Thursday, May 11, 2006

The challenge: How do we decide?

This post is reserved for a discussion of a specific issue. What would homebirth advocates need to see in order to believe that homebirth is not as safe as hospital birth? The answers should be restricted to science and statistics. This will be an insult and accusation free zone. No posts about " all obstetricians" or "you're twisting things".

Is anyone up to the challenge of explaining the scientific and statistical criteria that it would take to convince homebirth advocates of anything that they do not believe in advance? I'm willing to bet this is going to be a very short thread. I suspect that there really isn't any evidence that would be convincing, because homebirth advocacy is based in large part on personal beliefs and personal beliefs are not swayed by science.

111 Old Comments:

By the way, I am going to take the unusual step of deleting any comments that are simply personal insults or accusations. I have already deleted one.

By Blogger Amy Tuteur, MD, at 9:13 PM  

clean up birth and death records
since the inception of record keeping midwives and "others"/unintentional or intentional unattended births have been clumped together and has skewed that information as far as I can tell it is true in every state- and was true and commented on even in the 1920's stats -
I am thinking that there are some other pieces of info that need to be changed/standardized so that a national accurate as possible data base is available
because of the legal standing of midwives differ in each state- this alone can obscure accurate information.

By Anonymous Anonymous, at 12:49 AM  

Anonymous:

"clean up birth and death records
since the inception of record keeping midwives and "others"/ unintentional or intentional unattended births have been clumped together and has skewed that information"

Okay, that might improved the quality of the data.

So if I could show you studies where unintentional homebirths were excluded and mortality rate at home was higher, would you accept that homebirth is not as safe as hospital birth?

By Blogger Amy Tuteur, MD, at 6:40 AM  

Training.

I think part of the problem might be explained like this:

Designing accurate, unbiased studies, that will produce reliable and repeatable results, is HARD. Running studies is, if anything, even HARDER. People spend years just learning how to run a good study! I can't think of how many studies I've been a part of in some way and there are always people who are far better than you.

And incidentally, merely having a degree doesn't mean you're good at it. I don't have an M.D., but I'm better than many M.D.s as reading data and drawing accurate conclusions from the data.

Anyway, midwife training doesn't include "how to run a study" or "how to design a study" classes. (And why would it? You don't need those classes to deliver a baby!)

So you end up with a class of midwives who are extremely intelligent, but not trained to recognize good studies from bad studies, or to perform them.

And since scientists rely on "good" studies to prove their point and compare to "bad" studies: If you don't have training, you can't tell good from bad. And if you can't tell the difference, you can't be convinced.

If everyone involved took a good statistics class or two, this would be a lot easier.

By Blogger sailorman, at 9:40 AM  

"So if I could show you studies where unintentional homebirths were excluded and mortality rate at home was higher, would you accept that homebirth is not as safe as hospital birth?"



Every study that has compared midwives and obstetricians has found better outcomes for midwives for same-risk patients. In some studies, midwives actually served higher risk populations than the physicians and still obtained lower mortalities and morbidities. The superiority and safety of midwifery for most women no longer needs to be proven. It has been well established. (Madrona, Lewis & Morgaine, The Future of Midwifery in the United States, NAPSAC News, Fall-Winter, 1993, p.30)

In The five European countries with the lowest infant mortality rates, midwives preside at more than 70 percent of all births. More than half of all Dutch babies are born at home with midwives in attendance, and Holland's maternal and infant mortality rates are far lower than in the United States... ("Midwives Still Hassled by Medical Establishment," Caroline Hall Otis, Utne Reader, Nov./Dec. 1990, pp. 32-34)

In the U.S. the national infant mortality rate was 8.9 deaths per 1,000 live births [in 1991]. The worst state was Delaware at 11.8, with the District of Columbia even worse at 21.0. The best state was Vermont, with only 5.8. Vermont also has one of the highest rates of home birth in the country as well as a larger portion of midwife-attended births than most states. (Stewart, David, International Infant Mortality Rates--U.S. in 22nd Place, NAPSAC News, Fall- Winter, 1993, p.36)

The international standing of the U.S. [in terms of infant mortality rates] did not really begin to fall until the mid-1950s. This correlates perfectly with the founding of the American College of Obstetricians and Gynecologist (ACOG) in 1951. ACOG is a trade union representing the financial and professional interests of obstetricians who has sought to secure a monopoly in pregnancy and childbirth services. Prior to ACOG, the U.S. always ranked in 10th place or better. Since the mid-1950s the U.S. has consistently ranked below 12th place and hasn't been above 16th place since 1975. The relative standing of the U.S. continues to decline even to the present. (Stewart, David, International Infant Mortality Rates--U.S. in 22nd Place, NAPSAC News, Fall-Winter, 1993, p.38)



I'm not a homebirth advocate myself, and I'm trying to be open to your argument, but I just can't find the support for your claim that homebirth is more dangerous than hospital birth.

One (unnecessary) study showed that high-risk home deliveries result in higher mortalities; is this what you mean?

By Anonymous Anonymous, at 12:52 PM  

I have thought about this overnight and am not sure I can answer it because "safer" is too broad and undefined. Safer for whom, safer how? Safer in terms of whether the baby is alive at one day/ one month/ one year? Quality of life short and/or longterm? Are we only concerned about survival at birth? Or are we concerned about preventing child abuse with optimal mother/baby bonding, promoting breastfeeding to save however many lives a year from illness or from diseases of adulthood that are caused by feeding artificial baby milk? (Because aren't home birthers more likely to nurse and for longer terms?) If you can give me a more narrow definition of safer, I can give it a try.

MM, Georgia

By Anonymous Anonymous, at 12:59 PM  

Anonymous:

None of the references that you provided are studies. They are opinion pieces by homebirth and midwifery advocates. That's the problem. The claims in the opinion pieces are not supported by actual studies. Usually, the claims in opinion pieces leave out very important information.

Take the case of Vermont, for example. It is extremely unfortunate, but race is a risk factor for neonatal death. Vermont may have a higher than average proportion of home births, but more importantly, it has among the lowest if not the lowest proportion of African American residents. That is why the neonatal death rate is so low. If you don't believe me, just check the neonatal death rates for white women in other states and you will see that they are the same as Vermont.

By Blogger Amy Tuteur, MD, at 3:11 PM  

I couldn't resists. So I wrote a blog post called

"How to design a Homebirth Study that Scientists Will Believe".

It's here:
http://moderatelyinsane.blogspot.com/2006/05/designg-tests-for-homebirth-safety.html

By Blogger sailorman, at 3:26 PM  

Take the case of Vermont, for example. It is extremely unfortunate, but race is a risk factor for neonatal death. Vermont may have a higher than average proportion of home births, but more importantly, it has among the lowest if not the lowest proportion of African American residents.

What do you think affects outcomes more: use of prenatal care, or race? Women who have home births are receiving prenatal care from their midwives; if their outcomes are statistically better, then that proves my point.

Home births are safe.

By Anonymous Anonymous, at 3:38 PM  

The truth is that race seems to be a stronger factor than prenatal care, regarding neonatal outcomes. Black women receiving prenatal care still have increased neonatal death rates over white women.

However, this disparity does nothing to prove that homebirth is unsafe, in Vermont or elsewhere.

By Anonymous Anonymous, at 3:50 PM  

"What do you think affects outcomes more: use of prenatal care, or race? Women who have home births are receiving prenatal care from their midwives; if their outcomes are statistically better, then that proves my point.

Home births are safe.
"

No, it doesn't prove your point. This is a faulty conclusion

It MAY BE that midwives provide better prenatal care--I personally believe that may be true. It MAY EVEN BE that the prenatal care is so much better that the overall neonatal rate is better.

This says NOTHING about whether the same midwives, delievering the same babies, with the same prenatal care in a hospital would have better rates.

You should really read my blog posts if you dno't understand this.

By Blogger sailorman, at 4:37 PM  

People are getting away from the topic.

I asked a very specific question. What would you need to see in terms of statistics in order to acknowledge that homebirth is not as safe as hospital birth.

The answer to the question would go something like this:

If I saw the following statistics ... then I would think ...

Once you tell me the TYPE of statistics that you would believe, I will try to find them.

Please do not respond by citing irrelevant statistics. I'm not talking about what you think are the statistics (that discussion is being carried on in several different threads), but what type of statistics would convince you IF THEY EXISTED.

Thusfar, of the homebirth advocates, only MM has responded directly to the question. As I understand the response, she would like to include morbidity statistics and other statistics in order to arrive as a decision about whether homebirth is as safe as hospital birth.

Thank you, MM, for taking the time to think about it and respond. Now I'd like to ask you an even more specific question. What type of data would you need to see to be convinced that the neonatal mortality rate associated with homebirth is higher than neonatal mortality rate associated with hospital birh?

By Blogger Amy Tuteur, MD, at 4:48 PM  

You know, I'm not locked into the idea that homebirth is going to be as safe as hospital birth forever and ever amen. Things could change. Hospitals could do a lot to improve services for laboring women. There could be a significant advance in perinatal care that's not portable. It doesn't seem all that likely to me, but it's not implausible. And let me say again: I am only an interested mother and not a childbirth professional, so I don't know if my opinion is useful or interesting.

But I guess I'd be looking for a large study with input in the planning stages from both sides -- researchers who prepared by talking to both OBs opposed to homebirth and midwives attending homebirths, to address both sets of concerns about data collection. I'd want more information on the issue of congenital anomalies. Is Cawthon's speculation correct, that homebirthing mothers are more likely to continue a pregnancy in that situation? I'd want to study credentialed midwives, practicing legally.

If a study could meet those criteria and show a statistically significant difference (p=.01, let's say) in perinatal death rates for low-risk women with vertex babies, I'd re-evaluate my position.

While I'm waving my Epidemiology Magic Wand, I'd also love to see follow-up one and two years down the line. How's the correlation btw planned breastfeeding duration and actual breastfeeding duration? How's satisfaction with the maternal role? How's the child's cognitive and neuromotor development? (Those things all have sticky confounding factors, but that's what magic wands are for.)

By Blogger Jamie, at 5:30 PM  

Hmmmm. Interesting question. I will concede your point that homebirth advocacy is based in large part on personal beliefs. I would go one step further and say that it is based on personal beliefs AND personal experiences. I am a "homebirth advocate" and my belief in the ideal of homebirth is a combination of deeply held personal beliefs and my reading of scientific literature. I could not personally disentangle the two. I am particularly biased after experiencing two very bad hospital births and one very pleasant home birth.

That being said, I will take a stab at your challenge. In order to change my mind, for similar populations I would like to see #1 lower rates of infection... both mother and baby; #2 Lower c-section rates; #3 Lower rates of postpartum depression; #4 Higher rates of breastfeeding. Those are some biggies that come to mind.

As far as scientific studies, I would need to see the "gold standard" double-blind randomized trial in which women and babies fared better in the hospital than at home. But, that will never happen. Impossible and unethical.

I must say though, that your contention that "personal beliefs are not swayed by science" could equally be applied to the obstetrical community in many instances. Old practices die hard! But, perhaps that should be a new thread!

By Blogger Mama Liberty, at 5:49 PM  

Thanks for the response. It's still a complicated question but this narrows it to more manageable. I'm struggling with the apples/oranges aspects of in the US as there's both the caregiver factor, with the standard of care for the DEMs doing most of the home birth being different from the standard of care for hospital based CNMs and OBs; and then the location factor. It certainly would be satisfying to design a study where we had OBs/CNMs provide their customary prenatal care and then attend randomly seleted women at home and compare those to matched DEM home birth outcomes; and then to move some DEM moms into the hospital and allow the DEMs to function completely independently with the only difference being that tranfer of care at complications would be much faster.

