Home > News & Opinion > Blog > Blame it on the rain

Blame it on the rain

Popularity tracker
By Apoorva Mandavilli
4 November 2008

Autism rates are higher in states where it rains more, according to a report thatʼs in newspapers everywhere today.

Say what?!

When you write about autism, and the many avenues scientists are exploring to pin down its cause, you get used to seeing all kinds of nutty theories bandied about. But this one really takes the cake.

Iʼm particularly baffled because the research is published in a reasonable journal, the Archives of Pediatrics and Adolescent Medicine, one of the JAMA/Archives journals, and by scientists and ‘policy analystsʼ from the more-than-reasonable Cornell University. Not that that makes it any less absurd.

The researchers looked at annual rainfall for California, Oregon and Washington State between 1987 and 1999, and concluded that autism rates are higher when there is a higher rate of rainfall during the childrenʼs first three years.

They come up with all sorts of potential explanations: maybe the children watched more TV because they were indoors, or were more exposed to household chemicals, or maybe they didnʼt get enough vitamin D ― or chemicals “in the upper atmosphere that are transported to the surface by precipitation" are to blame.

Or maybe, what theyʼre seeing is, as any novice scientist or reporter can tell you, a classic case of correlation ― not the same thing, by a long shot, as cause.

As one of my journalism colleagues pointed out, the accompanying editorial is delightfully called, “Do these results warrant publication?”

I would have voted no.

Comments

Name: Matthew BELMONTE
3 April 2009 - 12:41PM

The interpretation in the published version of their article is a lot more reasonable than the one with which these authors began (for discussion see "http://cornellsun.com/node/19538"), and these correlational data may in fact lead to more specific, prospectively testable hypotheses. That's why it was published: those who read more than just the title of the accompanying editorial will discover that the editorial author's answer to his rhetorical question "Do these results warrant publication?" is, in fact, affirmative.

Add a Comment

You can add a comment by filling out the form below. Plain text formatting.