Tags
[Content warning: brief discussion of weight gain; lots of discussion of food in pregnancy.]
There are a lot of books and websites in the genre “Mom Without Medical Background But With Quant Background Does Lit Reviews About Pregnancy.”
If you take the this seriously, a lot of pregnancy advice is useless or actively counterproductive. For instance, pregnant people are regularly advised not to change kitty litter. But– they argue– in reality, toxoplasmosis infections are almost always due to undercooked meat and unprotected contact with soil. Transmission due to cat litter is extraordinarily rare. An infected cat can only transmit toxoplasmosis for two weeks, and it’s unlikely those two weeks would happen while you’re pregnant. And if you change the kitty litter on anything approaching a regular schedule you’ll change it well before you can catch toxoplasmosis by touching poop. (Debunking the Bump, chapter one; Expecting Better, chapter nine.)
Or take weight gain. Pregnant people often worry about gaining too much weight. But gaining a little bit too much weight during pregnancy isn’t correlated with any particularly interesting outcomes, while gaining a little bit too little weight leads to low birth weight, which leads to every negative consequence imaginable. It’s much, much, much better to err on the side of gaining too much weight than too little. (Debunking the Bump, chapter four; Expecting Better, chapter ten.)
Fish consumption is the single most embarrassing example. Many pregnant people limit their fish consumption for fear of mercury poisoning, which lowers IQ. But eating fish in pregnancy is correlated with an increased IQ in children if you eat up to eight servings a week– well above the FDA’s recommended level of fish consumption. This is probably because fish is rich in omega-3 fatty acids. (Debunking the Bump, chapter two; Expecting Better, chapter six.)
And it goes on and on like this! For pages and pages! Which leaves me, a prospective parent, with the question: why is all pregnancy advice so bad?
I have thought of three possible explanations.
First, maybe random mothers with quant backgrounds are an unreliable source of information about pregnancy health. Perhaps I should pay attention to the government, my doctor, and What To Expect When You’re Expecting, instead of women with no medical background and access to Sci-Hub and Google Scholar. But when I’ve spot-checked the books they’ve generally been pretty accurate. And they mostly seem to agree with each other about what the data says, even though as far as I know they aren’t talking to each other. Most of the time, when people are making up random bullshit, they tend to make up different random bullshit (see also: nutrition advice). Also, I don’t like this explanation because Emily Oster says I’m allowed to eat sushi and What To Expect says I’m not and I don’t want to go through my nine mouths of pescetarianism sushiless.
Second, maybe all medical advice is like this. The general quality of information about medicine in general is exactly the same as the general quality of information about pregnancy, it’s just that I’ve bothered to actually look into pregnancy. This is very worrisome. What if this whole time I’ve been exercising because I thought it would make me live longer and actually I’m shortening my lifespan? What if I have been trying to cut down on neon-colored high-fructose-corn-syrup snacks for absolutely no reason? What if the true secret to fixing depression was doing handstands every day to encourage blood flow to the brain, and all the “fix your depression with yoga” people have been right all along?
Of course, the very most important pregnancy advice is generally accurate: it is really really bad to smoke or to drink to excess, and it’s much better to have a pregnancy before 35 than after 35. And a lot of the research is relatively new. We only started figuring out how pregnant people got toxoplasma gondii in 2000, and before then it was “like idk cats probably?” The FDA found that fish consumption was correlated with higher IQ in 2009. Folic acid was really really important before mandatory food fortification began in 1998, and it took a while to realize that very few people are folic acid deficient anymore. Perhaps it takes a while for this information to filter down to the general public, and in thirty years pregnant people will devour two plates of salmon and then go change the cat litter. So maybe the take home here is that medical consensus is correct about really really important things, but tends to be slow to adopt new information. (Which might be a good thing! There is, after all, a replication crisis.)
Third, maybe pregnancy advice is unusually bad for some reason. For instance, people do seem to be unusually risk-averse about pregnant people compared to other groups. So maybe people come up with recommendations (“don’t eat fish! don’t change cat litter!”) based on relatively weak evidence, because what if a bunch of women get mercury poisoning and then we have a huge epidemic of mercury babies? And when it turns out that cat litter or fish is actually fine, people tend to keep giving advice: what if the study was wrong? They want to err on the side of caution.
I think that risk aversion plays a nonzero role: for instance, I expect that’s behind folic acid supplementation (best to be on the safe side about whether the pregnant person is eating enough Cheerios) and telling women to avoid sushi (parasite infections from sushi are very rare but they do happen). And it probably even explains some of the weird downplaying of risk: if pregnant people are told to avoid both unpasteurized cheese and deli meat, without a clear explanation that deli meat is far far more dangerous than unpasteurized cheese, they might forget that they’re not supposed to be eating deli meat. But I’m not sure that risk aversion fully explains the fish thing or the weight gain thing.
Paul Crowley said:
Say a little more about drinking? I’ve heard it argued that while heavy drinking is very bad, the evidence that moderate drinking is bad is very poor.