How about: show in a first world country where they book women for home births within the same health system as hospital births, that neonatal mortality for planned home births, including completed home births and births where moms transfer AFTER a typical time for hospital admission for planned hospital birth in that culture (after onset of active labor, whatever the standard of care is in that culture for moving labor into the hospital in planned hospital birth), is higher than neonatal mortality for planned hospital births completing in the hospital; exclude unattended home/car/etc births b/c I don't think reporting is reliable and other reasons. Match moms up in the two groups as closely as possible for risk factors, parity, baby's size, socioeconomic status, race, gestation length, etc; decent size groups.

Morbidity risk for both myself and my offspring was a concern of mine in planning home births.

MM, Georgia

By Anonymous Anonymous, at 6:21 PM  

"Morbidity risk for both myself and my offspring was a concern of mine in planning home births."

I agree, Georgia! I will go one further... as someone planning a large family, I was also concerned with morbidity risk for my future children. Should I end up with a c-section, the risk of future placental problems and stillbirth are increased.

I've noticed in observing political debates about home birth that the physicians commonly insinuate that home birthing mothers care more about their own health and experience than that of the baby. I find the idea that a doctor who sees me for a series of six minute visits a handful of times is more concerned about the health of my precious baby than I am. Mother and baby are not competing entities. I feel that midwives understand that.

By Blogger Mama Liberty, at 8:22 PM  

Oops. I suppose Georgia might not be your name but your location! Tee hee.

By Blogger Mama Liberty, at 8:23 PM  

Jamie:

"I guess I'd be looking for a large study with input in the planning stages from both sides -- researchers who prepared by talking to both OBs opposed to homebirth and midwives attending homebirths, to address both sets of concerns about data collection."

I agree. Although I'm not sure the participation of both sides is necessary as long as the concerns of both sides are met.

I'm not aware of such a study done in the US, but there may be studies from other countries that meet your criteria. I'll take a look.

"How's the correlation btw planned breastfeeding duration and actual breastfeeding duration? How's satisfaction with the maternal role? How's the child's cognitive and neuromotor development?"

It would be great if such a study could include follow up, especially about the child's development. I'm not sure how valuable information about breastfeeding and satisfaction with maternal role would be since those are so dependent on the mother's pre-existing psychological issues.

For example, since women know that they are "supposed" to be breastfeeding because it is better for the baby, they often will not acknowledge that they stopped because they didn't care to make the sacrifices required. Yet when you ask them about it, they have all sorts of creative excuses. I would be willing to be that the factor that is most predictive of breastfeeding success is maternal resolve. Those who really intend to breastfeed almost always do so successfully and for long durations.

By Blogger Amy Tuteur, MD, at 9:07 AM  

Anonymous:

"show in a first world country where they book women for home births within the same health system as hospital births, that neonatal mortality for planned home births, including completed home births and births where moms transfer AFTER a typical time for hospital admission for planned hospital birth in that culture"

There are no American studies like this, but I believe there are several studies from Great Britain and Australia that might meet your criteria. I will look and let everyone know.

By Blogger Amy Tuteur, MD, at 9:11 AM  

It seems to me that we have the best and broadest study right before our eyes. In the developed world, the top ten countries that have the lowest infant mortality rate all use the midwife model of care and birth a large percentage of their babies at home or in birth centers. As you go down the list the percentage of infant deaths goes up as the percentage of homebirths go down.
It's not a carefully organized, double-blind experiment, but it covers hundreds of millions of women; a feat no study could beat.
How can we argue against that? Is the data for millions and millions of births just wrong? That certainly doesn't make sense.

By Anonymous Anonymous, at 1:13 PM  

Anonymous:

"In the developed world, the top ten countries that have the lowest infant mortality rate all use the midwife model of care and birth a large percentage of their babies at home or in birth centers. As you go down the list the percentage of infant deaths goes up as the percentage of homebirths go down."

Remember we are not looking at infant mortality (1 month to 1 year), we are looking at neonatal mortality (from labor through one month).

The reasons that the US lags in neonatal mortality has absolutely nothing to do with midwives or birthing centers or homebirth.

If you look at the statistics for white women, they are comparable to all the other countries. The US has the highest proportion of black women of all the countries with the best neonatal mortality. The African American statistics are much worse (more than double the same statistics for white babies) and therefore increase the overall US neonatal mortality rates. If the other countries had higher populations of black women, their statistics would be very different.

So, for example, the neonatal mortality rate in 2000 for babies born to white women was 3.8/1000 while the neonatal mortality rate for babies born to black women was 9.6/1000.

By Blogger Amy Tuteur, MD, at 3:13 PM  

How about this paper from Australia:

Perinatal death associated with planned home birth in Australia: population based study
BMJ 1998;317:384-388 ( 8 August )

Abstract

Objective: To assess the risk of perinatal death in planned home births in Australia.
Design: Comparison of data on planned home births during 1985-90, notified to Homebirth Australia, with national data on perinatal deaths and outcomes of home births internationally.
Results: 50 perinatal deaths occurred in 7002 planned home births in Australia during 1985-90: 7.1 per 1000 (95% confidence interval 5.2 to 9.1) according to Australian definitions and 6.4 per 1000 (4.6 to 8.3) according to World Health Organisation definitions. The perinatal death rate in infants weighing more than 2500 g was higher than the national average (5.7 versus 3.6 per 1000: relative risk 1.6; 1.1 to 2.4) as were intrapartum deaths not due to malformations or immaturity (2.7 versus 0.9 per 1000: 3.0; 1.9 to 4.8). More than half (52%) of the deaths were associated with intrapartum asphyxia.
Conclusions: Australian home births carried a high death rate compared with both all Australian births and home births elsewhere. The two largest contributors to the excess mortality were underestimation of the risks associated with post-term birth, twin pregnancy and breech presentation, and a lack of response to fetal distress.

By Blogger Amy Tuteur, MD, at 4:02 PM  

" If the other countries had higher populations of black women, their statistics would be very different."

I find this to be racist. I think you mean higher populations of POOR black women. The reasons that the statistics for African Americans are so much worse is because of poverty, not race.

Anyway, if you look at the statistics for the Netherlands, with a very high rate of home births (33% I think), they also have a disproportionately high poor immigrant population, yet they still seem to do much better than the U.S. I would also point you to Taos County, N.M., with about a 25% (or higher) out-of-hospital birth rate. The midwives in Taos County have excellent statistics even though they also take care of a large number of poor immigrants.

By Blogger Mama Liberty, at 4:15 PM  

I apologize that I do not have the time, nor the expertise of Amy or Sailorman to dissect studies down to minutia, but here is a list of references to studies, discussions, opinions, etc.: References on Home Birth. Perhaps more fodder for discussion...

By Anonymous MetroMidwife, at 4:44 PM  

As far as I can tell from a couple of minutes' googling, Australia is/was not a system where they book women for home births within the same health system as hospital births.

I haven't read this study, and have no idea if it's any good, but in my opinion it would be booking in the same health system (since the home births were all with US CNMs).

MM, Georgia

Murphy PA. Fullerton J.
[Department of Obstetrics and Gynecology, Columbia University College of Physicians and Surgeons, New York, New York 10032, USA. pam15@columbia.edu ]
Outcomes of intended home births in nurse-midwifery practice: a prospective descriptive study.
Obstetrics & Gynecology. 92(3):461-70, 1998 Sep.
OBJECTIVE: To describe the outcomes of intended home birth in the practices of certified nurse-midwives. METHODS: Twenty-nine US nurse-midwifery practices were recruited for the study in 1994. Women presenting for intended home birth in these practices were enrolled in the study from late 1994 to late 1995. Outcomes for all enrolled women were ascertained. Validity and reliability of submitted data were established. RESULTS: Of 1404 enrolled women intending home births, 6% miscarried, terminated the pregnancy or changed plans. Another 7.4% became ineligible for home birth prior to the onset of labor at term due to the development of perinatal problems and were referred for planned hospital birth. Of those women beginning labor with the intention of delivering at home, 102 (8.3%) were transferred to the hospital during labor. Ten mothers (0.8%) were transferred to the hospital after delivery, and 14 infants (1.1%) were transferred after birth. Overall intrapartal fetal and neonatal mortality for women beginning labor with the intention of delivering at home was 2.5 per 1000. For women actually delivering at home, intrapartal fetal and neonatal mortality was 1.8 per 1000. CONCLUSION: Home birth can be accomplished with good outcomes under the care of qualified practitioners and within a system that facilitates transfer to hospital care when necessary. Intrapartal mortality during intended home birth is concentrated in postdates pregnancies with evidence of meconium passage.

By Anonymous Anonymous, at 5:23 PM  

Amy,

I have a challenge for you. A good debater must be able to take either side. Please analyze the Australia study, from the perspective of a "homebirth advocate" with the same fervor and critical eye as you did for the recent BMJ study.

This might be fun!

Thanks, Danielle

By Blogger Mama Liberty, at 5:24 PM  

"I find this to be racist. I think you mean higher populations of POOR black women. The reasons that the statistics for African Americans are so much worse is because of poverty, not race."

And I find your comment to be insulting. The fact is that it appears that being black is a risk factor in and of itself. There are plenty of poor white people and plenty of poor Hispanic people, but they do not have such high perinatal mortality.

There are gender differences and racial differences in health. For example, for many years cardiologists and cardiac surgeons assumed that whatever worked for men would necessarily work for women. Eventually, they realized that though the cardiac mortality rate was dropping for men, it was not dropping for women. Nowadays, if a cardiologist did not consider the gender of the patient in formulating a plan of care, it would be considered negligent.

Not all black women are poor. Wealthy, well educated black women who have prenatal care have significantly higher perinatal mortality than comparable white or Hispanic women. That suggests to me that we are ignoring somthing very important in caring for them. Perhaps they have additional risk factors we don't understand or they react differently to certain standard treatments. There is no doubt that they have a baseline level of hypertension that is much higher than in the white and Hispanic population and that can be a risk factor for poor pregnancy outcomes.

It is all too easy to ascribe the poor outcomes for black infants to economic or social factors. That relieves us of the burden of looking at the way we practice medicine and essentially placing the blame on them. I don't know what the cause is, but I feel very strongly that we should make a large financial commitment to researching this issue and find out how we can improve.

By Blogger Amy Tuteur, MD, at 6:03 PM  

MM, regarding the Fullerton study:

The study subtracts deaths from the homebirth group in order to reach its conclusions. By the rules of statistics (check a statistics website), they are not allowed to do that. Here are some more specific criticisms:

1. The first problem is that the results of the study are self reports by midwives. The authors of the study appear to have made extensive efforts to corroborate the midwives reports. Nonetheless, they could not obtain hospital charts for 20% of the women who were transferred in labor to the hospital.

2. Of 1221 beginning labor with the intention of delivering at home, 5 babies died for a neonatal death rate of 4.1/1000. According to the midwives report, 4 of the 5 deaths were stillbirths. Again, according to midwife personal report, 2 of the stillbirths occurred in labor but before the midwife got there; this is where the self-reporting becomes problematic. Did those deaths really occur in labor, but before the midwife got there, or did the midwife choose to protect herself by claiming that she had nothing to do with the death?

The authors of the study choose to remove these two stillbirths from their calculations. I imagine that there reasoning was that since it couldn't be ascribed to the midwives (if their self reports were true), they couldn't be ascribed to homebirth. They arrived at the quoted neonatality rate only because they subtracted these deaths. The actual neonatal mortality rate was 4.1/1000. Removing these deaths is inappropriate. They occurred in women intending to deliver at home and laboring at home.

Leaving aside the first 2 deaths for a moment, there were 2 other intrapartum deaths (stillbirths). Both occurred in women who were greater than 42 weeks gestation and meconium was present during labor. These were entirely preventable deaths.

