Thoughts prompted by having a baby
Altriciality, creationism, moral non-naturalism, confirmation bias, experimenting on pregnant women, elective c-sections, the best pregnancy books, Seamus Heaney.
I have a baby! Little Ernie is objectively the cutest baby in the whole world, and I will not listen to any slanderous lies like ‘there are cuter babies’ or ‘you’re just blinded by personal attachment’ or ‘Peter you should probably rewatch the Seinfeld episode where they go to the Hamptons’.
Having a baby prompts many thoughts. Some of them are just ‘awwwww, wook at da widdle baby!’, but some of them are more substantial. I found as I was writing my ‘interesting things I read or thought about in November’ post that the vast majority of entries were baby-related; so I have separated out all the baby thoughts and put them here. This is one of the reasons why you should have a baby: they are fucking fascinating little things, kind of the anti–Grecian urn, in that this extremely loud form will tease you into thought.1 I would be keen to hear from other parents about what thought their kids have teased them into.
All babies are premature
The most important thing to know about human babies is that, when they come out, they are not done.
Most other mammals have precocial babies. The babies are fully ‘cooked’ in the womb, meaning that when they come out they’re basically just small versions of the adults of the species. Think about calving or lambing if you’re familiar with those, or (if not) look at this video of a giraffe birth for an example: the baby basically just falls out, and is up and walking around within five minutes.
But some other mammals have altricial young, which are very much not able to take care of themselves after birth. You are probably familiar with newborn puppies and kittens. The ‘purest’ altricial mammals are probably the marsupials like kangaroos and wallabies, whose babies are so unfinished that they literally have to live in a pouch. Here’s a photo of what a wallaby baby (‘joey’) looks like; obviously that thing can’t take care of itself!
It should be quite obvious that human babies can’t take care of themselves. But we’re weird. Our babies are more developed at birth than most altricial species: still not developed enough to be precocial, but compared with other altricial species we’re investing a lot of effort and resources into gestation. Other altricial mammals spend only a short time in the womb and so usually have non-functional sensory organs, often can’t thermoregulate, and in general are basically foetuses when they come out. Humans, however, invest a huge amount of resources in babies before their birth: human babies spend a long time in the womb, are well-developed in many respects, can thermoregulate (to a certain degree) and have functioning sense organs, etc. But that investment doesn’t seem to ever pay off, and our babies are still helpless. We don’t look like a normal altricial species; we look like a precocial species where (for some reason) all our babies are born prematurely.
And indeed, according to the leading hypothesis, this is exactly the case. Our ancestors seemingly had precocial babies. But we evolved to have a combination of big skulls (for our big brains) and small pelvises (for walking upright), meaning that if our babies stayed in the womb until they were pretty much done, they’d literally just not fit out: their head would be too big for the mother’s pelvis. And rather than reinventing the wheel and changing our whole reproductive strategy to be properly altricial, evolution just said ‘fuck it’ and triggered childbirth early.
Obviously, childbirth can’t be triggered too early, or the baby would never survive even with parental care. We have our babies when their heads are just barely able to fit through the pelvis, and thus—as the Lᴏʀᴅ himself once said—‘in pain you shall bring forth children.’ (Thanks to a failed epidural, Rachel has now experienced this particularly acutely.) But this isn’t enough cooking, and so all human babies are essentially premature. (What we call premature babies are, in a sense, doubly premature.) The evolutionary strategy of Homo sapiens is to just hope that humans will figure out ways to allow at least some of our premature babies to survive.2
This is the most important insight I have kept coming back to to understand what Ernie is doing. For the first few days, for example, Ernie wasn’t taking enough milk. We eventually figured out that he his suck reflex wasn’t super strong: although he was latching on ok, he just wasn’t sucking enough. (Bottle-feeding fixed this because we could see how much milk he was taking in, and adjust positioning if we saw he wasn’t getting enough.) Such difficulties in breastfeeding are not uncommon, as a quick perusal of parenting subreddits will make clear. In situation like this, you begin to ask yourself: how did people manage this before bottle feeding? The answer is: they didn’t. Sometimes babies are born before they’re ready to take milk; humanity didn’t evolve a solution to this problem for any individual case, our evolutionary strategy is just to hope that things work out in aggregate.
