Previously, I had written this post about the CDC statistics for neonatal and infant mortality for the years 2003-2004. It included a link to the online searchable query, so you can do the computations and searches yourself.
Now, I’d like to take a closer look at them. This is partly because of the ensuing comment-debate on “Who’s Catching Your Baby?” at the RHRealityCheck blog. It is quite an interesting discussion, which was, of course, started by the ubiquitous Dr. Amy Tuteur (does that woman never sleep?? It seems like whenever there’s a homebirth discussion anywhere, she’s one of the first commenters; in addition to quickly responding to or deleting comments I’ve posted on her blog as late as midnight or as early as 6 a.m. Maybe she’s a clone…) Anyway, now that the CDC stats are able to be queried online, she’s been able to go from beating the dead horse of the Johnson & Daviss 2000 CPM study to riding the 2003-2004 stats to death. In a comment on July 21, she says that the CDC data, “shows that homebirth with a DEM is the most dangerous form of planned birth in the US!” She follows it up with a triumphant table with the following death-rates per 1,000 live births: CNM — 0.37, MD — 0.61, “other midwife”– 1.15.
And she says there’s nothing to “interpret” but there is. (I encourage you to read several of the rebuttal comments, and preferably all of them, because there is some extremely important information given in them.) Let’s take a closer look at these numbers which ostensibly show that homebirth with a DEM has a threefold death-rate over women in a similar risk category (hospital birth with a CNM).
First, I want to point out that there may be some “bad apples” in every profession, including obstetrics and midwifery. The laws on midwifery vary from state to state. About half of the states have legalized non-nurse midwives, some states have them declared illegal, and other states don’t have laws one way or the other. Mississippi is one such state. I could be wrong, but I believe that it would not be illegal for me in my state to start calling myself a midwife and start attending births. My qualifications? None. I’ve given birth twice in my home and I’ve read a lot of books and other information. While most of the time nothing would go wrong, if it ever did, I would be completely unequipped to handle anything. Yet in compiling statistics, my incompetence would be grouped in together with competent non-nurse midwives, skewing the statistics and giving the whole profession a black eye.
A brief aside — midwives can be grouped into two broad groups: certified nurse midwives (CNMs) and non-nurse midwives, also known as direct-entry midwives, since they bypass nursing school and enter midwifery directly. Of these non-nurse midwives, some are certified and some are not. Not being certified does not necessarily mean that one is not qualified or competent; but certification does show that one has passed a certain rigorous education. Some states have much stricter regulations than others on who midwives can have as clients — women with certain previous or current obstetrical histories “risk out” and must seek care from an obstetrician. While some of these restrictions are debatable, many are just plain common sense, and even in unregulated states, many midwives will not take such clients — for their own good and for the sake of their babies. But some might. So there may be some high-risk births in the “other midwife” set as compared to the CNM set. Dr. Amy would like to pretend that the home-birth group is homogeneous and uniform, but it may not be — I’ve heard enough anecdotal accounts of riskier women giving birth at home to verify it. I’ve also heard several accounts of midwives who resisted a transfer when she should have transferred; and others when the midwife strongly urged the woman to transfer and she refused to go to the hospital — sometimes with deadly results. Whether this changes the statistics or not, or whether it is “significant” or not, is for someone else to determine.
Now onto the numbers.
In the above “Oh, my goodness, homebirth kills three times as many babies as CNM-attended hospital birth!!” scare, Dr. Amy has made a few correct restrictions of data.
The first was maternal age, which is restricted to the low-risk 20-44 year old sets. Also, for some reason, the majority of women who give birth at home are white. (I just ran a query for all births attended by “other midwife” — regardless of cause of death, length of gestation, age of mother, etc. — it was as broad as I could make it — and found that 93% of women who gave birth with “other midwife” as the birth attendant were white, while CNMs and MDs had 79%. Since there is a recognized difference in birth outcome based on race — although we may question and debate the reasons — it is correct to look at just white births, to more closely approximate the groups.