The final death occured within the first 24 hours and the authors provide no information on the cause.

The 4 intrapartum deaths are hardly a ringing endorsement of homebirth, since the intrapartum death rate in a hospital setting is vanishingly small.

3. A curious, but unexamined finding is that of the 97 women transferred to the hospital prior to the onset of labor, 10 had IUFDs (intra-uterine fetal demise). Since there were 1318 women enrolled in the study, the pre-labor stillbirth rate was 7.6/1000. The baseline IUFD rate (for all women, not just low risk) is 5.1/1000, this is pretty remarkable. It is 50% greater than the population as a whole. It raises the possibility the the prenatal care in the midwifery group was substandard.

So, the bottom line is:

The actual neonatal mortality rate was 4.1/1000, almost twice as high as what the authors claim.

At least 2 of the 5 deaths were preventable and could have been avoided by following standard obstetric recommendations.

The pre-birth stillbirth rate of the midwife group was 50% higher than expected, raising serious questions about the prenatal care offered by the midwives in this study.

By Blogger Amy Tuteur, MD, at 6:15 PM  

There are physiological differences between blacks and whites, you know. Vitamin D production, for example. Impact on pregnancy?

http://www.medscape.com/viewarticle/530908

MM, Georgia, USA

By Anonymous Anonymous, at 6:24 PM  

Oh, Dr. Amy, when I saw you mention Australia earlier I was going to say, "Not that Bastian study, please!" I agree with you about breech babies and twins being safer in the hospital. I agree with you about babies less than 37 or more than 42 weeks being safer in the hospital. In my very first comment at Neonatal Doc's I said that I have a healthy respect for what meconium can do to a baby's lungs, as a result of sitting at my oldest son's bedside while he was recovering from meconium aspiration. (He was born in the hospital, with continuous monitoring, FWIW.)

This study is an indictment of some Australian midwives' standards of care (not listening to FHTs during second stage? disregarding fetal distress? yikes!). Better standards of care lead to better outcomes.

Still curious about how much you think the mortality risk rises with homebirth for a low-risk woman with a singleton vertex baby at term.

By Blogger Jamie, at 9:54 PM  

Dr. Amy says: I would be willing to be that the factor that is most predictive of breastfeeding success is maternal resolve. Those who really intend to breastfeed almost always do so successfully and for long durations.

Careful there. Motivation and breastfeeding duration are certainly correlated, but there are a lot of factors at work. I have seen women really saddened by the premature end of a breastfeeding relationship, and they were even more upset by people who assumed they must not have worked hard enough to figure things out.

If you get the chance to see Linda Smith speak, do.

By Blogger Jamie, at 10:11 PM  

I didn't see you respond to some of Maribeth's points about the Fullerton study over at Neonatal Doc's, Dr. Amy. Would you do so here? I haven't been able to find a clear online distinction btw IUFD and intrapartum death. Suppose a woman planning a hospital birth shows up and says, "I've been contracting all day and I felt the baby kicking an hour ago," but when she's hooked up to the monitor there's no heartbeat. IUFD or intrapartum death?

What about Maribeth's observation that the women she serves are more vulnerable to lethal genetic conditions? What about my observation that I know a number of homebirthing women who would carry to term after a lethal anomaly diagnosis (not hard data, but a consideration nonetheless)? Homebirthing women are also more likely to decline routine ultrasound screening, which might mean that they were unaware of a lethal anomaly.

There are plausible explanations on the table for the increased IUFD rate. Is it really necessary to cast aspersions on midwives' prenatal care?

By Blogger Jamie, at 10:26 PM  

to comment about self reporting- that isn't exactly true- they were enrolled first of all so had to report the records, of everyone enrolled . If they were going to fake records wouldn't they have just not included any deaths? Thin about it--
how do you have proof if a mom comes to your office and she has a a demise? you are just self reporting-- then she goes to the hospital -- what if your client refused to go to the hospital, and wants to wait until spontaneous labor starts.

By Anonymous Anonymous, at 11:09 PM  

Amy writes: Wealthy, well educated black women who have prenatal care have significantly higher perinatal mortality than comparable white or Hispanic women

Where is the evidence for this Amy? I'm not disagreeing with it, I just didn't know that (controlled for hypertension, of course)

Jamie, that's a good point about lower rates of prenatal testing contributing to a potentially higher neonatal mortality rate. Amy, what would you say is the percentage of women who get a Quad screen in a modern OB practice? My practice percentage rate is about 1-2%. And we have had lethal trisomies. I bet the abortion / termination rate is significantly lower for women planning homebirths.

MM, do read the thread at Neonatal Docs, where the Fullerton Murphy study was fully debated.

By Anonymous maribeth, CNM, at 1:25 AM  

"And I find your comment to be insulting. The fact is that it appears that being black is a risk factor in and of itself. There are plenty of poor white people and plenty of poor Hispanic people, but they do not have such high perinatal mortality."

I'm sorry. It was insulting. I don't like it when people throw that term around when discussing racial issues, but I did it.

That being said, I still strongly disagree with you. The differences have much, much more to do with poverty than with physiology. I would like you to define the term and prove that there are SIGNIFICANT differences between well-educated wealthy black women and white women .

Even if you match two socioeconomic groups, you still might have cultural differences that affect the statistics. For example, far fewer African American women have abortions than do caucasion women. Another cultural difference that might affect outcomes would be diets.

Again, apologizing for insulting you earlier... but you do have to be careful in the territory you are treading in. White racists have typically believed that there were inherent, physical qualities that made blacks inferior. If you contend that a black woman can't carry and deliver a healthy baby just because of her race, you are in dangerous territory.

The movie "Born in the USA" spotlights a CNM attended birth center in Brooklyn or the Bronx (can't remember which) and they serve a very high population of poor blacks. One of the mothers said something that has really stuck with me. I can't quote it verbatim, but she was having her third child (I think) and said that she was taking responsibility for her pregnancy, prenatal care and health for the first time. She said the midwives really made her feel confident in herself. I think the poor black population would be much better served if the Midwives Model of Care were the standard, of course with OBs taking care of the truly high risk cases.

By Blogger Mama Liberty, at 9:52 AM  

Jamie:

"This study is an indictment of some Australian midwives' standards of care (not listening to FHTs during second stage? disregarding fetal distress? yikes!). Better standards of care lead to better outcomes."

I agree with some of your concerns about this study but it does raise some very important points. At the time that the study was done, the midwives involved in it thought that what they were doing was acceptable. As far as I can tell, they did not violate the standards of Australian midwifery as they existed at the time; in other words, they did not commit malpractice according to Australian midwifery standards.

This study raises important concerns about the judgment of the homebirth midwives. They ignored research very well defined by doctors for decades: breech births and twin birth carry significantly higher risks, monitoring in the second stage is extremely important, and meconium is often a sign of distress.

In that sense, this study is similar to the Pang study where homebirth midwives insisted on delivering babies between 34-36 weeks and had poor results.

It's not as though this doesn't happen even today. Maribeth has clearly expressed her conviction that postdates is an arbitrary medical designation. Furthermore, while searching in Google blog search, I came across a blog post from this week, from a woman rejoicing that her homebirth midwife has agreed to deliver her twins at home.

By Blogger Amy Tuteur, MD, at 10:54 AM  

Jamie:

"I haven't been able to find a clear online distinction btw IUFD and intrapartum death. Suppose a woman planning a hospital birth shows up and says, "I've been contracting all day and I felt the baby kicking an hour ago," but when she's hooked up to the monitor there's no heartbeat."

That would be an intrapartum death. Once labor starts, and the baby is alive before it starts, the death counts as intrapartum. So the deaths in early labor in the midwife group truly are intrapartum deaths and cannot be subtracted.

Fullerton et al. insist they can subtract those deaths because they occured before the midwives got there. However, because the study was done by self reporting, there is no way to verify what really happened. As Sailorman has stressed repeatedly (and correctly), when researchers begin manipulating their data AND the inevitable result of the type of manipulation can only be improvement of their statistics, it raises a red flag.

Intrapartum death in early labor is quite unusual, but of course, it can happen. If Fullerton, et al., had described one such early intrapartum death I would have looked at it and said, "well, that's unusual, but it could happen". However, reporting TWO occurences of a very rare event in a relatively small study group makes me extremely suspicious.

Fullerton et al. were never entitled (by the rules of statistics) to subtract those deaths anyway, so it doesn't really matter what reason they gave. However, the reason is so implausible that it suggests deliberate manipulation of the data to make them look better.

By Blogger Amy Tuteur, MD, at 11:16 AM  

Jamie:

"What about Maribeth's observation that the women she serves are more vulnerable to lethal genetic conditions?"

Maribeth does serve a population with a higher level of genetic conditions that are incompatible with life. However, that has nothing to do with homebirth. It is because the population that Maribeth serves intermarries only with each other. Any small, isolated population will have an increased number of severe genetic anomalies.

There is no reason to think, and I have seen no evidence to date, that women who choose homebirth are more likely to have genetic anomalies or are less likely to know about or less likely to terminate such a pregnancy.

All of this is secondary to the main point, anyway. The authors of these studies describe the anomalies as lethal. Lethal simply means that the anomaly killed the baby. It does NOT mean that the baby would not have survived with appropriate medical care. If the baby could not have survived, the anomaly would be described as incompatible with life.

So, for example, if a baby born with a congenital heart defect is born at home and promptly dies, it has died of a lethal anomaly. However, unless the specific defect is known to be incompatible with life, the death can be considered as preventable in the hospital setting.

By Blogger Amy Tuteur, MD, at 11:25 AM  

"That being said, I still strongly disagree with you. The differences have much, much more to do with poverty than with physiology. I would like you to define the term and prove that there are SIGNIFICANT differences between well-educated wealthy black women and white women"

Gee, I know you don't trust me, but I haven't made up anything yet.

Racial differences in birthweight for gestational age and infant mortality in extremely-low-risk US populations
Authors: Alexander G.R.1, *; Kogan M.D.2; Himes J.H.3; Mor J.M.4; Goldenberg R.5

Source: Paediatric & Perinatal Epidemiology, Volume 13, Number 2, April 1999, pp. 205-217(13)

Abstract:

Using national data, we develop and contrast the birthweight percentiles for gestational age by infants of extremely-low-risk (ELR) White and African-American women and examine racial differences in the proportion of small-for-gestational-age (SGA) births. We then scrutinise racial variations in infant mortality rates of the infants of ELR women. We further compare the infant mortality rates of infants at or below the 10th percentile of birthweight for gestational age of each race group to determine whether infants with similar restricted fetal growth have comparable risks of subsequent mortality. Single live births, 34–42 weeks’ gestation, to White and African-American US-resident mothers were selected from the 1990–91 US Linked Live Birth–Infant Death File (n = 4 360 829). Extremely-low-risk mothers were defined as: married, aged 20–34 years, 13+ years of education, multiparae, with average parity for age, adequate prenatal care, vaginal delivery, and no reports of medical risk factors, tobacco use or alcohol use during pregnancy.

Marked racial variation in birthweight percentiles by gestational age was evident. Compared with ELR White mothers, the risk of an SGA infant was 2.64 times greater for ELR African-American mothers and the risk of infant mortality was 1.61 times greater. For the ELR group, the infant mortality rates of African-American and White infants at or below the 10th percentile of birthweight for gestational age of their respective maternal race group were essentially identical after controlling for gestational age. In conclusion, race differences in fetal growth patterns remained after controlling for risk status. Efforts to remove racial disparities in infant mortality will need to develop aetiological pathways that can explain why African-Americans have relatively higher rates of preterm birth and higher infant mortality rates among term and non-SGA infants.