LikeLiked by 2 people
ozymandias said:
No one has shown any negative side effects from consuming one to two drinks a week in the first trimester or up to one drink a day in the second and third trimester, ideally sipped slowly (no shots!) and taken with food. There is a good deal of disagreement about whether that’s because light drinking actually isn’t bad for your fetus or because people who drink lightly during pregnancy tend to be upper-middle-class, better educated, etc. My understanding is that following those recommendations would be a fairly significant reduction in drinking for most people who drink regularly regardless.
LikeLike
tcheasdfjkl said:
“My understanding is that following those recommendations would be a fairly significant reduction in drinking for most people who drink regularly regardless.”
I disagree with this – for me “one to two drinks a week” would be barely a reduction while “no drinking” would feel like an actual sacrifice, so there’s a big difference between these recommendations.
Thanks for the info!
LikeLiked by 1 person
ozymandias said:
I revised the post to say “drink to excess”. Thanks, nitpickers. 😛
LikeLiked by 1 person
Paul Crowley said:
I’m guessing you mean those thanks sincerely 🙂 To nitpick further, I’d be tempted to say “it is really really bad to drink to excess or to smoke”.
LikeLiked by 1 person
Molly Lynn Stratton said:
“Or take weight gain. Pregnant people often worry about gaining too much weight. But gaining too much weight during pregnancy isn’t correlated with any particularly interesting outcomes…”
Theory: society is so fat-shamey it even affects pregnant women?
LikeLiked by 7 people
Jack V said:
That was my obvious interpretation… I just assumed you were SUPPOSED to put weight on, because, like, you need a bunch more internal organs, I was shocked (though no longer surprised) that anti-weight advice had infected even pregnancy.
LikeLike
benquo said:
“All advice is terrible because the epistemic environment is hostile” seems like the simplest explanation. It matches up pretty well with things like this: http://andrewgelman.com/2017/02/09/pizzagate-update-dont-try-trick-twice/
LikeLike
Lawrence D'Anna said:
I wish we had some kind of catchphrase for “the epistemic environment is hostile”, because in a lot of domains this seems to be the number one impediment to actually getting right answers.
LikeLike
tcheasdfjkl said:
Actually what does “the epistemic environment is hostile” mean? Does “hostile” mean something beyond just “bad” here?
LikeLike
danarmak said:
I think a hostile epistemic environment means an anti-inductive one. An intelligent optimization process (mostly but maybe not only other humans) will adapt to your efforts to learn the truth and try to counteract them.
I don’t know what benquo meant, but I hazard that e.g. there are a lot of people incentivized to provide pregnancy advice (books sell, articles promote your career), they end up optimized for popularity not truth, and create market and institutional forces fighting the truth insofar as the truth wouldn’t sell as well.
This might be because the truth would be to sacrifice something harder than eating sushi; or because people are invested in advice that actually doesn’t help; but perhaps most of all because there isn’t a lot of advice to be given, pregnancy mostly just works, while there’s a big market demand for advice, especially if it can be presented as a surprising secret that will put you ahead of everyone else. Also, people who have already followed this advice and smugly told their friends about it won’t want to admit they were suckered by a charlatan.
I have no experience with pregnancy advice, but I think this description is definitely appropriate for weight-loss diets. We don’t fully understand how to do it or why many people are overweight to begin with, but there’s a huge market demand for answers, there are many careers invested in providing advice, many people whose identity is bound up with believing in certain diets, a lot of money being made from special foods and supplements and workshops and books, and all of these work against anyone promoting the truth that 1) we don’t really know how to lose weight and 2) almost all diets being promoted don’t actually work.
LikeLiked by 1 person
Aapje said:
A lot of advice also seem direction pushing, rather than target hitting, which makes it only correct for people who are on one side of the optimum.
LikeLiked by 1 person
ozymandias said:
Pizzagate seems like a different thing. Both the Quant Moms and the government/people’s doctors/What To Expect are assuming that scientific studies are, generally speaking, accurate. It seems to me like “are scientific papers generally truth-tracking?” is a different question from “is the scientific consensus, as explained to me by authoritative sources, truth-tracking?” For example, it might be that a lot of scientific studies are inaccurate, but the CDC etc. do a good job of filtering out inaccurate studies, ignoring small studies that might not replicate, etc.
LikeLike
Emma said:
“it’s much better to have a pregnancy before 35 than after 35”
Articles like this – http://www.bbc.co.uk/news/magazine-24128176 – suggest that might not be accurate either. I’d be keen to know your thoughts.
LikeLiked by 1 person
ozymandias said:
Having a baby over age 35 increases risk of stillbirth and miscarriage by a lot, and a lot of other bad things by a little bit.
LikeLike
qwep said:
Because people over 35 are fatter than under 35, the problem isn’t the age, the problem is the obesity
LikeLike
ozymandias said:
That is not true. The primary cause of negative consequences from advanced maternal age is ovarian aging, which affects thin people and fat people alike. (For example, thin people also go through menopause.)
LikeLike
Mari said:
Agree. There’s a lot of scaremongering circulating the internet about age, most of which is outdated and agenda-driven. If we weren’t meant to have children past 35 then it wouldn’t happen as often as it does. I have yet to meet a woman who has given birth in her late 30s or 40s who had issues with her pregnancy. And these women are usually ready for kids, therefore more responsible.