Likewise, Ernie’s breathing is super weird sometimes. This is eminently normal: all newborns breathe in irregular patterns (the kind of irregular breathing that would be a terrifying symptom in adults), sometimes they make extremely concerning noises, and sometimes they just forget to breathe altogether for a few seconds. Why? Because their respiratory system isn’t finished yet! Or: Ernie is very bad at regulating his own hunger, he will gorge himself on too much milk and then sick half of it up; us trying to enforce breaks or burping doesn’t help. The connections between his digestive and nervous systems aren’t done. Final example: babies’ skulls aren’t even done yet! They have holes in their skulls! They need to finish growing their skulls outside the womb.
On the one hand, this is a reassuring thought, because it helps me put a lot of observations into a helpful framework and make sense of them, meaning I feel less lost at sea with my little guy. On the other hand, it is a framework that keeps the risk of infant mortality front and centre: there’s no biological guarantee that babies are ‘designed’ for this, that if you do everything right that your baby will be ok. Thankfully, living in a rich western country, I can feel pretty confident that Ernie will be ok because of to centuries of medical progress. But not everyone has access to the fruits of that progress.
Creationism bad
Anyway. I feel like an authentic dad already, because I’m already finding myself taking my experiences to be evidence for whatever my pre-existing opinions were.
For example, I found myself saying to Rachel that I don’t know how you can be a creationist if you have a baby. I use the term ‘creationist’ broadly, I don’t just mean the six-days-6000-years-ago people: I mean anybody who believes in the doctrine of creation in its widest sense, including Catholics and those Protestants whose experience with Darwin’s legacy has led them to learn the virtues of equivocation a few centuries late; and indeed including Muslims and Jews.
My point is that human babies are badly designed. They suffer and die at appalling rates, they have all sorts of random unfinished bits, they literally have holes in their skulls, they’re basically stuck together with sellotape. Therefore, they must be the product of some process that is kludgey and constrained and maybe semi-random—to put it another way, a process that relies on bodges. An infinite intelligence with no constraints creating something in its own image would not rely on bodges; QED. An appeal to ‘guided evolution’ doesn’t help you, because the explanatory power of evolution in this case depends crucially on the fact that it’s unguided, going to the local maximum rather than the global one: we don’t have an optimal altricial reproductive strategy, we invest a lot of resources into babies in the womb, but then they come out helpless anyway.
Christian creationists might appeal to the Fall (maybe human babies were originally optimally designed, but then we sinned) except that the ways in which babies are suboptimal don’t correspond to any kind of punishment. Indeed, quite the opposite, the bad bits of baby design seem to have a close link with our ‘very good’ intelligent and rational nature (big head → altriciality). In response to this, you could basically say ‘the Lord works in mysterious ways,’ and insist that we are not very good at interpreting the way created things reflect the act of creation; but that wouldn’t be very Romans 1 of you!3
Of course, Rachel points out that only someone who already believed what I believe would interpret the experience of having a baby in these terms. Most people handed a baby would, in the first instance, be in awe of this amazing bundle of joy and how wonderful it is, and would only secondarily (if at all) reflect on the various bits of its design that are suboptimal. The awe might well prompt them to give thanks to a creator. I am also in awe of my amazing bundle of joy, but already before Ernie was born I was convinced that ethical worth and beauty and sublimity can result from blind gradient ascent in trait space. If you didn’t have that belief when your baby was born, it’s unlikely that you’d suddenly be converted with one look! Of course, I know that I am right and the creationists are wrong; but I must resist the dad-coded urge to act as if I’m right because I’ve had a baby, when really I’m right because of mountains of biological evidence.