By Blogger Amy Tuteur, MD, at 11:39 AM  

Amy, this isn't about trust. I trust that you believe strongly in your position. But, clearly as per the section you bolded, we are talking about questions and issues that are very complex. Are you saying you are the one that has the definitive answers? Yes, that one study does show that there are differences in outcomes. But it doesn't prove that African American women can't carry and birth healthy babies simply because of their race. Maybe I am misunderstanding exactly what your assertion is. As I stated in my previous post, there are other factors to consider such as cultural issues. Btw, I would suspect that the differences in infant mortality have alot to do with breastfeeding rates.

By Blogger Mama Liberty, at 12:09 PM  

"But it doesn't prove that African American women can't carry and birth healthy babies simply because of their race."

What do you think it shows?

I don't know why you insist on describing these findings in a way that is pejorative to black women. It is just as easy (and certainly less judgmental) to say that black women, through no fault of their own, have a higher incidence of prematurity and low birth weight.

Afterall, it is pretty well known that black people (like other ethnic groups) are afflicted by specific diseases like sickle cell anemia. We aren't suggesting that they are in some way inferior because of this, just that they have different medical risk factors.

"I would suspect that the differences in infant mortality have alot to do with breastfeeding rates."

Well, then you would need to show that wealthy, well educated black women breastfeed at a lower rate than comparable white and Hispanic women. Can you do that? Otherwise, you are simply making assumptions about these women based on their race.

The bottom line is that you are not going to get very far with an argument that suggests white European countries have lower neonatal mortality that the US because of midwifery. The differences are almost entirely due to the proportion of black women in each population. Everyone (physicians, public health advocates, etc.) knows and acknowledges this.

If we really are committed to lowering the neonatal mortality rate of black infants, we will need to study the medical causes. Blithely suggesting that midwives could solve this problem is absurd and suggests an unwillingness to take their misfortunes seriously.

By Blogger Amy Tuteur, MD, at 12:30 PM  

I'm all about lowering the IMR for black babies, of course. I suspect that there are confounding factors as well as some things we don't understand. First, women of color do not seek primary health care as often as white women. I think there is a lot of under-diagnosed chronic HTN - causing many of those LBW babies, of course. I support you in your efforts to work towards this important goal, Amy.

By Anonymous maribeth, CNM, at 12:52 PM  

It does NOT mean that the baby would not have survived with appropriate medical care

Amy, I know you're standing by your definition. But not everyone in the medical feild uses lethal to mean "likely preventable with appropriate care". 35% of the Murhpy-Fullerton study participants were at high-risk for lethal, incompatible with life, whatever you want to call it - anomalies.

Here is a list of the congenital disorders we face for you to review (http://www.clinicfor
specialchildren.org/
research.html). Some are treatable, some are not.

I disagree that homeborn babies do not have a higher incidence of (untreatable) congenital disease. There are close to 130,000 Amish in America, with an incredibly high fertility rate. 90% + of them have their babies at home. Likewise, midwives in Utah and Idaho serve conservative Morman sects (including polygamous families) in homebirth.

Trust me, the rate is higher than the average public. I also suspect the incidence of lethal trisomies is higher, owing to an avoidance of antenatal genetic testing.

And the amazing thing is that, even given this disproportionately high burden of genetic disease, home birth statistics are still so great!

By Anonymous maribeth, CNM, at 2:31 PM  

Furthermore, while searching in Google blog search, I came across a blog post from this week, from a woman rejoicing that her homebirth midwife has agreed to deliver her twins at home.

I am not endorsing this practice, but I know that midwives who do attend twins at home are not solo. There is a team for each baby and the mother. It's not a midwife and a mom hoping for the best.

By Anonymous MetroMidwife, at 3:03 PM  

Amy, I told you that I think it shows that there are differences due to complex factors that we don't completely understand. I gave you such examples in previous posts. But I'm not sure you are reading them. As per my breastfeeding statement, African American mothers of all socioeconomic levels DO have lower breastfeeding rates.

I was initially excited about this debate forum because I thought it would be a place for an honest exchange of ideas from people with differing opinions. There are many good points that have been made by posters here and I think there could be some very interesting conversations, but you ignore many questions and are only willing to take the conversation into areas you feel comfortable with. Furthermore, you engage in debate tactics that you've asked others not to use. Throwing around studies is one of them. Breastfeeding and childbirth are very hard things to understand quantitatively, because there are so many confounding factors. Emotions, cultures, passions and pre-conceived notions (pardon the pun) are as much a part of this as biology. The subjects of breastfeeding and childbirth are impossible to study with the "gold standard", double-blind randomized trials because as you know conducting those kinds of studies would be impossible and unethical.

I wish you would have taken me up on my challenge to critique the Australian study with the same voracity and critical eye as you do the studies that seem to support the safety of homebirth. If this is merely a forum for critiquing research, I must admit that you will probably do better than I. I am more of a generalist and thrive on ideas and conversations rather than raw statistics.

By Blogger Mama Liberty, at 7:07 PM  

"I wish you would have taken me up on my challenge to critique the Australian study with the same voracity and critical eye as you do the studies that seem to support the safety of homebirth. If this is merely a forum for critiquing research, I must admit that you will probably do better than I. I am more of a generalist and thrive on ideas and conversations rather than raw statistics."

First of all, it seems that Jamie is capable of reading and understanding a paper, even though she does not have an extensive statistics background. It would be more appropriate for you to ask her for her analysis, than for you to ask me.

Second, this situation poses a bit of a quandary. On the one hand, advocates of homebirth claim to be better educated about childbirth than others. They claim to have large amounts of research to support their assertion that homebirth is safe.

On the other hand, it appears that most advocates of homebirth have actually never read the research, and don't have any response to the charge that the data was manipulated. Furthermore, when I offer a paper in precisely the same way that papers are offered to me (the abstract), you ask me to analyze it.

Which is it then? Are homebirth advocates truly relying on research and medical knowledge? Or are they really just trumpeting their personal beliefs without regard to whether there is scientific evidence or not?

By Blogger Amy Tuteur, MD, at 7:53 PM  

"Furthermore, you engage in debate tactics that you've asked others not to use. Throwing around studies is one of them."

No, that's not true. I asked others not to cite papers that they have not read. I'd be very happy if everyone were throwing around papers they HAD read. I have read the papers I cited unless I noted otherwise.

By Blogger Amy Tuteur, MD, at 7:57 PM  

Metromidwife:

"I am not endorsing this practice, but I know that midwives who do attend twins at home are not solo. There is a team for each baby and the mother. It's not a midwife and a mom hoping for the best."

The risk is not going to be reduced by having a second midwife. The problem is that twins are inherently high risk due to the different problems they present. For example, there is a significant risk of abruption of twin #2's placenta when the uterine size shrinks after delivery of the twin #1. This is a life and death situation and can only be treated by an emergency C-section.

The central problem here is that by agreeing to deliver twins at home, the homebirth midwife has demonstrated definitively that she has neither the appropriate knowledge or judgment to do so.

Anyone who has been following these homebirth discussions for awhile will notice a theme that comes up over and over again. The homebirth midwives and their advocates don't seem to have the fundamental knowledge about risks and safety until AFTER a doctor does a study to show that this lack of judgment killed a few babies. We have repeatedly seen this in the case of breech, twins, premature babies (the Pang study), and postdates pregnancies.

There is obviously no doubt that a homebirth midwife can manage an uncomplicated delivery. That's because ANYONE can manage an uncomplicated delivery. Yet as soon as the situation gets complicated, homebirth midwives begin to lose babies that could have been saved in the hospital. That doesn't even include the babies they lose because they did not anticipate the need for resucitation and immediate treatment of congenital anomalies.

By Blogger Amy Tuteur, MD, at 8:09 PM  

On the other hand, it appears that most advocates of homebirth have actually never read the research, and don't have any response to the charge that the data was manipulated

that's an unfair generalization, Amy. And what have we been doing this whole time, other than putting forth our analysis that the data has NOT been manipulated?

By Anonymous maribeth, CNM, at 8:44 PM  

"No, that's not true. I asked others not to cite papers that they have not read. I'd be very happy if everyone were throwing around papers they HAD read. I have read the papers I cited unless I noted otherwise."

You are the one that brought up the Australia study! I asked you to critique it, not Jamie, because I wanted to see if you could "take the other side" (a hallmark of a good debater.)

Amy, I really feel that you are not familiar with how midwives practice. I would put much more stock in what you said if you had spent any real time with midwives or tried to understand the Midwives Model of Care.

"There is obviously no doubt that a homebirth midwife can manage an uncomplicated delivery. That's because ANYONE can manage an uncomplicated delivery. Yet as soon as the situation gets complicated, homebirth midwives begin to lose babies that could have been saved in the hospital. That doesn't even include the babies they lose because they did not anticipate the need for resucitation and immediate treatment of congenital anomalies."

Huh? You really are not familiar with midwifery. Midwives absolutely anticipate the need for resucitation. You do know they bring equipment with them? You do know that Ina May Gaskin has a maneuver named after her for dealing with shoulder dystocia? I would rather have a midwife present in the case of shoulder dystocia than an OB.

I had two AWFUL hospital births, one of which ended with a two week stay in the NICU for my son. The experiences were humiliating and degrading. I had one, incredible, healing, healthy, safe home birth. Yes, in the rare case of a sudden catastrophic emergency, my home birthed son could have lost his life. But my oldest son almost lost his life in a hospital!!! You trade risks. I felt my risks were higher in the hospital.

By Blogger Mama Liberty, at 9:32 PM  

"The homebirth midwives and their advocates don't seem to have the fundamental knowledge about risks and safety until AFTER a doctor does a study to show that this lack of judgment killed a few babies."

If midwives don't have fundamental knowledge, why do they teach labor and delivery to medical residents in some programs?

Lack of judgment on the part of obstetricians has killed babies as well. But we are really starting to digress here.

So basically, as I understand it, your position is as follows:

The World Health Organization is full of crud.

The American Public Health Association is full of crud.

European nations with high utilization of midwives and/or high rates of homebirth don't have black people so their statistics are better.

Homebirth midwives kill babies.

Is this more or less accurate?

By Blogger Mama Liberty, at 9:43 PM  

Mama Liberty:

"You are the one that brought up the Australia study! I asked you to critique it, not Jamie, because I wanted to see if you could "take the other side" (a hallmark of a good debater.)"

We are not discussing my debating skills. We are discussing the safety of homebirth. Let's be honest, here. You asked me to do the analysis because 1. you don't know how to analyze it yourself and 2. you planned to criticize anything I said.

Yet if midwives cannot analyze the research themselves, how can they be sure that homebirth is safe? If midwives apparently don't understand the special risks posed by twins, the special risks posed by breech deliveries, the special risks posed by premature babies and the special risks posed by postdates babies, how can they possibly practice safely?

And if practicing midwives apparently cannot understand the most fundamental information about childbirth, how on earth can patients be expected to "educate" themselves about the risks of homebirth?

"I would put much more stock in what you said if you had spent any real time with midwives or tried to understand the Midwives Model of Care."

I was physician backup for literally hundreds of CNM deliveries. I know all about the Midwives Model of Care. However, that isn't relevant, anyway. We are talking about the safety of homebirth and research demonstrates that it is not safe, not my personal experience (although I happen to have extensive experience).

"Midwives absolutely anticipate the need for resucitation. You do know they bring equipment with them?"

Equipment is meaningless if you don't have extensive experience in using it AND the judgment to know when it is needed.

"Homebirth midwives kill babies."

That's what the research shows.

I want to return, though, to a fundamental point. Homebirth midwives are not knowledgeable about the research. They don't know what it is and they don't know how to evaluate it when it is shown to them. They base their faith in the safety of homebirth on their personal opinions or because someone else told them it is so.

Equally important, they don't seem to understand the risks in high risk situations. They seem to think (without citing any evidence) that doctors made it up. Ignorance is never a virtue, and in caring for pregnant women, it can be deadly.

