Looking for something? Start here.
Custom Search




« New Study Identifies Need to Distinguish Planned from Unplanned Home Births | The Seedy Underbelly of Jill's Hard Drive »

OB/GYN Journal Fast Tracks Anti-Home Birth Study in Advance of Pro-Midwife Legislation

International Expert Calls Study Deeply Flawed and Politically Motivated

Bookmark and Share


Original press release available at TheBigPushforMidwives.org


WASHINGTON, D.C. (July 7, 2010) – As New York and Massachusetts moved to pass pro-midwife bills in the final weeks of their legislative sessions, the American Journal of Obstetrics and Gynecology fast-tracked publicity surrounding the results of an anti-home birth study that is not scheduled for publication until September. Described as unscientific and politically motivated, the study draws conclusions about home birth that stand in direct contradiction to the large body of research establishing the safety of home birth for low-risk women whose babies are delivered by professional midwives.

“Many of the studies from which the author’s conclusions are drawn are poor quality, out-of-date, and based on discredited methodology. Garbage in, garbage out.” said Michael C. Klein, MD, a University of British Columbia emeritus professor and senior scientist at The Child and Family Research Institute. “The conclusion that this study somehow confirms an increased risk for home birth is pure fiction. In fact, the study is so deeply flawed that the only real conclusion to draw is that the motive behind its publication has more to do with politics than with science.”

Advocates working to expand access to out-of-hospital maternity care questioned the timing of AJOG’s public relations efforts on behalf of a study that won’t be published until next fall.

“Given the fact that New York just passed a bill providing autonomous practice for all licensed midwives working in all settings, while Massachusetts is poised to do the same, the timing of this study could not be better for the physician groups that have been fighting so hard to defeat pro-midwife bills there and in other states,” said Susan M. Jenkins, Legal Counsel for The Big Push for Midwives Campaign. “Clearly the intent is to fuel fear-based myths about the safety of professional midwifery care in out-of-hospital settings. Their ultimate goal is obviously to defeat legislation that would both increase access to out-of-hospital maternity care for women and their families and increase competition for obstetricians.”

The United States recognizes two categories of midwives: Certified Nurse-Midwives, who are trained to practice in hospital settings and who also provide primary and well-woman care, and Certified Professional Midwives, who undergo specialized clinical training to provide maternity care in out-of-hospital settings. Research consistently shows that midwife outcomes in all settings are equivalent to those of physicians, but with far fewer costly and preventable interventions, including a significant reduction in pre-term and low birth weight births, and as much as a five-fold decrease in cesarean surgeries.




PrintView Printer Friendly Version

EmailEmail Article to Friend

Reader Comments (16)

there is also a good article about the new study in midwifery today online http://www.midwiferytoday.com/articles/ajog_response.asp

July 7, 2010 | Unregistered Commenteratara

Can you please link to an online copy of the Wax paper? I have looked everywhere for an online coy and not been able to find one. Thank you!

July 7, 2010 | Unregistered CommenterPhilippa


July 8, 2010 | Unregistered CommenterDeleted

Amy, do you even go here?

July 8, 2010 | Unregistered CommenterKatherine

Anyone can join MANA, a 13 yr. old Irish setter can join, so join- get the data and go to town. Write a gazillion blogs about how planned home birth has triple the rate of death/demise ....yadda yadda yadda. I mean whatrealy are they going to do- sue you if you expose them????

The better more important news from that press release is that legislation is passing in states where home birth midwives remain unregulated and essentially unlicensed as care providers. This is critically important to the safety of mothers and babies. ACOG does not need to hand wring about us all flinging open the L&D doors and having ALL our babies at home. Home birth parents comprise approximately 1% of state populations we owe it to them to protect them from charlatans and imposters.

ACOG as a professional trade organization as well that always protects the interests of its members- I have pointed out before the "you asked, we answered" section under the practice management, professional liability tab, and one cannot even peek at the answers to those juicy questions if they are not a member so essentially, you can't even be their friend without a medical license so at least the midwives at MANA will let anyone be their friend. (Herein lies the problem- that is another topic for another time.)