Metaethical non-naturalism bad
Likewise, I have been thinking about my experience with Ernie in light of something my friend HenryShevlin once said. When I take care of Ernie, I am filled with awe and joy and love. For example, I have now experienced the magical thing all parents make reference to: he has vomited on me and indeed shit on me, yet my love for him is somehow so great that it doesn’t even register as a bad thing at all, not even mild annoyance. It’s incredibly special, goes way beyond any other kind of love I’ve experienced. (Yes, my childless friends, I am becoming that type of annoying arsehole.)
But phenomenologically speaking, this magical love doesn’t feel like anything qualitatively new—as strange as it may sound, it’s not a brand new emotion, just a rearrangement of emotions I already had. I’m not noticing anything about him beyond his the biological facts of his life, his pissing and shitting and vomiting and eating and and breathing and his jerky little movements.
Further, the awe and love are not transcendent or qualitatively unique, and they are compatible with boredom and stupidity and all the usual mundane feelings. (When Ernie first shat down my arm, my first thought was ‘just like Jay’s Dutch bird!’, a completely inappropriate reference that it is would have been advisable to keep entirely to myself, but instead I am putting in on Substack for illustrative purposes. How lucky you are, my readers!)
To put this in the way inspired by Henry: when I look at Ernie, I’m not perceiving any special non-natural moral properties, over and above ordinary perception of natural properties, and natural dispositions to react to certain natural properties.4 But I am loving him: so there can be nothing to love over and above what I am experiencing. So love is a purely natural matter, and so non-naturalist moral realism is false, QED.
But of course, the way I have described the phenomenology here is extremely question-begging. What I think I can say is that, given that I already knew that non-naturalist moral realism was false, becoming a father has made it clearer to me exactly how a complex arrangement of purely natural dispositions to react to perceiving purely natural properties can give rise to a complex and ethically-laden emotion like love. But a moral non-naturalist could definitely strain to describe the phenomenology differently and ensure a place for perceiving non-natural properties; and they might even have a story that is more prima facie plausible when it comes to the phenomenology of mother’s love (as opposed to father’s love), because that often does seem qualitatively new. I can explain mothers’ experience away by saying ‘hormones’, but this explanation isn’t a priori obvious, it rests on a whole bunch of prior assumptions about human psychology that I think are true but which require argument.
Confirmation bias bad
So yes, having a baby doesn’t cut the Gordian knot of longstanding philosophical or theological disagreements. At most, it gives you an extra few datapoints. But what has struck me is how easily I assumed that my experience did cut the Gordian knot! I really did find myself saying the words ‘I don’t know how you could be a creationist with a baby.’
When you’re a dad (or a mum), you only have a few extra datapoints, but they feel extremely important. (They probably are extremely important.) Confirmation bias loves when you feel like this, and it gets to work right away, shaping the way you interpret your experiences to confirm your pre-existing opinions, but making you think that you’re just straightforwardly learning the lessons of these incredibly important datapoints. In my case, because of what I’m interested in, the pre-existing opinions in question were philosophical and theological; but for many people they’re more practical or down-to-earth. This is why so many people—me included—have kids and suddenly start claiming epistemic authority they have no right to, telling you stuff like ‘you’ll understand once you have kids’. This is completely infuriating for the listeners, who know full well that the person hasn’t impartially processed the objective truth about having children!
The problem is that, indeed, I have experienced something that non-parents haven’t; I do have access to (Bayesian) evidence they don’t have. It is therefore entirely right to expect that non-parents—and, when he grows up a bit, Ernie himself—should listen to what I have to say: it is simply true that I know things they don’t. But the minute I got the new evidence, confirmation bias got to work straight away; and when I get around to talking about this stuff, the evidence has already been twisted horribly to fit my pre-existing opinions.5 So non-parents are also entirely right to think that I am being a horrible and entitled bore, declaring my opinions as if they are gospel when really they’re just my opinions. And, almost inevitably, when Ernie’s a teenager he will rebel against my (epistemic) authority, and not for no good reason.