By Blogger Amy Tuteur, MD, at 11:01 PM  

I can't help noticing that through out the many topics being debated on this blog that no one has brought up the moms choice in making the decision to take on the risk of the home birth and that the Midwives are honoring the choice that their client is making. In the end it is not the Midwife or the Doctor who can make the choice for the women, it is her and her partners choice. All a Midwife and Doctor can do is explain the benefits and risks and leave the choice to the person who it effects!!

By Anonymous Doulagal, at 12:09 AM  

Doulagal:

"I can't help noticing that through out the many topics being debated on this blog that no one has brought up the moms choice in making the decision to take on the risk of the home birth and that the Midwives are honoring the choice that their client is making."

We actually have discussed it; I'm not sure which thread it is it. We are all in agreement that it is the mother's choice. What we are discussing is whether homebirth midwives give women true and accurate information with which to make their choice.

If homebirth midwives are telling women that homebirth is as safe or safer than hospital birth, they are telling them something that is compeltely untrue. All the studies that are touted as showing homebirth is safe, actually show an excess of neonatal deaths compared to a low risk population delivering in a hospital.

By Blogger Amy Tuteur, MD, at 6:47 AM  

So, one other question. If you take the "good" studies (the one's not flawed) what is the statistical differences between home and hospital births when looking at infant mortality?
And have you considered that the women would still choose the Home birth option knowing the differences, due to their mistrust of the patriarchal system in place at a hospital?

I live in a very large city and we can not have free standing birth centers because the AMA made sure that it became illegal to have one! So women's choices, in where they birth, is limited to home or hospital.

By Anonymous DoulaGal, at 8:55 AM  

Doulagal:

"I live in a very large city and we can not have free standing birth centers because the AMA made sure that it became illegal to have one!"

Free standing birth centers are not illegal.

By Blogger Amy Tuteur, MD, at 9:08 AM  

Here is a relatively recent study from Canada which fulfills many of Jamie's criteria for a good study. It shows a higher neonatal mortality rate in the homebirth group as well as a higher rate of adverse outcomes. Interestingly, the majority of the hospital births were attended by midwives, not doctors.

Outcomes of planned home births versus planned hospital births after regulation of midwifery in British Columbia
Janssen et al; CMAJ, February 5, 2002; 166 (3).

Results included:

Perinatal death rate
homebirth 3.5/1000
hospital birth 0.8/1000

Babies needing ventilators for more than 24 hours
homebirth 5/1000
hospital birth 0

maternal hemorrhagic shock
homebirth 2.3/1000
hospital 0

Maternal blood transfusion
homebirth 3.5/1000
hospital 0.8/1000

Hospital birth is clearly safer in this study than homebirth. This is despite the fact that the authors manipulated their data to improve it. They removed anomalies from both groups, but they failed to tell us what the death rate for anomalies was in the home and hospital groups. I would be willing to bet that the death rate for babies with anomalies was dramatically higher at home than in the hospital. Had the study included babies with anomalies (by the rules of statistics, it should have), the results probably would have been far worse for the homebirth group.

By Blogger Amy Tuteur, MD, at 9:42 AM  

NOBODY here thinks that women should not be allowed to birth at home. The choice between a home and hospital birth is a choice between apportionment of risk.

The debate is purely about INFORMATION provided to the expectant mother.

Informed consent requires that the mother have accurate information regardig her risk. If the mother does not have accurate information, her consent cannot be informed.

Don't you see that?

Everyone keeps saying we don't want home births. This is not true. I think mothers (sadly) have the autonomy and right to make a lot of risky choices, and I would no more try to ban home birth than I would try to make it illegal to drink or smoke during pregnancy.

We just want people to stop lying to mothers about how safe it is.

By Blogger sailorman, at 9:48 AM  

Amy,

They are in Illinois!( I will post a link as soon as I find it!)

By Anonymous DoulaGal, at 9:50 AM  

Amy,

Someone on another list made a post reminding me of an issue with anomalies. It FEELS (I have no data) more common in mainstream hospital based obstetrics for women to receive extensive prenatal testing to rule out anomalies, and then be offerred or encouraged to terminate when certain types of anomalies are detected, even some compatible with life (trisomy 21. OTOH, it feels that it is more common than in the general population for homebirthers to forgo this testing and choose to birth whatever baby they are carrying. If true, that would skew some stats and require that either that all elective terminations for anomalies be put back in or some neonatal deaths by anomalies be removed. Right?

MM, Georgia

By Anonymous Anonymous, at 10:25 AM  

"MM, Georgia":

Yes, you are someehat correct. IF there were a proven/known statistical difference in the populations with respect to # of anomalies which are brought to term, this would have to be accounted for in the analysis. Any difference between the populations gets accounted for.

I However, to my knowledge this has not been proven. You may "feel" it's true, but it has not been shown to be true. That's a big distinction. Also, the choice of statistical method you propose to "balance" things is not really correct.

I wrote a longer explanation but deleted it, as I'm not sure exactly what your question is. If you have a more specific question regarding the statistics I'll be happy to try to address it.

By Blogger sailorman, at 11:50 AM  

One quick addition:

In order to have an effect, even theoretically, you wouldn't have to have a difference in abortions per se.

Rather, you'd have to have a difference in abortions which resulted in a population difference in babies who would, even theoretically, be unable to respond to hospital treatment.

The stats are fairly complex.

By Blogger sailorman, at 11:56 AM  

Sailorman said: "Yes, you are someehat correct. IF there were a proven/known statistical difference in the populations with respect to # of anomalies which are brought to term, this would have to be accounted for in the analysis. Any difference between the populations gets accounted for.

"I However, to my knowledge this has not been proven. You may "feel" it's true, but it has not been shown to be true. That's a big distinction. Also, the choice of statistical method you propose to "balance" things is not really correct."

100% correct about knowing no proof and I said this up front. I can anecdotally say from significant familiarity with the subculture, that in the hb community in and around my large city, it is quite typical for hb mw clients and UCers to have no visual ultrasounds, no quad screens, no amniocentesis. This definitely seems much less common in OB practice and OBs are still the main hospital birth care provider in this country. Totally anecdotally, before leaving my first OB I told him I did not want a second trimester US because I would not be terminating either way and he said something like "We are going to have a real problem with that." So I think it's a hypothesis that would need to be explored before we decide what to do with including/excluding hb babies who die from anomalies in studies. We need to entertain the possibility of this unproven factor: hospital survival stats benefit from medical terminations removing babies at higher risk of neonatal death from the cohort.

So now my question is, assume the unproven hypothesis, how do we then need to treat the data?

MM, Georgia

By Anonymous Anonymous, at 12:15 PM  

Amy: the link as I promised!


http://www.hmprg.org/birthcenters.html

By Anonymous DoulaGal, at 12:21 PM  

"So now my question is, assume the unproven hypothesis, how do we then need to treat the data?

MM, Georgia"


FIRST, you have to design a test for, and prove, your hypothesis. THEN, you carefully consider adjusting your data, IF such adjustment is supported by your hypothesis AND IF your study really strong.

Chaging data is one of the most dangerous things you can do--if not THE most dangerous thing you can do. There is no better way to obscure reality than fiddling with data. You never fiddle with or change with your data based on a suspected belief.

If you do, you have replaced "scientific study" with "guess".

Your ideas about "exclusion" aren't really correct. You seem to have a good starting idea of how this works though, and I think I can explain the problem to you.

Let's start by narrowing down your hypothesis. I'll give you a hint and comment that abortion is completely irrelevant. It doesn't matter which babies DON'T get delivered, as they're not included in any of these stats anyway. (inclusion/exclusion again). It only matters which babies are ultimately delivered.

Anyway, I think your hypothesis is this:

"The population of women which is treated by midwives has a higher incidence of anomalies incompatible with life".

Is this correct? If not, try to state it as clearly as you can so we can continue.

By Blogger sailorman, at 12:44 PM  

Dr. Amy, the most important criterion for me is statistical significance, and it looks like none of the differences you mention even hit p=.05. You also neglect to mention the ways in which hospital birth came off worse than homebirth, some of which are statistically significant. For instance, significantly more (p=.01 in the PA column) hospital-born babies had APGARs <7 at 1 minute. That confirms what I've been saying all along -- hospitals can fix more problems, but they also see more problems. Those babies are probably going to be fine, since they bounced back, but it's a scary thing to be that mother, watching them work on your baby with the mediocre APGAR.

Here is the table with neonatal interventions and outcomes, for anyone daunted by wading through the full text.

It is not clear that including congenital anomalies would make the hospitals look better, since there were 18 cases in the hospital group and 5 in the homebirth group.

Where are the studies in which the author agrees with you about the import of his results?

By Blogger Jamie, at 12:49 PM  

MM:

"it feels that it is more common than in the general population for homebirthers to forgo this testing and choose to birth whatever baby they are carrying. If true, that would skew some stats and require that either that all elective terminations for anomalies be put back in or some neonatal deaths by anomalies be removed. Right?"

No, here's why. Even if it were proven that homebirth mothers had a higher incidence of anomalous babies at term, there is no evidence that they want those babies to die. The issue is not whether there are more anomalies, but whether babies with and without anomalies are more likely to die at home than in the hospital.

Furthermore, none of the studies we have discussed (including those written by homebirth proponents) has even suggested that there is a higher incidence of anomalies in the homebirth group. They simply subtracted the anomalies without any explanation.

By Blogger Amy Tuteur, MD, at 1:01 PM  

"The population of women which is treated by midwives has a higher incidence of anomalies incompatible with life".

Hypothesis:

A higher % of the pregnancies of women recieving prenatal care from hospital providers are electively terminated on the basis of results of prenatal screens and/or tests than are the pregnancies of women receiving prenatal care from home birth providers.

Propositional logic has a place both in science and this debate. Also, I can't imagine, if I was designing and executing a homebirth safety study, that it would cause me to *change data*. The data are what they are but they must be interpreted in a way that is logical.

MM

By Anonymous Anonymous, at 1:14 PM  

Jamie:

"Dr. Amy, the most important criterion for me is statistical significance, and it looks like none of the differences you mention even hit p=.05."

Awesome pickup! Doesn't help in this case, though. That's because the authors choose to apply a Bonferroni correction to the numbers (essentially making it more difficult to reach statistical significance) when there is no apparent reason to do so.

This study suffers from what all homebirth studies suffer from: insufficient numbers of participants. Since there are so few participants, the differences between the groups have to be quite large to attain statistical significance. We have essentially agreed to ignore that fact because there are NO studies that reach the desired number of participants.

I am analyzing this paper following the same rules we have used to analyze the others. If you want to change the rules, we'll have to go back and look at everything again.

If statistical significance were the most important thing to you, you'd have to ignore all the studies done of homebirth to date, including those that claim to show that homebirth is safe.

By Blogger Amy Tuteur, MD, at 1:22 PM  

Jamie, I just read that study in full.

C'mon.

I mean, Dr. Amy posts aboput the various DEATHS and you counter with "well, they were less Apgar at 1 minute!" DO you think those are equivalent?

And you're not even quoting the study accurately!! There was no significant difference there at all. You may have been fooled by the statistics the authors used; the numbers are tricky. The proper p value is not 0.01.

You are correct on one fashion: the number of major problems is too small to provide much statistical support. HOWEVER, the numbers for home birth are atrocious (see Dr. Amy's post for specifics).

You can really do one of two things with these numbers: 1) Ignore them entirely; or 2) Use them in support for hospital birth. It's fair to say "they don't count". But you can't in good faith use them in support of home birth.