Here are some of the greatest hits from ACOG's site...
"YOU ASKED, WE ANSWERED: How to Evaluate Your Defense Attorney (Sep 21, 2009)"
"YOU ASKED, WE ANSWERED: Resident Feels Unqualified to Perform Certain Procedures (Oct 3, 2008)"
"YOU ASKED, WE ANSWERED: Protecting yourself while working locum tenens (May 2, 2008)
"YOU ASKED, WE ANSWERED: Terminating the Physician-Patient Relationship (Aug 14, 2008)"

hmmmmmm so let's not be flinging mud all over each other- it's time to fix the broken system.

[This (insightful) comment partially edited]

July 8, 2010 | Unregistered CommenterSaanenMother

I don't know if anyone else has read the meta-analysis, but I did and I have a few questions for clarification. I am almost done with a Ph.D. in educational psychology, so I understand the research parts of the study. But a few terms are leaving me confused. First, "nonanomalous". In the article, the researchers say that when only nonanomalous neonates are included, the neonatal mortality rate triples. That doesn't make any sense to me--I understand nonanomalous to mean "without any obvious anomalies"? Am I wrong? Because if I were the researcher and I found a tripling of a death rate when removing the "participants" that had obvious anomalies (e.g. when we removed patients with cancer from the sample, we found a tripling of the death rate in the sample, WTF?)...well, that is so backwards that I'd be looking at whether I miscoded variables or interpreted data incorrectly. Anyone else who has read the study, can you clarify?

Also, I understand perinatal mortality to mean death between 22 weeks gestation and 7 days post-birth, and neonatal death to mean death up to 28 days. Why then would a need for resuscitation and failure to resuscitate increase the neonatal death rate but not the perinatal rate? Why is that the problem these researchers focused on?

One last comment about the article. If I found a tripled neonatal death rate but no significant differences between home vs. hospital for perinatal death rates, as a researcher I'd be speculating what other factors might correlate with homebirth (but not be an essential part of homebirth) that might cause neonate mortality...e.g. what is prenatal care for these babies? Do the mothers feel that they can take their babies in for care if something is wrong? Do homebirth midwives need more training in triage for postnatal complications? My point is, as a researcher it seems premature to blame neonatal death on homebirth. The researchers cannot make causal statements because it was not a true experiment. Their broad sweeping generalization that mothers are sacrificing safety for less intervention is unwarranted; instead they should recommend further research into what might be mediating the relationship between homebirth and increase neonatal mortality.

July 8, 2010 | Unregistered CommenterKathie

Here's a breakdown of the Wax paper which is scheduled to be published in September.

Usually the PR machine on studies don't begin months before an article is published, which begs the question asked in The Big Push's press release... what's the rush? Something political going on that might benefit from influencing public opinion?

July 8, 2010 | Registered CommenterJill

In response to Saanen Mother's excellent questions:

I have also read the entire metanalysis, as well as each of the 12 studies reviewed.

You are correct, "nonanomalous" would refer to a baby with anomalies that could be the cause of death. I suspect the reason that the home birth death rates were higher after removal of the infants with anomalies is that most of these were likely identified by ultrasound before birth, and delivered in the hospital because of the known anomaly. This would skew the hospital's neonatal death rate unfairly because of delivering a higher proportion of babies with known anomalies.

<Also, I understand perinatal mortality to mean death between 22 weeks gestation and 7 days post-birth, and neonatal death to mean death up to 28 days. Why then would a need for resuscitation and failure to resuscitate increase the neonatal death rate but not the perinatal rate? Why is that the problem these researchers focused on?>

There are varying definitions of perinatal and neonatal mortality. The study investigators defined perinatal death as "still birth of at least 20 weeks or 500 g or death of a liveborn within 28 days of birth", and neonatal death as "death of a liveborn within 28 days of delivery". Using these definitions, the perinatal death rate could include a significant number of very premature babies who were not resuscitated, and stillborn babies who were not resuscitated. Since the neonatal death rate only refers to babies born alive, we can assume that any of these babies that was born with respiratory difficulties was (or should have been) resuscitated.

