Catching up with yesterday’s fascinating autism story…
The story, from the University of California, Davis, is about a study reporting that older mothers–and older fathers, in some circumstances–might be more likely than their younger counterparts to have children with autism.
The study generated all kinds of numbers, and when I started to read through the press release, the study’s abstract, and some of the news stories, I was sorry I’d promised my colleagues here that I would post on this. The study is full of devilish details.
Part of the problem lies in sorting out the contributions of fathers’ age and mothers’ age. Previous studies have found that the risk of autism increases with a father’s age, but this one says that’s true only if the woman in question is under 30. Another issue is the question of relative risk versus absolute risk. Most stories led with the finding that women over 40 are more likely to have a child with autism than are younger women, and some put a number on that: A 40-year-old woman’s risk of having a child later diagnosed with autism was 50 percent greater than that of a woman between 25 and 29 years old, the press release said. But few said up high that the overall risk of having a child with autism remains low.
Except for the trusty AP, which eased my anxiety about posting on this. The fourth graf of the AP story included the crucial fact that “the risk of a woman over 40 having an autistic child was still less than 4 in 1,000, one expert noted.”
I also thought the AP’s lede, by Lindsey Tanner, was clear, and reasonably well packed with information and some context:
A woman’s chance of having a child with autism increases substantially as she ages, but the risk may be less for older dads than previously suggested, a new study analyzing more than five million births found.
(A minor quibble on the writing: This is what I call a lede with a caboose. We’re moving along the track just fine, and we expect the sentence to end with the attribution, “a new study found.” That’s a nice, one-breath sentence when read out loud. But the lede, we discover when we get to the attribution, is dragging along an extra car that we didn’t expect–a new study “analyzing more than five million births” found. That’s the caboose. Better to de-couple it and put it in the next graf. Commenters are free to berate me over my slightly ridiculous and over-extended train metaphor.)
What’s interesting about the AP lede is that it puts fathers’ age in a secondary position, far less important than mothers’ age. In contrast, look at how Roni Caryn Rabin of The New York Times handles this:
Both Parents Ages’ Linked to Autism Risk
Older mothers are more likely than younger ones to have a child with autism, and older fathers significantly contribute to the risk of the disorder when their partners are under 30, researchers are reporting.
That makes the fathers sound more important than the AP lede did. As far as I can tell from looking over the study and the press release, both ledes are accurate. But the emphasis is quite different. The AP story, it seems to me, would reassure fathers, while the Times story would alarm them, at least those older fathers married to women under 30.
I found the story by Jennifer Thomas of HealthDay confusing. She buries us in too many numbers too high up in the story. And she puts even less emphasis than the AP did on the finding regarding fathers: “While previous research has shown that older dads are more likely to have a child with autism, this study found no such link between autism and paternal age, with one exception — older men who had children with much younger women.” Again, it’s accurate, but different.
A few others:
Daniel J. DeNoon on WebMD: Autism Risk Rises with Mother’s Age.
AFP: Study Confirms Link Between Maternal Age and Autism. Note the caboose.
Julie Steenhuysen of Reuters: Age of mother affects child’s autism risk: study. (Editing by Cynthia Osterman.)
Grist for the mill: UC Davis press release.
– Paul Raeburn
Leave a Reply