I don’t trust science anymore. Especially not when it’s presented by the New York Times.
Here’s an article about The Risks to Babies of Older Fathers, and I instantly thought of my Bloomberg Opinion colleague Ariel Procaccia’s recent post about the need to teach statistics to scientists. See, anyone who has observed the slightest amount of animal husbandry knows that this NYTimes article is bullshit. Stud animals are bred until they can’t get it up anymore; in fact breeders prefer older males so they can avoid propagating animals that develop health problems later on.
I feel like a lot of our misconceptions regarding genetics and the heritability of traits could be totally avoided if more journalists had participated in 4-H.
Anyway, back to this research regarding the Association of Paternal age with Perinatal Outcomes in the United States. It’s published in the BMJ, a reputable British medical journal. The authors are physicians and professors at Stanford, an institution more infallible than the Pope. The study claims a sample size of 40.5 million births — that’s a lot!
Of the 40.5 million births, only 1.2 million (about 3%) had fathers over the age of 45. Out of that 3%, 61% of the babies had mothers over the age of 30. Of the 97% of babies with fathers under the age of 45, only 14.5% had mothers over the age of 30. So this is in fact a relatively small population with a skewed maternal age distribution.
The study claims to adjust for maternal age. The “adjustment” is a stratification of mothers into three groups: <25, 25-34, and >34. The <25 bucket has wide error bars, because there are very few babies born with young mothers and old fathers. The 25-34 and >34 buckets do show an association between adverse births and paternal age, but the buckets themselves are ridiculous. There’s a huge difference between a 25-year-old and a 34-year-old mother, and a massive difference between a 34-year-old and a 54-year-old (the oldest mother used in this study!). Older fathers are more likely to pair with older mates, so by lumping the mothers into such wide groups, the study does the exact opposite of adjusting for maternal age.
In conclusion, The Risk To Babies of Older Fathers is that their mothers tend to be old too. #FakeScience
The thing is, I would love for the NYTimes article to be true. Biological differences are the root cause of so much gender inequality in the workplace, and I honestly think the world would be a better place if men succumbed to the same biological clock as women. Of course by “better place” I mean, A Better Place For Me. Isn’t that what equality is all about?
Huh. Is this news? The idea has been around for quite a while.
Common points I’ve seen mentioned are that:
1. Older fathers tend to have a higher age-at-birth-of-first-child, not just a higher age-at-birth-of-the-child-that-qualified-them-as-an-“older father”. This suggests that their overall genetic fitness is low — a fitter man would be able to attract a mate more easily, and reproduce sooner. Parents with problems having offspring with problems is to be expected.
2. The mate these men do eventually attract may have problems of her own (or why settle?). Your observation that these women tend to be older falls into this category.
As to the observation that genetics journalism could be improved if more journalists had participated in 4-H, you remind me of something I read somewhere suggesting that smaller family sizes and the modern tendency for family members to work in radically different fields may help explain why people attribute so little to genetics. If you have several children, you’ll notice that they’re quite different and you can’t get them to react similarly to the same things. If you frequently interact with several members of the same family, you’ll notice that (compared to unrelated strangers) they’re all pretty similar. But it’s unusual today to have a lot of experience with a person and also several of that person’s close relatives. You’re more likely to have experience with a person and several of that person’s unrelated coworkers.
But there is a source of fake experience of what families are like — television! And family members on television are chosen, for dramatic reasons, to have very different personalities from each other. There’s no point in giving a TV family three children who quietly do their homework; the extra two don’t open up much in the way of story potential.
Fascinating! The NYTimes presented it as new research (and the paper itself claims that a comprehensive study has never been done) — it’s quite possibly that others had looked into it before, and found no meaningful causal relation, and dropped it.
The fake-TV family experience is a very good point, as is the idea about smaller family size. Maybe this is media-induced brainwashing, but I still lean towards nurture over nature on some things. I mean, just look at Google, Facebook, and every academic institution in this country! Somehow they sourced a population with equal representation from every race and gender, all of whom think exactly alike.
No prior comprehensive study — possible, but I’ve been reading about this for years through following Razib.
2009: https://scienceblogs.com/gnxp/2009/03/10/old-fathers-have-duller-childr (“Older fathers have duller children [than expected from measured paternal characteristics]”)
2012: http://evoandproud.blogspot.com/2012/09/on-paternal-age-and-iq.html (“high average age at first marriage shouldn’t be considered a predictor for low societal IQ, because while it is characteristic of Africa, it is also characteristic of north/west Europe”)
2012: http://blogs.discovermagazine.com/gnxp/2012/08/beware-of-the-ancient-of-days/ (links and comments on a paper concluding that children of older fathers have more de novo mutations, and that “these observations shed light on the importance of the father’s age on the risk of diseases such as schizophrenia and autism”)
2013: http://blogs.discovermagazine.com/gnxp/2013/09/older-fathers-and-the-iq-of-their-children/ (“Reports of older fathers having duller children appear to be overblown”)
I don’t see author overlap between the various papers linked in those posts and the one currently under discussion by the NYT, but I didn’t check very thoroughly.
As noted in the NYT piece, the hypothesis suggests itself by virtue of the fairly radical difference in gamete generation between males (generated throughout life) and females (generated before birth and stored).
I left a reply with several links; maybe check the comment spam filter?
Sorry about that, the spam filter gets triggered by anything more than 3 external links. Thanks for the references, will check them out.