20 January 2008

Dubious Conclusion Spun From Slightly Larger Sample . . .

. . . this time a NYT story with the headline "Pregnancy Problems Tied to Caffeine". That's an alarming headline, and suggests a cause for concern, but at the end of the article, they quote a sceptic of the study:
Dr. Li said he and his colleagues had determined that the risk from caffeine was real and could not be explained away by different rates of morning sickness.

Dr. Carolyn Westhoff, a professor of obstetrics and gynecology, and epidemiology, at Columbia University Medical Center, had reservations about the study, noting that miscarriage is difficult to study or explain. Dr. Westhoff said most miscarriages resulted from chromosomal abnormalities, and there was no evidence that caffeine could cause those problems.

“Just interviewing women, over half of whom had already had their miscarriage, does not strike me as the best way to get at the real scientific question here,” she said. “But it is an excellent way to scare women.”

She said that smoking, chlamidial infections and increasing maternal age were stronger risk factors for miscarriage, and ones that women could do something about.

“Moderation in all things is still an excellent rule,” Dr. Westhoff said. “I think we tend to go overboard on saying expose your body to zero anything when pregnant. The human race wouldn’t have succeeded if the early pregnancy was so vulnerable to a little bit of anything. We’re more robust than that.”

Nice to see the scepticism within the article, but why the alarming headline then? And when this story gets picked up by CBS Nightly News, and all the various local TV newsers, how much you wanna bet they leave out the scepticism?

So much more fun to freak out people without conclusive evidence.

No comments: