not at all unreasonable to speculate that

  21. An analogy for a point from Dr. Orosz’s analysis, that I feel Ms. Bernstein and some of the commenters are missing concerning mortality rates based on a comparison of the homebirth group in the hospital birth group.

    There is a trip that needs to be made and one person believes that it takes an hour to drive and another believes that it takes a half an hour to drive. Let’s assume that there are no shortcuts and that there are five steps to the trip to this location. The five steps are reviewed and it is agreed that each step takes at most five minutes to drive. We can then conclude that the overall trip should in no way take more than 30 minutes to drive … much less then the hour that one person thought it would take.

    In effect this is what Dr. Orosz has done in her analysis of the MANA paper. At every possible opportunity when selecting criteria for the CDC hospital birth group, Dr. Orosz opted to choose variables that would make mortality rates for the homebirth group look more favorable compared to the hospital groups rates. This is often done in studies, when the comparison groups are likely not to be really well matched. In effect giving an extreme benefit of the doubt to the study group, in this case the homebirth group. Dr. Orosz conclusions maybe off, but will only be likely to under estimate the difference between homebirth and hospital birth mortality rates, not overestimate them. Ms. Bernstein’s critique is that the criteria chosen do not match well the patient population of the MANA study. However, what she fails to note is that the differences all favor over estimating the mortality rate of the hospital group relative to the homebirth group and yet the homebirth group still has mortality rates much higher than the hospital group. I am having a hard time understanding why this is such a difficult concept to grasp.

  22. I have never been to this site. Followed it from SOB and quite interesting actually.
    I wanted to share because as a CNM who has been intimately involved for three years in Portland Oregon and more recently from my home in the south the issue at hand is incredibly complex in many ways. However, in others quite simple.

    I wrote a paper for the College, American College of Nurse Midwives, it was quite extensive 150 pages addressing midwifery; Midwifery in The United States Verification of Facts and Data. It was debated by the elders at the National Level without my permission or inclusion. Obviously things were exposed that probably did not want to be exposed by the elders of my College.

    Things from there get brutally abusive and I end up getting more politically active. Hence, OSHA, Joint Commission, ACOG and the perinatal credentials nationally being implemented. And now The World Health Organization is involved. The bottom-line is WE MUST CHANGE. Midwifery cannot exist anymore broken as we have been for the past 40-50 years. We must take a stand and address difficult issues within our community in order to protect the public and it is a very difficult task to tackle. But if we do not do this with complete honesty, we are in essence, HOSED!

    Now about the paper. One cannot critique such data in my opinion for numerous reasons.
    First, homebirth and the lack of regulation from state to state means the data is irrelevant. Furthermore, it certainly cannot be compared to a hospital setting with any congruence.
    Second, the data is from a database that is voluntary. And majority of the data is entered retrospectively, hence, rendering the data full of bias.
    Third, majority of states that self collect and bypass the MANASTATS PROJECT prove that the data is absolutely inaccurate. The College knows this because I wrote about this in my paper that was debated.
    Fourth and most important to me: I LIVED THE HORRIFFIC ABUSE OF HOMEBIRTH AND MORTALITY OF NEONATES. PTSD would be the word to describe most of in hospital providers from the havoc. It was absolutely horrible that is why I became so heavily involved. This is the opening in my paper.

    If anyone would like a copy of my paper contact me and I will absolutely share. We must change things and it must begin with HONESTY WITHOUT FEAR.

  23. I’ll begin by quoting Ms. Bernstein: “There are so many things wrong…that it’s hard to even know where to begin.”

    1) So, how do I know that i09 did not bother to read the study or the Skeptical OB post? Their title was ‘Statistics Professor Challenges Midwives’ Math on Home Birth Safety.’ There are so many things wrong just in this title that it’s hard to even know where to begin. First of all, let’s take the term ‘midwives’ math.’”

    This complaint is ridiculous. Leaving aside the fact that several of the authors were, in fact, midwives, it’s incredibly silly to criticize someone for referring to a professional organization that represents midwives as “midwives.” If a trade association representing cattle ranchers published a paper that was criticized by a statistics professor, it would be completely reasonable to write a headline that said, “Statistics Professor Takes Issue with Ranchers’ Claims,” even if the people who actually wrote the report work in an office and not a ranch.

    2) “Note: I’m not linking to the authors’ bios for their privacy and protection.”

    More silliness. No one has a reasonable expectation of privacy in their publicly-accessible bio. We’re not talking about their okcupid profile, for Pete’s sake. And your use of the word “protection” is ridiculous. First of all, your high-minded refusal to provide a simple link wouldn’t prevent anyone from finding this information anyway with a simple google search. Second, you seem to imply that they are in need of protection, as if someone is going to harm them physically for having written a report on home birth. Do you think you’re putting Tuteur at risk by mentioning her name? Of course not.

    3) “The Journal of Midwifery and Women’s Health study did not attempt to make any relative risk comparisons between homebirths and hospital births.”