She also limited it to 37+ weeks of gestation and 2500+ grams of birthweight. Since it is known that babies born pre-term or low-birthweight fare worse, this is a logical step to take. I would like, however, to see a study which takes women of equal risk at the beginning of pregnancy, and see how they fare when comparing midwives to doctors — in other words, do midwives do a better job at keeping women pregnant until term, and having normal-weight babies? This is only a tangential discussion of neonatal mortality; but babies do better when they’re born at term and at a healthy weight. If midwives do a better job at keeping babies healthily on the inside until they’re ready to be born, then that is significant. There was a recent article I blogged about, which talked about “the worrisome rise in underweight babies,” and included the fact that many doctors are contributing to this by inducing or giving women C-sections before they reach term. If doctors are contributing to the neonatal death toll by inducing or sectioning women before they reach 37 weeks, or if these babies are born weighing less than 2500 grams, then that simply won’t show up when doing a CDC query of babies born at 37+ weeks and 2500+ grams. Now, obviously, there are many reasons why some babies should be born preterm, so I’m not advocating for refusing to do inductions or C-sections when it’s obvious that babies would do better on the outside, but I’m talking about when they are done for non-medical or quasi-medical reasons. I’m also talking about the possibility that fewer women end up with these medical concerns if they choose midwifery care. But of course, to look at this, would require a study which looks at women at the start of a pregnancy, matching women in as many characteristics as possible, to ensure a “matched cohort” for doctors and midwives. It would be interesting to find, for instance, that the rate of preeclampsia was 1% in the midwife group while 5% in the doctor group, even though the women had identical risk characteristics at the beginning of the pregnancy.
I do have a word more about the “37+ weeks of gestation.” You will find when you go to the CDC query that the gestational ages are broken down into several groups, including “37-39,” “40,” “41,” and “42 weeks and above.” Term is generally considered to be 37-42 weeks, and most women will naturally give birth at some point in that time. Obviously, there will be preterm births, but generally if a woman makes it to term, she will give birth before the end of 42 weeks. Only about 10% of women, will naturally go longer; in 95% of these cases, without a problem. However, placentas will not last forever, and in some cases, their function decreases so that babies are put at risk, due to the lack of blood flow — primarily oxygen, but possibly also nutrition. Some have suggested that part of the reason for going into labor is some mother-baby signal that the baby has outgrown the placental supply and needs to be born. Brief labor lesson — when the uterus contracts to dilate the cervix, no blood exchanges between the mother and fetus — the baby lives off of the oxygen currently in the placenta. If the oxygen level is already low or failing, then the baby could suffer. Consequently, many midwives will automatically transfer care (either from their own protocols or because of legislation) to doctors, because the woman is no longer considered “low-risk.” Since all births from 42 weeks onward (and I’m unsure if that’s the start or the completion of 42 weeks of pregnancy) are categorized together, it’s possible that some of these births happened even at 44 or 46 weeks of pregnancy. But I would point out that there are 9 cases of babies dying from “extreme prematurity” in the 42+ weeks of pregnancy group, 2 of which supposedly weighed 500-749 gm! I think somebody made a clerical error, although it’s also possible for some babies to have been induced or sectioned due to being over-due when in fact they were premature (the remaining 7 babies’ births were all attended by MDs).
It’s a little difficult looking at mere birth and death certificate data to get a concrete understanding of what happened in these cases. In some cases, the cause of death may not be due to the place of birth at all, but rather due to outside causes (such as car wrecks), or would have happened regardless. Genetic defects and congenital abnormalities such as anencephaly and thanatorphic short stature have a near-100% neonatal death rate, regardless of care. Other defects and abnormalities may have a high survival rate, but it is still possible to be lethal — some terms comprehend a wide range of abnormalities, from mild to severe. Take Down Syndrome, for instance. Most of the time it is non-lethal, but sometimes the heart is so badly affected that the baby cannot survive. It is also possible that some women who choose to give birth at home will also refuse to go to the doctor for any reason, including religion, fear, or sheer willfulness. It is indeed possible for a midwife to attend such a woman, and urge her to take her baby to the doctor because something is wrong, but the mother will refuse. Such a death would go against the midwife in these statistics, even though she did everything right. Also, if a baby is kept alive by a hospital until the 28th day (just past the neonatal period), it will not count as neonatal mortality… but the baby will still be dead.