LikeLike
Mari said:
Sorry, agree with both Quep and Emma, not Ozymandias. Are you an expert by chance, Ozymandias, because citing a couple of studies means very little. Many are rife with flaws, are biased, or there are a myriad of other studies with conflicting results.
LikeLike
MarginalCost said:
Perhaps one explanation is that the feedback loops are very poor? Today, many women will only go through one to three pregnancies, and even for those with multiple pregnancies it can be over a decade before any conclusive evidence emerges, by which time you’ve long stopped worrying about pregnancy.
If this were true, we’d expect it to be the case with other one-time or few-time events in a person’s life. Home-buying, wedding planning, and funeral planning come to mind. None seem especially terrible, but most do tend to fall in the category of 1) things people usually regret spending too much money on, and 2) often navigated by a third party who is able to do the same activity frequently. All have shorter feedback loops than pregnancy though.
LikeLiked by 1 person
Jared said:
That’s an extraordinary statement. I thought your HAES philosophy acknowledges correlations but then insists that weight isn’t causally important.
There are plenty of problems listed on the Mayo Clinic page for fetal macrosomia (high birth weight). It should not be surprising that high pregnancy weight gain causes high birth weight, but here is a source (and note that this source is defending the IOM recommendations against criticism that they are too lenient).
Nonetheless, pregnancy advice against excessive weight gain makes sense when excessive weight gain is 2.3 times more common than inadequate weight gain, particularly since textbook advice is a healthy range rather than “gain less” or “gain more”, and so it’s doubtful that the “healthy range” advice is actually contributing much to the problem of inadequate weight gain.
LikeLiked by 2 people
Julia Wise said:
I found that even the advice for gaining more weight during pregnancy was terrible (“eat low-fat diet products.”) The advice on children’s nutrition also seems super slanted toward preventing overweight and totally ignoring underweight. Like the Women Infants and Children nutrition program won’t cover full-fat dairy products post age 2 without a prescription.
LikeLike
Julia Wise said:
* dairy products, not diet.
LikeLike
loki said:
Every one of the maternal risks is solved by cesarian, which is super common in the US anyway, so that might be why that doesn’t register as a big deal. Meanwhile, the risks to the child are things that don’t cause any harm now but might maybe be correlated with non-fatal health problems later, more research required.
Meanwhile low birth rate is straight-up strongly correlated with infant mortality.
I can kinda see why one might err on the side of the former.
LikeLiked by 4 people
Rachael said:
What do you mean by deli meat? Like ham and stuff? I ate ham all through my pregnancies and have never heard any recommendation against it. I avoided unpasteurized cheese and pate.
I think there are some US/UK differences in pregnancy advice – maybe this is one of them?
LikeLike
ozymandias said:
Deli meat has generally been left out for a while without being heated. That means it is a breeding ground for listeria, which can cause miscarriage, stillbirth, and newborn death. Of course, most people who eat deli meat will not contract listeria from it.
LikeLike
Jack V said:
My guess is that most advice is a mix of “fairly certain” and “really random, but someone told me that once, so I’ll pass it on”, but when you’re pregnant, people give you a LOT MORE advice, and most people don’t feel ok just ignoring it.
LikeLiked by 1 person
Jsfik Xujrfg said:
One theory: Making sacrifices during pregnancy (like not drinking) prepares you psychologically for the sacrifices you will have to make with a baby (like having no time to go out drinking). Pregnancy is mostly a long wait, and many feel an urge to do something for the baby, do the latch on to things that are actually pointless.
LikeLike
Julia Wise said:
I looked through some pregnancy advice lists to see what proportion seems bad. The “correcting pregnancy myths” genre is obviously going to focus on the parts that are wrong, but I wouldn’t say “all pregnancy advice is so bad.”
Some examples:
http://www.babycentre.co.uk/a536361/10-steps-to-a-healthy-pregnancy
http://www.parents.com/pregnancy/my-body/nutrition/healthy-pregnancy-tips/
http://www.healthnavigator.org.nz/healthy-living/pregnancy/pregnant-what-you-need-to-know/
Some of it seems stupid – avoid artificial coloring? But most seems reasonable – eat nutritious food including oily fish, sleep enough, drink enough fluids, don’t smoke, take a vitamin.
It seems reasonable, given that we can’t really come up with a controlled trial for alcohol exposure, to conclude that something that’s definitely bad in large quantities might be a little bad in small quantities. It takes high exposure during the right part of pregnancy to produce diagnosable symptoms of fetal alcohol syndrome, but I can totally believe that there are low-level effects that aren’t clearly attributable to maternal drinking.
LikeLiked by 1 person
dndnrsn said:
About fish and mercury exposure: different fish caught in different places have different levels of mercury content – generally the higher on the food chain the more mercury. So canned tuna is a bad idea – it also tends to be really lean, and so it doesn’t contain much Omega-3 fat anyway.
Fatty cuts of wild salmon caught in certain places are probably the best bet. And most fish oil supplements at least claim that they’ve filtered out the mercury and so forth.
LikeLiked by 1 person