    That may technically be true, but there’s no doubt that they aggressively promoted this study as showing that homebirth is “safe.” What possible meaning could that have besides “as safe as hospital birth?” The fact that they didn’t choose to draw the obvious (and damning) comparison between homebirth and hospital birth doesn’t make it wrong for others to do that.

    4) “Statistics Professor Writes a Blog Post that Builds on Math Regarding Homebirth Risks from a Peer Reviewed Study in a Well-Respected Journal by Researchers from Oregon State University.”

    Not only is that headline far too long, it also needlessly refers to the Journal of Midwifery and Women’s Health as “well-respected.” What makes Ms. Bernstein think that this journal is “well-respected?” By whom is it respected? Is there any evidence for that statement? Or does the mere fact that a journal exists make it “well-respected?”

    5) “(Note: quotes are because I think it’s weird that Prof. Orosz keeps being referred to as “a statistics professor” instead of by her name, not because I’m doubting that she’s really a statistics professor.”

    What’s weird about it? With all due respect to Professor Orosz, she isn’t exactly a household name. What’s relevant is the fact that she’s a statistics professor.

    6) “I want to know how problematic it may be that the hospital cohort defines “hospital births” as any birth taking place in a hospital while the homebirth cohort defines “homebirths” as a birth that was planned to be at home regardless of where it actually took place. In fact, over 10% of the homebirth cohort gave birth in a hospital. How might this affect the comparison?”

    Others have already pointed this out, but it’s really not that difficult to figure out how this might affect the comparison: it would substantially harm the stats for hospital birth and help those for homebirth. Bernstein is taking an entire category of births that can logically be assumed to be far higher risk (assuming that a substantial percentage of women who transfer to the hospital in the middle of a homebirth do so precisely because there is a complication) and throwing it in the hospital group.

    7) “We currently do not have a comparable hospital birth cohort to use to calculate relative risk between homebirths and hospital births.”

    I’m still having a hard time figuring out why this is so. Is there a chance the two groups differ in some way besides the relevant variable (choice of birth location)? Sure. But that’s always true in any study where we’re comparing two groups of people that have made different choices (i.e. who weren’t randomly assigned a course of treatment). Does that render any comparison between the two a haphazard exercise in “back-of-the-envelope calculations?” I don’t see any reason for why it does.

    8) “In this case the cohort groups are very different.”

    To the extent that this is true, it only contradicts the claim that homebirth is safe relative to hospital birth. Everything we know about women who choose homebirths suggests that they, as a group, are likely to be healthier and more educated than mothers who give birth in hospitals.

    9) “Therefore, there may be increased risk of neonatal and intrapartum death associated with homebirths although further research using a comparable hospital cohort will have to confirm.”

    I don’t see how it is possible to find a “comparable hospital cohort” that would meet Ms. Bernstein’s criteria, since she argues that women who choose homebirth simply can’t be compared to women who choose hospital birth.

    10) “In other words, Prof. Orosz is assuming that in all cases in which outcomes were missing for mothers who transferred to a hospital with a breech birth, the newborn died, though she gives no evidence for why this may be true.”

    No, that’s not even close to what she’s doing. The operative terms that Orosz uses here are “if” and “would.”

    It’s really beyond galling that Bernstein would have the nerve to accuse others of not doing their reading.

    By the way, one quick suggestion for the author: in both of your posts you’ve made errors by simply missing a link, a quote, or a citation. It might not be a bad idea to send an advance copy of any future writings to Dr. Tuteur. Obviously you wouldn’t and shouldn’t give her the right to edit what you’re saying, but it might help you avoid making these kinds of very clear mistakes.

  24. I can’t reply to the comment above, unfortunately. 🙁

    Alli said “I’ve been trying to think of how to explain why I didn’t and still don’t find Dr. Tuteur’s comparison misleading, and I think I finally put my finger on it–it’s really a comparison of home versus hospital *midwifery* care.” Yes yes YES. Thank you for spelling it out better than I did.

    Women choosing to have a high risk home birth is one thing…. but women unknowingly having high risk home births is another… women who *think* they are low risk. As I have said before, I can’t help but wonder if the pre-e and GDM moms in the MANA study knew ahead of time of their issues… specifically the pre-e and GDM moms who ended up losing their babies due to their high risk issues. Did they know? Were they screened? Did their midwives just miss or ignore or dismiss the red flags?

    I watched on youtube a breech home birth… an ACCIDENTAL breech home birth. I was FLOORED! The mother had no clue and when the midwife found out the baby was breech, there was no discussion of risks, there was no talk of transfer…. nothing. Just the discovery of breech baby and then kept right at it. So much for informed consent! The shocking part is that it is really not shocking anymore. B/c THAT is home birth in our country. It is a lack of informed consent, lack of screening, lack of monitoring, lack of acknowledging the real risks in high risk situations.

    Until there are changes made to home birth in our country to ensure that it is for TRULY low risk women only, then a woman has no idea what kind of midwife she is getting, no idea if she’ll be getting informed consent, no idea if she will be getting the truth about the importance of screening measures, no idea if she’ll be properly monitored, etc etc etc.

    I wish I knew how many women have home births each year who think they are low risk but are actually high risk.

Leave a Reply