Things Amy Romano of Science (Fiction) and Sensibility Doesn’t Want You To Read
Posted July 21, 2010on:
Here’s a quick compilation of all things sent as criticisms to Science (fiction) and Sensibility in response to her critique of a recent home birth meta-analysis that Romano would not publish, plus a few bonus points. http://www.scienceandsensibility.org/?p=1349 They cannot be reposted verbatim, because math geeks worry about being ignored for being too boring, more than deleted for controversy.
It’s a “discredited methodology” As mentioned in a prior post , Birth Junkies are also Cochran Review Junkies. Anything that deviates from their methods is bad. Rest assured, Cochran’s Paint-By-Numbers Review methods are not the only acceptable way to do a meta. In fact, it’s a very amateurish way. People who know what they are doing, don’t need a cookbook. Or perhaps she is referring to the methodology of the actual studies. Also, not true.
Birth Junkies complain that a well-known study from Washington state did not distinguish between unplanned and planned home births. But, they did limit it to near full-term. The percentage of full-term women who accidentally give birth at home is minsicule. Middle-of-the-snowstorm rural Washingtonians might be found here. This is different than precipitious birth that occur accidently occur outside the hospital, but not at home. The main reason they don’t make it is the birth is going so smoothly that it happens too fast. These actually have among the lowest death rates. But, both are very small part of the close to full-term crowd. The error is miniscule and favors the Junkies to boot. More importantly, given the huge increase in death rate in the homebirth, a huge percentage of full-term out of hospital births would have to accidental to shift the conclusion. They still lost.
Decades old? Again, this could only help home birth. The rate of low-risk infant death in hospitals has dropped dramatically over the years. In the days of old, the higher hospital rates would be closer to the home birth crowd, which haven’t changed much since dinosaurs roamed the earth. Again, a glitch that can only help them. They still lost.
Some Didn’t Count Intrapartum Death Again, this omission favors midwives whose own data shows they kill early and often. In physicians, the ratio of intrapartum asphyxia death and post-birth asphyxia death in low risk pregnancies used to be 1 to 1. Because of fetal monitoring and the dreaded c-section, it is about 1 to 3, today. In U.S. lay midwives, it is 1 to 2. The main reason it doesn’t get studied anymore is because in physicians and their low risk patients, the overall rate is almost non-existant (3/10,000) Lay midwives int he US have been shown to have rates of 10/10,000 or three times higher. Yes, it is very amenable to birth strategies, ones midwives don’t do! Third time is a charm. It favors midwives and they still loose.
Adding deaths that occurred between 8-28 days (which are less likely to be related to intrapartum events and therefore are less modifiable by birth setting), She was blissfully unaware that birth asphyxia kills in this timeframe and is very relevant to the discussion. Midwives deaths in the US, tend to occur later among those who make it out of the womb alive. Neonatal birth asphyxia death is largely a controlled decision to discontinue life support and can be delayed for social reasons. By eliminating the rest of the first month, you are eliminating a disproportionate share of the midwifery deaths.
She doesn’t give us a lot of detail what is so discredited. It is simply important to blanket the airwaves and ether with that notion whenever unfavorable data emerge.
The Forest Plot Too funny. See related post at this blog. The people who know what they are doing don’t need a paint-by-numbers research method. They don’t need pictures either. They certainly don’t need them where technically you are not supposed to have them.
And she is perplexed by the use of a fixed-effects model for the analysis of neonatal death. It says in the paper they tested for homogeneity, how they did it and why they did it. Romano is apparently perplexed because she doesn’t know what any of these big words found in the paper mean. But, saying some big word’s use perplexes you makes it look like you do know what they mean and that your badmouthing is very learned instead of more like than propaganda.
Meta-analysis is about Power and The Truth The tablets came down from the mount and told her this. Probably, at the end of the Cochran Paint-By-Numbers Review manual, the final instruction is “This is the Truth, you must implement this in clinical practice as mindlessly as you performed this review.” Midwives may have these monolithic beliefs, but the rest of us do not.
The Netherlands She suggests we should simply adopt the findings of a study from the Netherlands that showed that home birth and hospital birth had the same death rates.
Apparently the unflawed, not discredited, One True and Just methodology for meta-analysis is where the selection criteria is only one single solitary lonely study you can find that give you the results you want.
She notes, in great detail, the figures from the study. The death rate is 3.4/1000 in home and hospital. It doesn’t register that this is about three times higher than the US, at 1/1000. It doesn’t register that this is bad.
When it is pointed out that The Netherlands doesn’t have safe homebirth with midwives, but rather crappy hospital birth with midwives, as well, she evades the issues by saying this is not a cross-cultural comparison. Ok, let’s make US homebirth as safe as hospitals by taking everything out of the latter.
A confidence interval you could drive a truck through? She notes “a 95% chance that home birth increases the risk of neonatal death by somewhere between 32% and 625%” Ummm, uumm Amy, the entire, drive your truck through it interval is all way, way, way on the side against home births. This is fact is bad for your position. Very very very bad. Whatever the True Increase is, it means homebirth is worse, much worse. You looked at everything up and down, good and bad, things you like and things you don’t. The best it gets is 32% worse. It doesn’t register.
She concludes with notion we have to bring this practice which, unbeknownst to her, she has just shown is waaaay bad into the mainstream. She states that there is no “standard safety net in place for women who begin labor with the intention to birth at home but turn out to need hospitalization in order to birth safely” I believe this thing is called the emergency room. Do they not teach that is the “U.S. Maternity Care And How the Patriarchy Ruined It” class?