This is in many ways the problem of human reasoning. Information is local, different people have experienced different things and know different things, we really do need to be able to share our experiences, learn from each other, and (more generally) take part in a division of knowledge. Reasoning might even be inherently social. And yet—by the time anyone gets around to sharing the information they have taken in, it’s been so thoroughly twisted and mixed up by their own perspective and biases. When someone tells you their opinion, there’s often no way to know what parts of it reflect genuine information that you just don’t have access to, and what parts reflect contingencies of their perspective or even just straightforward reasoning errors. It’s slightly crazy to me that this deep and fundamental problem of social epistemology has hit me so hard and so quickly after I’ve become a parent.
Elective c-sections and risk-aversion
Last month Leah Pierson in The Argument declared that we need to experiment on pregnant women. She is completely correct. Much of the data we have on the effects of various drugs and interventions during pregnancy is, to use the precise technical term, total fucking shite, because generally pregnant women aren’t included in clinical trials.
The conventional wisdom, though, is that pregnant women are incredibly risk-averse and would not be willing to take risks with their babies. This is taken to mean that, even were scientists more open to experimenting on pregnant women, they’d find few pregnant women willing to be subjected to experiments.6
This conventional wisdom is definitely grounded in something! Many (not all, but many) pregnant women are indeed extremely risk-averse, and extremely worried about the health of their child in many decisions, to the point that it sometimes impacts on their own health.7 But the question is, what is this attitude grounded in?
One explanation might be that it’s innate and biological: evolution makes pregnant women act to protect their child at all costs to ensure their genes are passed down. Another explanation might be that it’s deep culture, the patriarchy socialising women into the role of ‘mother’ and teaching them to internalise the idea that their child is more important than them. Without being so bold as to completely dismiss evopsych or feminism entirely, I want to offer a different possible factor.
In order to rationally make decisions under uncertainty, you usually need to know how likely the risks are associated with your options. I know that when I take a plane journey I could maybe die or could maybe survive, and that if I jumped off the top of a tall building I could maybe die or could maybe survive. But in the case of a plane journey I know the risk is extremely small, so I’m happy with flying; whereas in the case of jumping off a tall building, it’s a very real risk that I die, so I don’t do that even if it would be fun. In the extreme case where you have precise probabilities, you want to make the decision that maximises the expected value of your ‘utility function’.
But what about the other extreme case, where you literally have no idea abut the level of risk you’re facing? You know that something might go wrong, but you don’t know if it’s basically guaranteed, basically impossible, or anything in between. In cases like this the normal approaches break down. You instead need to use crude heuristics or radically different decision-making procedures. John Rawls suggested that, in situations of such truly profound ignorance (such as his famous ‘veil of ignorance’), you should choose the option where the worst possible outcome of your choice—however unlikely—is less bad than the worst possible outcome of any other choice you could have made: you aim to avoid the absolute worst possible outcomes, ignoring both less extreme bad outcomes (irritations and frustrations), and any possible upside.
Pregnant women’s risk aversion is quite like Rawls’ ‘maximin’ rule! I’m not saying they’re exactly the same (for one thing, Rawls says nothing relevant to the question of how you should trade off risks to the baby vs risks to the mother), but I am suggesting that they play the same role. If you have no information about the level of risk, you can’t make rational decisions in the normal way, and have to follow some other guideline; radical risk-aversion is one.
And pregnant women, in general, aren’t given information about the level of risk. I now speak from (my wife’s) experience: the way that obstetricians and midwives talk about risk to pregnant women is extremely obfuscatory, misleading, and in many cases actively adversarial (they aren’t providing information in an attempt to be helpful, but rather in an attempt to cause you to make one choice rather than another.)8 They’ll sometimes tell you what the possible bad outcomes are—though sometimes they won’t tell you! if they want you to choose a particular option, they’ll often just not tell you about the downsides!—but in general they’ll not tell you how likely those bad outcomes are.9
My hypothesis is that, in a large proportion of cases, risk-aversion from pregnant women is neither a biological imperative nor a deep-seated internalisation of patriarchal attitudes, but rather is a sensible and adaptive tactic in an extremely low-quality (even adversarial) information environment.