By Blogger sailorman, at 1:31 PM  

"A higher % of the pregnancies of women recieving prenatal care from hospital providers are electively terminated on the basis of results of prenatal screens and/or tests than are the pregnancies of women receiving prenatal care from home birth providers."
MM


This won't work as a hypothesis. Not because it isn't true (it might or might not be true, who knows).

It's not relevant, though, because it doesn't address the matter at hand: "for a given birth, which is the safer location?". As stated, this doesn't even give support for considering it as a factor at all.

This isn't an issue of propositional logic, it's an issue of study design. By focusing on abortion, you're losing track of who matters.

We are studying neonatal deaths based on birth site.

You are (I think) claiming that part/all of a difference in deaths is explained by population differences.

Thus, abortions would ONLY be relevant if you draw a connection to population difference, AND suggest how that dfference could/would affect the neonatal death rate.

By Blogger sailorman, at 1:36 PM  

Sailorman, are you saying that the homebirth vs. physician-assisted hospital birth calculation is wrong, or are you saying the comparison doesn't reach statistical significance for homebirths vs. all hospital births?

By Blogger Jamie, at 3:59 PM  

"You are (I think) claiming that part/all of a difference in deaths is explained by population differences."

No, I never *claimed* that. I don't know. I wonder. It's a question I would want answered along with other questions about whether a study is well designed.

By Anonymous Anonymous, at 4:10 PM  

I'm saying that it wasn't significant. Which mirrors your apparent protest about the study, since you appear to pass quickly over the huge (though not statistically significant) differences in perinatal death, maternal need for transfusions, etc).

I'm also saying it is a tad disingenious, when the data appear to show numerous EXTREMELY dangerous issues with home birth (and no 5 minute Apgar difference at all) to focus on a short term indicator like that.

But hey, if you want to talk about actual numbers and not significant differences, I'm all for it. Let's start with perinatal deaths.

I mean, don't you want to at least try to address the issues Dr. Amy raised?

And most of all, don't you want to discuss the effect of removing the anomalies on the statistics?

One of the major benefits of a hospital over a home birth is the ability to respond to non-normal situations: congenital anomalies, maternal hemmorhage, etc etc. Why would you remove these from the study if you were actually looking to find out which site was safer?

By Blogger sailorman, at 4:23 PM  

"One of the major benefits of a hospital over a home birth is the ability to respond to non-normal situations: congenital anomalies, maternal hemmorhage, etc etc. Why would you remove these from the study if you were actually looking to find out which site was safer?"

You might need to remove some congenital anomalies because, as you know, you need the cohorts to be as similar as possible if what you are seeking to study is the safety of homebirth environment and practitioner versus hospital environment and practitioner as far as neonatal mortality goes. If hospital practices skew their cohort in the neonatal period by having more terminations in the second trimester, it's relevant to what I think the question is. MM

By Anonymous Anonymous, at 4:43 PM  

MM:

"You might need to remove some congenital anomalies because, as you know, you need the cohorts to be as similar as possible if what you are seeking to study is the safety of homebirth environment and practitioner versus hospital environment and practitioner as far as neonatal mortality goes. If hospital practices skew their cohort in the neonatal period by having more terminations in the second trimester, it's relevant to what I think the question is."

I have searched the medical literature and the lay literature. I cannot find even a single scientific reference to a study, survey, or anything else that even raises the possibility of increased congenital anomalies in the homebirth population.

Furthermore, none of the midwives or doctors who have published research on homebirth has ever mentioned the possibility that there is a higher rate of term congenital anomalies in a homebirth population.

There is a complete absence of any evidence for this theory and in the absence of evidence it is nothing more than wishful thinking.

Since almost all pro-homebirth studies deliberately exclude congenital anomalies without explaining why, it is almost guaranteed that the true perinatal death rate at homebirth is much higher than has been reported to date.

Essentially what homebirth advocates have been saying is that if your baby dies from a congenital anomaly at a homebirth, we don't care! We don't even think it is worth mentioning.

By Blogger Amy Tuteur, MD, at 5:14 PM  

Okay, Sailorman, let's talk about the deaths. Quoting from the article, one stillbirth had no obvious cause and appeared to have occurred before the onset of labor. Autopsy of the third baby showed "severe hypoxic ischemic encephalopathy with extensive hemorrhage and infarction in other major organs, indicative of a pre-labour insult." Maybe the hospital could have saved the second baby, who died in early labor; maybe not. (Would the mother, if she'd been planning a hospital birth, have arrived in the hospital at that point? Would she have been hooked up to the monitor at the time of the baby's death? Would there have been time to organize a C-section? We can't know.)

Let's also observe that it's a little unusual to have 1 perinatal death in the 1314 hospital births. I'm glad it worked out that way for those families, but it's not truly representative.

And let's talk about the anomalies, too. You saw that the incidence was more than double in the hospital groups (1.3% and 1.4% vs. 0.6% for home-born babies). I'm not sure why you think including those outcomes would make the hospitals look better. Hospitals can't do much for babies with trisomy 18. They can't treat anencephaly. Home-born babies with treatable cardiac anomalies may suffer as a result of increased transport time, true. That's a risk, small but real, of homebirth.

Oh, I've been meaning to say: another possible reason for excluding congenital anomalies is that parental response introduces another variable. In some situations, parents may choose not to resuscitate a very sick baby; other parents may opt for aggressive albeit fruitless intervention.

By Blogger Jamie, at 5:27 PM  

Jamie:

"Hospitals can't do much for babies with trisomy 18. They can't treat anencephaly. Home-born babies with treatable cardiac anomalies may suffer as a result of increased transport time, true. That's a risk, small but real, of homebirth."

Both trisomy 18 and anencephaly are extremely rare congenital anomalies. There is no reason to believe that any of the babies who died had either.

The vast majority of congenital anomalies are treatable. You can be sure that if the deaths in these studies were caused by congenital anomalies that were incompatible with life, the authors would have mentioned it. They sliced and diced their data to present as strong a case for homebirth safety as they could.

No, there is no reason to believe that these impaired babies died for any other reason than they were born at home, where no one had the requisite knowledge or skills to save them.

What's really amazing to me is the effort to ignore these babies, as if their deaths did not matter and only the deaths of healthy infants count.

By Blogger Amy Tuteur, MD, at 5:50 PM  

Jamie,

Because the numbers we're talking about here are so tiny (incidence well under 1% for many things) it's important to be specific.

In many cases, when a death occurs at home, we don't know if it was preventable.

We DON'T KNOW.

We never will. It is logically impossible.

The only way--the only way you will know if deaths are actually preventable is to compare two populations. If one population has fewer deaths, then they're preventable. there is no other statistically valid way to make the distinction. "a little in abundance is a lot". Even if a hospital will consistently save 1 in 1000 babies with a given condition, you need to capture that data. Because there are lots of conditions, and it adds up.

If it was "obvious" then you wouldn't need to include the anomalies, but then we wouldn't be having this conversation.

You can't perform autopsies and "guess" who would have been "saved". The autopsies are not especially reliable for that; they don't account for about a gazillion factors. And, of course, they have potential for bias, from both sides. Reliance on the autopsies is unfounded. They are merely an interesting source of discussion.

In terms of the anomalies: It's certainly true about anencephaly or trisomy 18. However, those weren't the causes of death here, are they? Every single death listed, could THEORETICALLY have been prevented.

(And BTW, where are you getting those numbers? I can't find them using a search function.)

Jamie, you are playing the "yeah, but" game:
1) We make a point.
You say "yeah, but" and show a study abstract.

2) We analyze the data faults and show the study doesn't work.
You say "yeah but" it's not statistically significant.

3) We note that all the studies in SUPPORT of your data aren't significant. We note that the DATA (not the abstracts) in studies cited by home birth people tend to show hospitals are safer.
You say "yeah but" hospital birthers abort more feti (WTF is up with that?)

4) We note this is pure conjecture.
You say "yeah but" maybe they let them die! Yeah, that's it!

Kee-rist, Jamie, you otherwise seem like an intelligent sort. I can't beleive you haven't read the study.

But instead of talking numbers you're now trying to assert that any difference is caused by selective abortions, or perhaps parents "mercifully" letting their babies die...?

Hopefully even you can see this is getting ridiculous.

When will this end? I used to think you could be convinced by science and logic but I'm beginning to think it's more like a religion to you.

By Blogger sailorman, at 5:59 PM  

Absolutely, those babies with congenital anomalies mattered. Whether they lived brief or long lives, whether they underwent aggressive intervention or died quietly in their parents' arms -- their lives mattered. This study doesn't tell us anything about the outcomes, though. So we can't assume that the hospital-born babies lived and the home-born babies died. We can't assume that the anomalies required intervention in seconds rather than minutes, and that transfer meant worse outcomes. We don't have the information.

Something that's interesting to me about this study is the authors' varied affiliations: family practice, ob-gyn, peds, midwifery, and epidemiology, by the looks of it. It seems unlikely to me that they're colluding to hide dead babies. Not impossible, but a definite stretch.

By Blogger Jamie, at 6:02 PM  

Sailorman, the numbers are in the footnote to Table 5, linked in a comment of mine up above. I'll respond to more of your comment later.

By Blogger Jamie, at 6:04 PM  

probably client population studies need to be done before further testing- so apples are compared to apples-
say a study that shows care in places that have Amish populations ---

--------------- the reasons I say this is because I would really like to know how often and what tests OB clients refuse-- compared to what home birth clients refuse
--------------
I would also like to see exactly which areas are problems areas in homebirth-- which is not what you are asking about - if there are certain practices that need to be changed in order to improve home birth I want to know- and just because the hospital has different answers does not mean that there is not room for improvement there either-
-----

By Anonymous Anonymous, at 6:10 PM  

Jamie,

I found the numbers, thanks. My search function apparently didn't search the footnotes.

When you respond as promised, let's stick to numbers, rather than wondering about abortions rates, OK?

Anyway:
"You saw that the incidence was more than double in the hospital groups (1.3% and 1.4% vs. 0.6% for home-born babies). I'm not sure why you think including those outcomes would make the hospitals look better."

I don't know. Maybe it would make the hospitals look worse. But we're looking for truth! Or at least, I'm looking for the truth. I'm having my doubts about a few other folks here.

5 babies were home birthed with anomalies. 18 babies were born in the hospital with anomalies. How many of those lived? If the hospitals are worse, I want to know. OTOH, if the hospitals are saving lot of babies who would otherwise die at home I want to know.

Don't YOU want to know, too?

By Blogger sailorman, at 6:35 PM  

Jamie:

"This study doesn't tell us anything about the outcomes, though. So we can't assume that the hospital-born babies lived and the home-born babies died."

The problem is that the studies are SUPPOSED to tell us about those outcomes. They just can't announce that they are excluding them for no apparent reason. Since there is no legitimate scientific reason to exclude them, and since it undoubtedly makes their results better (where do YOU think babies with congenital heart disease would do better?), we are forced to conclude that they excluded them to improve their data.

I'm not sure why you are resisting acknowledging this Jamie. In the absence of any other credible explanation (and no one has offered a credible explanation), we must accept the three obvious conclusions: homebirth leads to preventable neonatal deaths, there is no evidence that homebirth mothers have a higher incidence of congenital anomalies at term, that babies babies with congenital anomalies who are born at home are more likely to die than if they are born in the hospital.

By Blogger Amy Tuteur, MD, at 8:36 PM  

The logic here is confounding! You admit you don't know that they were anomalies 'incompatible with life'. (And do check with your local perinatologist, because as I mentioned, that term is out of favor. Why? Because, said my referral MFM doc, "only absence of a pulse or respirations are incompatible with life".)

Anyway, you are then taking this huge illogical, non-evidence based leap that well, they 'must have been' preventable, or treatable anomalies. And yet another leap saying that hospital care would have been better. You don't know that. You've never once responded to my post about the lack of tertiary level care available in the average hospital (making your argument weak, as if you're trying to avoid that reality).