Looking at the “other midwife” statistics then for 2003-2004 — for white women age 20-44, at 37+ weeks of gestation, with birthweights of 2500 grams or more, there were 29 deaths out of 29,493 births, or 0.98/1000. (I’m not sure where Dr. Amy got 1.15 — if you can figure it out, let me know the exact criteria.) Taking a closer look at the 29 deaths, we see that 19 of them (those with a cause of death code starting with “Q”) were due to congenital malformations and genetic defects. Back when I was still commenting on her blog, I pointed that out in a discussion on the Johnson & Daviss homebirth study, and she said that you can’t take out these deaths because most such babies survive in the hospital. That may be true — it may be that babies with such defects (heart problems, for instance) may survive with immediate high-tech care or perhaps surgery. But what is definitely reality, is that many home-birthing women (such as myself) decline all prenatal testing including ultrasounds and amniocentesis, which help to identify many affected babies. This has a two-fold effect — first, if immediate care will help affected babies, then that might affect the death rate of babies with severe problems; and second, and more important, women who refuse all such testing will not have abortions, which means that more of these babies will be born. It is a fact that women who find out they have a baby with a genetic or congenital defect are counseled to abort. Some of them will have an abortion– even among those who are nominally pro-life. Women who do not know they are carrying an affected baby will not have a reason to consider an abortion; therefore, more home-birthing women will give birth to an affected baby — without regard to their stance on abortion. Also, some women who know they are carrying an affected baby may be more likely to have an abortion if the prenatal diagnosis is that the defects are severe, as opposed to mild. Women who do not know that their babies are affected — perhaps severely — will not seek an abortion, as opposed to a certain rate of women who do know; therefore home-birthing women will be more likely not only to have a baby with a genetic or congenital defect, but also to have a more severe case, one more likely to end in neonatal death.
Taking out those deaths, we are left with 10/29,493. Looking at these deaths, we see that 4 occur in the 42+ weeks (which I’m near-100% certain is a higher-risk group, for reasons stated above). Three were due to “birth asphyxia” while one was due to an unspecified “complication of labour and delivery.” Taking out these higher-risk births, there are 6 deaths out of 26,496 births, or 0.23/1000, which is identical to the CNM rate (taking out deaths due by accident or homicide).
It is tempting to restrict deaths to those due to infection; pregnancy, childbirth and the puerperium; “certain conditions originating in the perinatal period”; and symptoms, signs & abnormal lab & clinical findings. For instance, one of the babies died of “diaphragmatic hernia without obstruction” — was this due to his or her being born at home? Somehow I doubt it. But if a baby dies of lung problems which may have been exacerbated by having been born by C-section, then that ought to be counted. Yet, to be fair, we will look at just the above causes of death. After all, hospitals are where sick people go; it’s where MRSA resides. People get hospital-caused infections all the time, and sometimes babies die of them in the neonatal period. While that doesn’t necessarily mean that it was due to their births being attended by doctors or CNMs, they would not have been exposed to those infections at home. This takes out 2 deaths from the midwife-attended group (because it takes out the above hernia, as well as a case of pneumonia), lowering the death rate to 0.15/1000. The CNM rate becomes 0.21 and the doctor rate becomes 0.25/1000.
Anyway, this is just playing with the numbers — I’ve provided the links so you can satisfy your curiosity, if you are curious. If you would like to take the time to look at the causes of death for hundreds and thousands of babies, you can look at the specifics of CNM- and doctor-attended births. I’m not saying that most or even all of these deaths were preventable — some things just happen. Some babies have lethal defects. If you want to look at anything, you are free to do so. If you think I’ve fiddled with the numbers too much, you can change it to just what you want.
Filed under: birth choices, studies & stuff Tagged: | amy tuteur, babies, baby, cdc statistics, childbirth, dr. amy, dr. amy tuteur, home birth, homebirth, infant mortality, labor and birth, midwife, midwifery, neonatal mortality, pregnancy, pregnant, statistics