This hypothesis predicts that, if there were an choice where pregnant women had an independent source of information about the risks, they would be much less risk-averse. We have such a test case: elective c-sections. Independently of becoming pregnant, you know a lot of people who have had c-section or who were born by c-section. Because it’s so common, the number of people you know born by c-section gives you a bound on your estimate of the risk of c-sections.10 This is an independent source of evidence. It’s not a perfect source of evidence (certainly nothing like a precise probability of stillbirth or anything like that); but it’s much more than you get in other cases, where the interventions are much rarer, and so it’s very possible that there’s nobody else who you know has undergone it. Because there is a significant amount of public information, midwives and obstetricians can only obfuscate the level of risk from c-section so much. This level of risk is quite low, but it is not zero: c-sections are generally much riskier than vaginal births, in terms of mortality risk both for the mother and for the baby.11 (It’s major surgery! They open the mother’s abdomen up and start rooting around in there!)
Sometimes, a c-section is medically indicated because of problems that might make a vaginal birth more difficult, but there’s also such a thing as an ‘elective’ c-section where women choose c-sections just because that’s what they want. So we have a test case. If risk-aversion is deeply internalised, whether biologically or culturally, pregnant women would still avoid elective c-sections. But if risk-aversion is at least sometimes an adaptive response to a lack of information, then in this case where women have more information, we’d expect them to give up the adaptive response, be less risk-averse and choose elective c-sections more often. Which is it?
I think the evidence somewhat favours my model! Elective c-sections have become way more popular in recent years, skyrocketing to around one in five births in the UK.12 This makes no sense if the roots of risk-aversion are deep-seated; it makes more sense if risk-aversion is at least partly just an information problem. Importantly, rates of elective c-sections have tracked rates of emergency c-sections: this doesn’t make a huge amount of sense on the face of it (we don’t normally choose to undergo medical procedures that have become more common in emergencies), but does make sense if you think of it in terms of ‘how many people do you know who have had a c-section?’13
More and more women are looking at the risks associated with c-sections, and because they have decent information about these risks, they are able to weigh them up against the benefits. And so they do weigh up the risks and benefits, and some of them decide to go for an elective c-section, because it’s convenient or avoids pain or means their vagina won’t be torn up. (I actually think that for most people this is probably the wrong choice; but that’s a different matter, and who am I to say, not ever having gone through it myself?)
My view is that that risk-aversion is at least 30–60% an adaptive preference in the face of the medical profession refusing to explain risks properly to pregnant women. Maybe the other 40–70% is deeper-seated; but even moving the needle on that 30–60% is enough to get clinical trial participants!
The Birth Partner and other recommendations
Objectively the best book about birth—indeed, the best book about any aspect of pregnancy?—is The Birth Partner by Penny Simkin (recent editions also feature contributions from Katie Rohs). It is so good that, even though it’s designed for husbands / wives / partners and not the pregnant woman herself, Rachel regularly grabbed it off me instead of the books that were actually intended for her. It’s respectful, smart, full of detail, pitched at the right level—unlike the vast majority of people writing on this topic, Simkin assumes that you’re an intelligent adult who knows what biology is!—and importantly, really clear on when you should stand your ground for your partner against doctor and midwife pressure, and when you should fuck off. (For example: if there’s a cord prolapse, shut the fuck up and stand the fuck back.)
To put it in slightly different terms: Some people don’t hate doctors, which is wrong, doctors are paternalistic and condescending and overconfident and don’t understand statistics. But other people hate doctors too much and are anti-scientific cranks. Simkin seems to hate doctors exactly the correct amount, which is rare and impressive.
My only criticism is that, when it comes time for her to discuss pseudoscientific guff like reiki and acupuncture, her tone changes entirely and she circumlocutes, saying stuff about how some women find this helpful. But I understand that that’s just what you have to say when writing pregnancy books: the audience believes in pseudoscientific guff at rates far above the population as a whole,14 and you don’t want someone to ignore your advice about induction or c-sections just because they’re offended by the truth about homeopathy. And this is only a couple of paragraphs across the whole book; otherwise, it’s brilliant. (Easily-offended Goodreads reviewers will tell you to get the fourth edition because the fifth edition is woke, but ‘woke’ in this case just means ‘not extremely awkward about the existence of lesbians’, so you’ll probably be fine.)
But, in spite of all of this, I have to confess that even with the book I really was not a great birth partner. You may take this to be a mark against Simkin, but I take it to be a mixture of my own failings and of the limitations of booklearnin’. Rachel and I considered hiring a doula for the birth, because the evidence base is extremely good that they improve outcomes; but when we looked at possible options we were extremely put off at the fact that almost all of them near where we live were extremely woo-woo mystical ‘listen to your body’ types. Rachel reasonably concluded that she wanted that energy to be as far away from her as possible in her moment of greatest stress. But it definitively would have been helpful to have someone in the room who was unequivocally on Rachel’s side (i.e., someone who didn’t work for the NHS) and who had seen a birth before—ideally more than one. I read as much about it as I could, but this is a ‘what Mary didn’t know’ case where actually having seen it makes all the difference.
In terms of other resources: Rachel is a huge fan of Emily Oster, and pays for ParentData; I have slightly more complicated feelings about Oster (as might be suggested in my extremely overcompressed and hard-to-understand Goodreads review of Expecting Better) but she is unquestionably a better source of information than the comparable alternatives, such as ‘government websites’ and ‘your wife’s midwives’. She is also the person who taught me about my single favourite example of confounding: studies that suggest a link between caffeine consumption and bad outcomes in pregnancy fail to account for the fact that serious morning sickness is an indicator of good outcomes, and if you’ve been throwing up all morning you might not have the stomach for a strong cup of coffee!
I also enjoyed this podcast episode, which features parenting advice from Oster, Ezra Klein, Bryan Caplan, and also a bunch of Effective Altruist types: some of it is quite weird, but it’s good fun.
Most pregnancy and parenting media is very bad, unfortunately.
Baby poems
Further to that, there are surprisingly few poems about babies? There’s a feminist point in here somewhere, that male poets paid little attention to what was happening in the ‘female’ world of pregnancy, birth, and childrearing; but still, meeting a little baby is an experience that is both close to universal and extremely emotionally affecting, and even taking into account the feminist point, you’d still think it’d be more common in poetry. To be sure, it’s becoming an increasingly common topic in more recent poetry, but that’s not yet made up for centuries of neglect.
Still, if I had to pick from the small number I was able to read, my favourite is Seamus Heaney’s ‘Route 110’. It’s a complex pastiche of Aeneid Book 6 with the underworld replaced with rural County Derry (not so different…). Heaney is travelling to meet his new granddaughter, who is one of the souls on the banks of the Lethe, ready to be reborn. You should read the whole thing, but here is the last section, which made me well up when I read it with Ernie:
And now the age of births. As when once
At dawn from the foot of our back garden
The last to leave came with fresh-plucked flowersTo quell whatever smells of drink and smoke
Would linger on where mother and child were due
Later that morning from the nursing home,So now, as a thank-offering for one
Whose long wait on the shaded bank has ended,
I arrive with my bunch of stalks and silvered headsLike tapers that won’t dim
As her earthlight breaks and we gather round
Talking baby talk.
As doth time? Warm cityscape!?
I was briefly tempted here to link the ‘some of you may die, but that is a sacrifice I am willing to make’ meme, but I made myself look at these statistics and it sobered up my attitude very quickly.
For Muslims: not very al-Jathiyah 3-5 of you.
Henry’s point was not about having a child, and I don’t know if he’d agree with me about the experience of having kids. His point was about what it was like to visit Auschwitz, expecting to ‘feel’ the evil in a special way—perceive the non-natural property of ‘genocide’—when in fact it’s just a load of brick buildings, and the knowledge of the evil that was perpetrated there isn’t much like perception at all. (I know that Henry is under even fewer illusions than me about this being a knock-down argument against non-naturalism.)