I believe the researchers are focused on resuscitation because they state, "2 cohort studies implicated intrapartum asphyxia in 31% and 52% of planned home delivery perinatal deaths. The past 2 decades have seen a significant decrease in such deaths, with evidence suggesting fewer fetuses experiencing intrapartum anoxia."

They go on to speculate that modern obstetric technology is the reason for the decrease in anoxia as a cause of neonatal death in the hospital. The study authors do not discuss any of the numerous studies that indicate continuous electronic fetal monitoring (EFM) has not improved the neonatal death rate (but has tripled the c-section rate). Intermittent fetal heart rate auscultation, done with a handheld monitor, has been shown to result in equally good outcomes as continuous EFM.

<My point is, as a researcher it seems premature to blame neonatal death on homebirth.>

I don't think the researchers specifically blamed neonatal death on homebirth. They do state that less intervention (which is common at homebirth) is associated with a tripled neonatal death rate. However, even this conclusion is faulty, as they admit in their own metaanalysis that the increased neonatal death rate disappeared when they studied only those births attended by certified midwives or nurse-midwives. Thus, the obvious conclusion to me seems to be that the qualification of the birth attendant has more correlation with neonatal death rate than place of birth or amount of interventions.

In fact, a study by Janssen, et al., titled 'Outcomes of planned hospital birth attended by midwives compared with physicians in British Columbia' (Birth. 2007 Jun;34(2):140-7) found that midwives who attended births in the hospital used much less intervention than physicians, had lower rates of perinatal morbidity, and equivalent rates of neonatal mortality. In fact, babies born to the midwife group needed less medication to reverse respiratory depression caused by painkillers given during labor, and less resuscitation. Other studies have supported the conclusion that midwifery care results in fewer routine interventions, lower morbidity, and equally good neonatal outcomes.

What is equally interesting about this metaanalysis is that 9 of the 12 studies reviewed concluded that planned home birth with a certified attendant had equally good neonatal outcomes, compared to hospital birth. One of the studies was too small to draw conclusions (only 5 women in each group) and one of the studies, the infamous Pang study, complied data from birth certificate information, which does not indicate planned place of birth or type of birth attendant (if any).

All in all, this metaanalysis is biased and flawed, and belongs more appropriately on the OpEd page of the journal.

July 8, 2010 | Unregistered CommenterJanelle

The Wax study is not yet available online because it is set to be published in September. You can view the abstract and information on requesting a full copy of the metanalysis at: http://www.elsevier.com/wps/find/authored_newsitem.librarians/companynews05_01582

July 8, 2010 | Unregistered CommenterJanelle

Regarding Kathie's questions and Janelle's responses:

Re: neonatal death rate tripling when removing nonanomalous babies -- these births/deaths are removed from both the numerator AND the denominator. Some babies with congenital anomalies live and some die, but regardless, all of them are removed from the numbers because they generally have nothing to do with location of delivery or attendant. Also, midwives may serve a higher proportion of women who decline genetic screening or refuse to terminate because of an anomaly, which often skews the deaths rates higher in the home birth population when anomalies are not addressed. So in the end, math-wise, when you remove the same number of babies from a rather small numerator and a very large denominator, the resulting rate can be very different.

Re: definition of perinatal mortality and neonatal mortality -- Wax et al did indeed define what they considered to be "perinatal" and "neonatal":
"perinatal" = 20 weeks gestation to 28 days of life
"neonatal" = live birth to 28 days of life
However, the studies they analyzed did not necessarily conform to that definition. For example, they used the massive Netherlands study (de Jonge et al, BJOG, 2009) with over 320,000 planned homebirths in their perinatal mortality calculations. [As an aside, here's another problem with Wax's analysis: even when combined with 3-5 other studies, the de Jonge data represented 97% of the total perinatal sample, which obviously skews the results very heavily toward the de Jonge findings (no difference in perinatal mortality rates between home and hospital) -- why even bother doing a meta-analysis when the other studies can't possibly influence the results?] de Jonge defined the perinatal period as intrapartum + 24 hours after a live birth and also intrapartum + 7 days after live birth. This is not even close to Wax's definition; he's not even comparing apples to apples, which is a big no-no in meta-analysis design.