Midwives carry O2, midwives are proficient at NRP (which isn't actually needed in most cardiac anomalies, if that's news to you), midwives DO transport, and get appropriate care in a timely matter. But there you go suddenly saying "these babies died because of bad NRP". Where's the data showing that? It is not evidence-based. It's pure conjecture. Methinks you are getting away from yourself in the heat of debate.

If you want to argue -- will a baby with an undiagnosed diaphragmatic hernia do better at home or in a tertiary care center, you get no argument from me. But it appears to me that you are interested in treating every single baby as 'a diaphragmatic hernia waiting to happen' and that is scary. That is exactly why OB medicine is so problematic in the US.

I have been reading all the studies, and the comparable low-risk hospital births, again, and I still cannot identify a higher mortality rate at home. All your arguments depend on "but if's" that may or may not be true, but are not fact.

And sailorman? When you said this: Kee-rist, Jamie, you otherwise seem like an intelligent sort... REAL nice. Better believe Amy would have deleted Jamie or my post, if we'd written that.

By Anonymous maribeth, CNM, at 11:01 PM  

babies with congenital anomalies who are born at home are more likely to die than if they are born in the hospital
Oh wait, and another leap: where was this data?

homebirth leads to preventable neonatal deaths

I don't think that is true. But even if you insist it is, would you admit that a 28% c/section rate -- likely 18% of them unecessary, per Amy's personal estimation -- with a fourfold increase in maternal mortality, and 50% higher incidence of fetal mortality in future pregnancies, sort of balances out the scale??

By Anonymous maribeth, CNM, at 11:07 PM  

Okay. I've been looking since this conversation started for statements from epidemiologists opposed to homebirth. If the American Public Health Assocation supports OOH birth but the Canadian Public Health Association says, "No, wait, too many dead babies" -- I want to know about that. If the epidemiologists agree, as they seem to, on the safety of homebirth, then I think the burden of proof is on those who say it's unsafe. And if you disagree with the authors, peer reviewers, and editors of every single paper we talk about except for Jenny Pang's (who acknowledges the limitations of her paper), I get increasingly skeptical. I'd still like to see a link that distinguishes IUFD and IP death, because my Stedman's doesn't and it's a central part of your critique of the Murphy-Fullerton study.

Gosh, there's a lot to respond to here, so maybe I'll break it up into a few different comments. I'll just offer this one as context for my vexation -- I was disappointed, Amy, in your comment at Dr. Crippen's blog. Is anyone here trying to say that childbirth is risk-free? You asked what kind of data would convince me that homebirth was unsafe, and I told you. I'm not leaving comments that say, "Dude, my naturopath saw my baby's beautiful golden-pink aura and told me that his fear of the hospital was palpable on the vibrational plane. That's why I stayed home with a footling breech presentation, and you can't tell me what's right for my family." If I had a breech baby, I would call an obstetrician to arrange a C-section and I would be grateful for his (or her) expertise. I agree with you that the risks of home breech birth are too large. But I'm still waiting for you to quantify the increase in risk that you perceive for me, as a low-risk multip half a mile from the hospital, with a vertex baby and an experienced CNM.

I'm curious, too, Amy, about whether there's anything interesting or new in this conversation for you. Has there been a moment when you said, "Huh, never thought about that before"? Did you wonder why the FPs had so many more low APGARs at one minute, before you quoted the five-minute stat in your comment? Let me turn your question around, and ask you what kind of data, specifically, might convince you that homebirth can be a safe option for low-risk women?

By Blogger Jamie, at 11:40 PM  

Sailorman, you are correct that deaths are much more important than one-minute APGARs. I chose one example to counter Dr. Amy's comment. I could have left a comment mirroring hers (e.g., there were four times as many cord prolapses in the hospital, and two and a half times as many placental abruptions, etc.) but I don't think that's an especially helpful approach.

What's your opinion on the use of the Bonferroni correction, btw? The authors' explanation seems consistent with other online sources.

You said: But we're looking for truth! Or at least, I'm looking for the truth. I'm having my doubts about a few other folks here.

I'm interested in truth, too; that's why I'm sticking around. My comment about parents' choices was not meant to cloud the issue. When a baby is born with a fatal anomaly, parents have more than one option, legally and morally. If, say, a hospital-birthing couple requests that hospital staff attempt to resuscitate a baby with an untreatable anomaly and a homebirthing couple doesn't, in this study the resuscitation attempt would go into the hospital's stats and make them look artificially worse. The neonatal team wasn't ethically bound to resuscitate the baby, but neither was it inappropriate for them to do so.

I don't mean to play the "yeah but" game, and you can go ahead and call me on it when you see it. I haven't been talking about abortion anywhere in this thread, have I? I'm trying to evaluate this paper using the parameters I mentioned up above.

By Blogger Jamie, at 12:10 AM  

The problem is that the studies are SUPPOSED to tell us about those outcomes. They just can't announce that they are excluding them for no apparent reason.

Here again I'd like to see a source for that assertion. I don't mean it to sound hostile, but I just don't know if that's standard practice or not. In this comparison of delivery mode, for instance, malformed fetuses are excluded from the calculations.

Since there is no legitimate scientific reason to exclude them, and since it undoubtedly makes their results better (where do YOU think babies with congenital heart disease would do better?), we are forced to conclude that they excluded them to improve their data.

I think outcomes for babies with congenital heart disease vary with the severity of the pathology more than they do with place of birth. In some cases, a baby with a CHD could die as a result of homebirth. In most cases, you hurry to the hospital when you see cyanosis. Some horrible CHDs aren't immediately symptomatic, putting homebirthing and discharged hospital-birthing families in the same boat. Without confirmation that the authors' methodology is unsound, I'm not persuaded that they're trying to hide any bodies.

...we must accept the three obvious conclusions: homebirth leads to preventable neonatal deaths,

Sometimes it does. So does hospital birth (you noticed there were two uterine ruptures in the hospital-birthing group and none at home, right? how much does induction/augmentation increase the risk of rupture?). I'm waiting for the data to show more deaths at home with good standards of care.

there is no evidence that homebirth mothers have a higher incidence of congenital anomalies at term,

Agreed -- I think it's an interesting possibility but no more.

that babies babies with congenital anomalies who are born at home are more likely to die than if they are born in the hospital.

Need to see some better data on that, Dr. Amy.

By Blogger Jamie, at 12:35 AM  

Jamie:

"If the epidemiologists agree, as they seem to, on the safety of homebirth, then I think the burden of proof is on those who say it's unsafe."

I see no evidence that epidemiologists agree that homebirth is safe. Certain organizations have made statements that homebirth is safe enough to allow. That does not mean that statisticians and epidemiologists agree with the following assertion that appears on every homebirth advocacy website I have ever seen, that homebirth is as safe or safer than hospital birth.

"I think the burden of proof is on those who say it's unsafe."

No, that is a cop out. First of all, I've already done that about 20 times over.

Second the burden of proof is on those who say it is safe. There is NO DATA to support that claim, none, zero, nada.

Lest you think that this is something special devised to confuse homebirth midwives, let me assure you that the burden of proof is ALWAYS on those who claim that something historically unsafe is now safe. It is the homebirth midwives who are asking for an exception to the rule that applies to everyone else and they are not going to get it.

"I was disappointed, Amy, in your comment at Dr. Crippen's blog."

I think I can live with that.

Seriously, Jamie, I can tell you are struggling with this. Clearly you are reading the papers, puzzling through them, and finding out that I am not making up what I say.

However, after literally hundreds of posts back and forth, you are virtually the only one (with perhaps one or two exceptions.) Homebirth advocates start with a conclusion and try to work back from that, violating every principle of science. They reject any evidence that doesn't conform to their personal beliefs. They are constantly accusing everybody of conspiracy theories (always a sign of desperation). Most importantly, to my mind, they decided long ago without ANY idea of the data; they just parrot what others told them.

Homebirth advocates are constantly praising themselves for being so well educated. That's absurd. They are among the least educated on the topic. NONE of them have read the literature, very few even know how to read it. NONE of them are aware that there is literally no evidence that homebirth is as safe or safer than hospital birth. NONE of them appear to know even the MOST BASIC statistics about neonatal and maternal mortality.

I have been debating this topic for over a decade, and I have NEVER seen an advocate who comes to the discussion with the following: "yes, there may be a slightly increased risk, but that is acceptable to me." It always starts with the same lie, that homebirth is as safe or safer than hospital birth. How educated can a group of people be when NONE of them know the truth.

Have you ever noticed that "the stars" in the homebirth movement NEVER debate experts in medicine or science. They always speak to and write for lay people who are unable to analyze what they say. That's because the "stars" know they will be cut to ribbons. If they really thought they had a case, they'd be happy to face off against real scientists, but they ALWAYS run in the opposite direction while loudly declaiming that they are being persecuted.

That is the sign of people who fear the truth, not people who know the truth.

By Blogger Amy Tuteur, MD, at 7:27 AM  

Let's talk about anomalies. in theory, this should be something pretty easy to discuss.

Jamie/maribeth/etc:

I will now post a series of statements.

What would be REALLY REALLY helpful in this conversation is fo you to agree/disagree wihth each statement individually.

If you disagree with ANY STATEMENT, please explain why FOR THAT STATEMENT.

1) Some infants have congenital anomalies.


2) Some of those congenital anomalies are undetectable by modern medical techniques prior to birth.


3) Some of those anomalies which ARE detectable (in theory) are nonetheless undetected prior to birth, either because they are 'missed' or the tests are not run.


4) Some of those anomalies will, in theory, be treatable at home using the equipment likely to be at a home birth.


5) Some of the anomalies require treatment using equipment/skills/personnel that are only available in a hospital, never at home.

By Blogger sailorman, at 7:37 AM  

You might also want to read the post
"More than Zero but Relatively Small"

Which doesn't talk about home birth per se but explains why we are getting so frustrated.

it's at
http://moderatelyinsane.blogspot.com/

By Blogger sailorman, at 9:20 AM  

To Sailorman re: anomalies-- no argument whatsoever on nos. 1-4. I don't accept #5 quite as written. In a few cases, your baby will be best off in a tertiary care center, no doubt about it. If that baby is born at the Pocatello Medical Center, he's not going to get the same level of care. So I'd readily accept #5 if it said, "Some of the anomalies require treatment using equipment/skills/personnel that are only available in specialized hospital units, never at home."

By Blogger Jamie, at 10:30 AM  

Amy, can you fix the link to the UK homebirth site? I think you need to add the www. Also, do you have access to the full text of this paper?

By Blogger Jamie, at 10:46 AM  

Jamie:

Your protest against #5 is unfounded. I'm using words which have an exact meaning, so there's no need to attempt to "qualify" my statements.

Not incidentally, I think this may be part of the problem with the discussions: If you think my words mean something else, then the care I take in making a point (and the accuracy of an argument) goes away.

Anyway... I said this:
"5) Some of the anomalies require treatment using equipment/skills/personnel that are only available in a hospital, never at home."

Your response seems to replace the word "only" with "always" or "often". That is not what I said.

I am sorry to lecture you. But this logical gap is frustrating. Saying something is "only available in a hospital" is completely different from saying it is "always available" or even "usually available".

Hmm, how to say this without being insulting...

If you, or anyone else, is applying this same level of reading exactitude and logic to the studies, it is entirely understandable that you might reach an inaccurate conclusion about what they actually say or prove. But just because it's understandable doesn't mean it's correct.

By Blogger sailorman, at 11:12 AM  

BTW Jamie:

I think you are rare among the posters here in that you DON'T seem to be doing this deliberately. I am trying to walk a fine line between explaining it (which requires me to be in 'teaching mode') and not scaring you off.

But I am going somewhere with this. So stick with the logic if you can.

By Blogger sailorman, at 11:24 AM  

"I have been debating this topic for over a decade, and I have NEVER seen an advocate who comes to the discussion with the following: "yes, there may be a slightly increased risk, but that is acceptable to me." It always starts with the same lie, that homebirth is as safe or safer than hospital birth. How educated can a group of people be when NONE of them know the truth."

Wait ... that is completely consistent with *my* position. I know you must be getting frustrated once the hyperbole starts. There may very well be a slightly increased risk of neonatal mortality in the choice to home birth. I'm not sure we have the ultimate study on the matter either way. I still call it within the bounds of "safe" and will advocate for preservation of birth choice and against myths in our culture's beliefs about birth. I care about the lives of any babies lost in home birth that probably would have been preserved in a hospital setting, and about babies that are lost because of hospital practice and conditions that probably would have lived at home. The whys of these are the meat of informed choice about whether to home or hospital birth, based on the particular situation of the birther who is the decision maker.

MM, Georgia

By Anonymous Anonymous, at 11:37 AM  

"So I'd readily accept #5 if it said, 'Some of the anomalies require treatment using equipment/skills/personnel that are only available in specialized hospital units, never at home.'"

Jamie, by my understanding of English and logical reasoning, if the above statement is true, then sailorman's #5 is also true.

MM, Georgia

By Anonymous Anonymous, at 11:53 AM  

Okay, point taken. But when you get to where you're going, Sailorman, please be sure to distinguish btw hospital care in the hypothetical Pocatello Medical Center and hospital care in a Level III NICU.

By Blogger Jamie, at 12:04 PM  

Jamie:

What I'm trying to do here is explain how I and many other scientists view this data.

What we establish through those statements, is what we might call a
"one-way" line.

This is to say that if those statements are true--and I am glad you agree they are true--SOME of these anomalies can only be treated in a hospital. It doesn't mean they WILL be treated.

Given those statements, you can't say that hospitals will treat more anomalies.

But it DOES mean that for the categories of anomalies we are talking about, hospitals can only be "better" than home birth, never worse.

And this, incidentally, is not affected at all by the hospitals in question.

Why? Well, the answer is actually fairly simple: If neither the midwife or the hospital can treat an anomaly, the net effect is zero. If the hospital can and the midwife can't, the net effect is positive for the hospital.

Because question #5 means those anomalies can never be treated at home, you can never have a situation where the effect is positive for the midwife.

And why doesn't the hospital in question change things? Two reasons:

1) Location: the hospitals which the midwives would "use" via ambulance transport are the same hospitals they'd be compared against.

2) See explanation above re question #5.

The hospital might change the AMOUNT of the "advantage" that a hospital provides, but it cannot change the logical FACT of the advantage in this instance.

And before I get jumped on: I am well aware this is only one factor in the overall safety argument.

By Blogger sailorman, at 1:11 PM  

"But it DOES mean that for the categories of anomalies we are talking about, hospitals can only be 'better' than home birth, never worse."

I think you mean better or neutral.

You assume that the survival of the anomaly is affected by the transport time from the home birth; or that the home birth practitioner/participant is less qualified to detect the anomaly and initiate intervention. Might be true often enough. Might not. Where's the data?

Also, in some cases, the home born baby can be transported to a Children's Hospital faster than the local maternity unit can transfer to Children's. I don't think this is a significant factor in favor of the safety of home birth, but logically, if this is a survival issue often enough, your conclusion that hospital birth must be safer in terms of neonatal mortality for anomalies requiring hospital care might fail.

MM

By Anonymous Anonymous, at 2:03 PM  

oops--sorry--"better or neutral" is indeed correct. I apologize for the error; the rest of my post uses the "or neutral" correctly.

MM, your points are certainly relevant. You're making one mistake though:

"You assume that the survival of the anomaly is affected by the transport time from the home birth;
or that the home birth practitioner/participant is less qualified to detect the anomaly and initiate intervention.


First, we're discussing anomalies which CANNOT be treated at home. By this definition, the whole detect/intervene thing you mention is simply not an issue.

Second, I don't think anyone will argue that transport time is a POSITIVE factor. Which is to say: it's either neutral, or negative.

I think most people would agree that it's a negative factor: If you're in an emergent state and heading for a hospital, delay is bad. Do you really want to argue otherwise?


As for "where's the data?":

The whole point of this discussion is that we can--and should--agree on a lot of thing BEFORE we start talking about data. These statements are true in the absence of data, using logic.

"Also, in some cases, the home born baby can be transported [from a home faster than from a hospital]...logically, if this is a survival issue often enough, your conclusion...might fail."

Yes, it might. IMO this is a pretty weak counterargument though, especially if you consider any superior ability of a hospital to stabilize a patientprior to transport, and/or to ensure that proper supervision is available during transport. But hey, it's possible. I don't think it's possible "enough" to really spend more time on though.

A confession: this is actually a psychological test of sorts.

I have picked what seems to be to be the EASIEST issue for a home birth person to concede: In the event of a 1) undetected; 2) serious anomaly which is 3) not treatable at home; and 4) requires hospital treatment, more infants will die at home than in the hospital.

It is a VERY limited statement. It is VERY well supported by logic as well as data, though really you shouldn't really need data to concede it probably true. It has NO BEARING AT ALL on the ultimate issue of comparative safety between hospital/homebirth.

And yet you--and every other home birth advocate I've asked about this--are unwilling to concede. ANYTHING. EVER.

I wish I could say I'm shocked.

P.S. Do you have easy "targets" too? Of course. You might, for example, try to get a hospital-birth advocate--like me--to admit similar issues about hospitals. Assuming you make properly limited statements like the ones I made, you'll have no trouble getting people to admit to them. Go ahead and try it.

By Blogger sailorman, at 2:46 PM  

What I meant by intervention in that instance was stabilization and transport, since our given was "can't be treated at home." Not that you could read my mind as I was typing :)

I don't need to go around pushing buttons. You presented a narrow issue, which is fun to play with but isn't an overbearing concern for many parents in choosing home or hospital birth. It's part of the neonatal mortality safety picture though, which is why it is on point in this thread.

You whinged at me about "claiming" something that I never claimed, I simply hypothesized. You act superior in your scientific/logical approach when you are clearly going outside of predictible logic into assumptions without hard data -- and as every fifth grader is taught, an assumption makes an ass out of u and me. And make studies easy to pick at and hard to rely on.

You're forgetting, I'm the one who's happy to concede that in terms of neonatal mortality, home birth may be slightly riskier. The fact is, I'm not emotionally attached to home birth being "safer" in some absolute way in terms of neonatal mortality across a population. I would still advocate for home birth choice.

MM

By Anonymous Anonymous, at 3:33 PM  

Jamie:

"In this comparison of delivery mode, for instance, malformed fetuses are excluded from the calculations."

The paper which you reference is a comparison of vaginal breech delivery and C-sections. Malformations represent a confounding variable and that is why it was removed. One of the more common malformations is hydrocephalus which would render breech vaginal delivery virtually impossible. The authors could have kept certain anomalies in the study (like heart defects), but rather than break the anomalies down into groups they just removed them all. Neither proponents nor opponents of breech vaginal delivery would object to this, since it is not going to affect the principle question of the study.

Removing congenital anomalies from a homebirth study does not have such a specific rationale. Now if you wanted to say to clients: look, homebirth is likely to be safe if your baby doesn't have any unanticipated problems, but it will probably kill the baby if it does have unanticipated problems, it would be fine to exclude anomalies. However, homebirth advocates don't say that. If they are going to claim that homebirth is safe, then they better include all the babies of all the women who are having a planned homebirth.

By Blogger Amy Tuteur, MD, at 5:13 PM  

Jamie:

"do you have access to the full text of this paper?"

No, I can't get it. The paper appears to be based on the National Birthday Trust study of 1994, which is a book. Dr. Chamberlain also wrote his own book about it, but I don't have access to that either. Furthermore, no matter how I search, I cannot find any actual neonatality statistics for homebirth in the UK.

By Blogger Amy Tuteur, MD, at 5:17 PM  

"However, homebirth advocates don't say that. If they are going to claim that home-birth is safe, then they better include all the babies of all the women who are having a planned homebirth."

here is where you are way wrong-- we have fairly extensive informed consents - and it is usually discussed in the first visit or second when a client signs up with a midwife-- for CPMs ( the credential included in the study) it is mandatory to have this statement as well as practice guidlines and an ethics statement as well -- additional consents for each procedure/or test as well

By Anonymous Anonymous, at 5:59 AM  

Anonymous:

"here is where you are way wrong-- we have fairly extensive informed consents - and it is usually discussed in the first visit or second when a client signs up with a midwife"

That's surprising to me since I have not yet seen a homebirth advocacy site that doesn't proclaim home birth to be as safe or safer that the hospital.

Is there really language in the consent form that says that in certain rare situations the baby might die for lack of resources that would be found in the hospital? Would it be possible for you to find one of the consent forms and quote the relevant passage?

Discussing the possibility in a visit is not enough, unless there is specific documentation in the chart of specific warnings.

If homebirth midwives are truly aware of the possibility of avoidable neonatal death at home, why don't they mention it on their websites and in their books?

By Blogger Amy Tuteur, MD, at 7:00 AM  

NEWSFLASH!!!!

Concede this little point, and change your practice in this little manner, and we'll leave you alone!!!

"Is there really language in the consent form that says that in certain rare situations the baby might die for lack of resources that would be found in the hospital? Would it be possible for you to find one of the consent forms and quote the relevant passage?"

We support choice! We only want INFORMED choice.

By Blogger sailorman, at 9:41 AM  

Is there really language in the consent form that says that in certain rare situations the baby might die for lack of resources that would be found in the hospital? Would it be possible for you to find one of the consent forms and quote the relevant passage?

The above-mentioned parents have chosen to have their baby at home or in the birth center. They understand that there are special responsibilities and risks associated with such a decision. Although most potential problems can be predicted, there are some complications that cannot be predicted at home, the birth center, or in the hospital.

Some of the possible complications that might occur at any birth are lacerations, hemorrhage, and cord prolapse, infection, fetal distress, multiple deliveries, non-vertex presentation (breech), birth injuries, and stillbirth. Some of these complications could result in death, permanent sterility or serious mental or physical disability for the mother and/or baby. Your Name Here cannot guarantee a normal birth of a healthy mother or a healthy baby. The acceptance of the mother for home birth or birth center birth in no way guarantees the final outcome.

Your Name Here will bring certain emergency equipment to the parent's home or will of such equipment available at the birth center, bur of course this will in no way be equivalent to the equipment, medicine and monitors available in the hospital. Your Name Here does bring or has on hand at the birth: oxygen, drugs to control hemorrhage, DeLee devices for deep suctioning of the infant, Doppler for electronic monitoring of the baby's heartbeat and other equipment.

There is no doubt that the hospital is the safest place for an abnormal birth. Your Name Here will make continual assessment to make sure the mother and baby remain low-risk and are suitable for an out-of-hospital delivery.

By Anonymous Anonymous, at 5:01 PM  

Anonymous:

Thank you so much for providing that info from the consent form. That's exactly what I'd like to see. A woman who signs that indicates that she understands the risks.

Is this consent form from a particular practice or is it uniform in your state?

By Blogger Amy Tuteur, MD, at 8:06 PM  

Is this consent form from a particular practice or is it uniform in your state?

My state requires I disclose the risks but leaves it up to the individual midwife to compose her informed consent paperwork. I have been guilty of the "safe or safer than a hospital" language but this discussion has inspired me to detail the risks rather than highlight the benefits alone. If they didn't know the benefits of home birth, they wouldn't be considering it.

By Anonymous Anonymous, at 8:10 PM