This is basically my explanation of why the Aumann agreement theorem doesn’t apply to human reasoners.
Probably even if this is true, we should still change our approach: allow pregnant women to sign up for clinical trials if they want, and if none of them want to, well that sucks but at least we tried. If the conventional wisdom is right, then the policy of not allowing pregnant women into clinical trials is pointless, because they wouldn’t sign up regardless; and if the conventional wisdom is wrong, then the policy is actively harmful. See also Pierson’s article.
The clearest case where the information relationship between pregnant women and the NHS is purely adversarial is actually post-birth, and relates to breast vs formula feeding. When Ernie was struggling to feed and had lost a huge amount of weight in the first couple of days, we had NHS staff make it quite clear to us that they could get in trouble for even suggesting the possibility of formula feeding to us.
There are a few exceptions for especially involved medical procedures (amniocentesis, epidural, etc.) where they are essentially legally forced to give you better information about the risks. But in these cases, the way they communicate risk is by giving you a piece of paper with estimates written down on it to satisfy the lawyers, while simultaneously talking to you about the risk. And Rachel’s experience has been that, whatever the information written down on the paper, the tone of what they say to you isn’t ‘here is what these numbers mean’ but rather ‘I can’t believe you’re taking this risk with your baby, you bitch’.
Specifically, you can use Laplace’s rule of succession to get an estimate. Not that the average pregnant woman is actually using Laplace’s rule of succession; but my point is only that they have the information that can be used to get an estimate for the level of risk, and are able to use that information as the basis of rough heuristics and ‘vibes’ about risk.
Though see here for a meta-analysis that found no difference in the rates of baby death between planned c-section and planned vaginal birth, and indeed found a lower rate of baby death in c-section when excluding twins / triplets / etc. However, the trials that are included in this meta-analysis are not fair comparisons: a c-section was somewhat indicated for all of the women included in the trials, in both the c-section and vaginal birth groups. Mostly this was because the baby was breech (upside down in the uterus). That is to say, the studies in this meta-analysis are mostly comparing c-section birth with breech vaginal birth, telling us is that the risk of baby death from a planned c-section is comparable to the risk of baby death from a breech vaginal birth. But we already know that breech vaginal births are really quite risky, much riskier than non-breech vaginal births! I conclude that c-sections are more likely to cause baby death than vaginal births unless there’s some complication like breech. Alas, there is no better study / meta-analysis comparing c-section risks to the risks of a normal vaginal birth, because (as Pierson’s article suggests) we don’t experiment on pregnant women enough.
18.7% is the English figure; I am not arsed right now to get all the incommensurable figures for the four nations and try to combine them. I asked Claude to do this and it told me ~17–19%; I haven’t checked its working but that sounds about right given that England is the vast majority of the UK.
You might suggest that there’s a possible confounder here: c-sections have become safer, which causes doctors to be more confident about ordering them in emergencies and causes women to look upon them more favourably as elective procedures. But note that ‘c-sections are safer’ just means that the level of risk of maternal death and stillbirth have gone down, not that these risks have been entirely eliminated. If pregnant women really were extremely risk-averse in making decisions about c-section, then ‘c-sections have become safer’ couldn’t be a confounder, because its impact on women’s decisions would be marginal. So either way it seems clear that decisions about elective c-section are not being made on the basis of extreme risk-aversion.
Probably because, as mentioned above, they’re in a low-quality and adversarial information environment!
Congratulations! This was perhaps the most Peter way to announce the birth of your first child.
Do you think the adversarial relationship between the expecting and the health system is particularly acute with the NHS, or part of the nature of the beast all over (not an imminent concern, as I am still not married)? Also, I very much enjoyed the experience of getting riled up over your take-downs of my philosophical beliefs, only to have the wind taken out of my sails by the last paragraph of the section
Very late to this, but congratulations!