Furthermore, Wax chose NOT to include the huge de Jonge study in the *neonatal* mortality analysis! Instead, he has a much smaller dataset, nearly 40% of which is from the Pang study which included preterm births from 34 weeks, as well as unplanned and unattended homebirths. It has already been well established that the rates of complications and deaths are much, much higher for preterm babies and unattended, unplanned home deliveries (Burnett et al, JAMA, 1980; Philipps, BMJ, 1984; Hinds et al, JAMA, 1985; Abernathy et al, Pub H Reports, 1989). Including the Pang data in his meta-analysis is the sole reason for the tripling of the neonatal mortality rate in the homebirth group. When he removes the Pang study from the analysis (as well as a small Australian study with <1000 women that showed no difference between home & hospital), the difference in neonatal mortality vanishes.

Cleverly, Wax describes the Pang study (and the other small Australian study) as "not clearly specifying home birth attendants OR in which home births were conducted by other than certified or certified nurse midwives" (which, by the way, is only an American distinction; no other country delineates nurse-midwives and direct-entry midwives, so technically, they should have all been thrown out). In describing it this way, Wax is able to imply that it is the lesser trained "other" midwives who are to blame for this tripling of neonatal mortality, when in fact it is the unattended, unplanned and preterm births in the Pang study that confound the results.

Wax also *suggests* that intrapartum asphyxia may be a problem in many of these deaths, citing two out-of-date studies that "implicated intrapartum asphyxia in 31% and 52% of planned home delivery perinatal deaths." We have much newer and robust data that is now widely-accepted that we in fact cannot often distinguish between a neurological insult that occurred prenatally versus during labor (Fahey & King, J Mid Wom Health, 2005). That throws the "intrapartum asphyxia" argument out the window as well, although it sure sounds good when you're making a case that midwives don't know how to resuscitate babies and therefore should not be allowed to attend births at home.

So, problems with Wax's conclusions.

Stating over and over that there is a tripling of the neonatal mortality rate for homebirths, when removal of a highly flawed study eliminates that difference, is highly unethical.

Suggesting that there is a higher neonatal mortality rate at home because midwives don't have the skills or equipment to resuscitate babies is 100% deliberately misleading speculation because his data includes a large portion of unplanned homebirths and preterm deliveries that midwives had nothing to do with. Of course preterm babies have higher rates of respiratory distress, we know this, it's well established.

Analyzing studies from all over the world, where the integration of midwives into the healthcare system is vastly different, healthcare systems in general are vastly different, and transport times from home to hospital are vastly different, does NOT make for a homogenous population with which to study differences in birth location. The women themselves may be low risk (or, perhaps not at all in the case of the Pang study), but the systems in which midwives must operate are not, and this matters.

Suggesting that the lower rate of interventions is a cause of higher neonatal mortality rates, and that more interventions result in fewer cases of intrapartum anoxia is so misleading to women trying to make an informed decision. There is no place for that kind of speculation in a research article or in evidence-based care.

Being surprised that they found no difference in perinatal mortality (when one study made up 97% of the sample) but *did* find a difference in neonatal mortality when they excluded that giant study, is just ridiculous. Then speculating that another reason why babies die more often at home is because they have higher birth weights and longer gestational ages is shockingly ironic, especially given that a large reason for this supposed difference is the inclusion of preterm babies in the sample.

There's no good reason for this to be getting as much fast-tracking from AJOG and media attention all over the world, other than for political reasons.

July 8, 2010 | Unregistered CommenterWendyCPM
Comments for this entry have been disabled. Additional comments may not be added to this entry at this time.