Mild vexation at science reporting

« previous post | next post »

Donald G. McNeil, Jr., "New Lines of Attack in H.I.V. Prevention", NYT Science Times today:

Because 95 percent of gay men and 40 percent of heterosexual American women have had anal sex at least once during their lifetimes, according to surveys, rectal versions of the [microbocidal] gel are being developed.

Where to start?

"According to surveys": not referenced in the article, of course (newspaper reporting almost never cites sources, even in articles on technical fields), but also not referenced or linked to on-line. So we have no idea of how these surveys were done. Presumably, self-report (with all of the problems with that), but from what pool of informants, how selected? (Note that a huge number of men who have sex with men do not self-identify as gay.) And how interviewed (on the telephone, electronically, on a printed form, face-to-face, and if the last, in what context)?

Then, since the heterosexual women surveyed were presumably reporting on receptive anal intercourse (and not on fucking their male partners with strap-ons or free-standing dildoes), we have to guess that RAI is what was being reported on for the gay men too, though if you ask a gay guy if he's had anal sex, he'll say yes whichever role he's taken in the encounter. As always, the wording of the interview questions is absolutely crucial.

(I'm passing over the peculiarity that the heterosexual women are identified as American, while the gay men have no identifiers.)

Then, "at least once during their lifetimes". So, like, "did it once and didn't care for it" or "did it a couple of times to please my boyfriend" (or, much worse, "once got raped up the ass") counts as a yes? Is that reasonable?

Both figures sound high for me — not an order of magnitude high, but still high. But we have no way to judge.

The idea of a protective gel for rectal use, even if it's not 95% effective, sounds admirable to me, but the real trick is going to be getting it used by the people who need it. And that's a public health policy that needs careful development, with attention not only to people who engage in RAI with some frequency but also to their insertive partners.

[update on November 11: Robert Hay writes to point to two pieces by William Saletan in Slate last month: here and here. These are breezy but informative discussions of the results of the National Survey of Sexual Health and Behavior (from the Center for Sexual Health Promotion at Indiana University), as described in Volume 7, Supplement 5 of the Journal of Sexual Medicine, October 2010. And there's a summary discussion on the NSSHB's site, here.

Saletan does a good job surveying "the mainstreaming of anal sex" for women, and of course notes that the 40% once-in-a-lifetime figure doesn't give a good picture of these practices.

I still haven't found the source of the 95% figure for gay men. What I see in the NSSHB site is figures for RAI for men in general, which of course is drastically smaller than 95%, even for the once-in-a-lifetime question, while only 4% of men in the survey reported RAI in the previous year.]



Comments are closed.