"'Abstinence only' education does not slow the spread of AIDS"

Read the full article in The Economist of September 20th 2007.

"Abstinence-only does not work. Abstinence-plus probably does."

Last month Dr Underhill published a review of 13 trials involving 16,000 young people in America. The trials compared the sexual behaviour of those given an abstinence-only education with that of those who were provided with no information at all or with whatever their schools normally taught. Pregnancies were as numerous in both groups. Sexually transmitted diseases were as widespread. The number of sexual partners was equally high and unprotected sex just as common.

Having thus discredited abstinence-only teaching, Dr Underhill and her colleagues decided to evaluate the slightly more complicated message of “abstinence-plus” using 39 trials that involved 38,000-odd young people from the United States, Canada and the Bahamas. Their results are published in the current issue of Public Library of Science Medicine.

This tuition—compared, as before, with whatever biology classes and playgrounds provide—reduced the number of pregnancies in three out of seven trials (the remaining four recorded no difference). Four out of 13 trials found that abstinence-plus-educated teenagers had fewer sexual partners, while the remainder showed no change. Fourteen studies reported that it increased condom use; 12 others reported no difference. Furthermore, in the vast majority of cases, abstinence-plus participants knew more about AIDS and HIV (the virus that causes the disease) than their peers did. And the tuition often reduced the frequency of anal sex (which brings a greater chance of passing on HIV than the vaginal option). In contrast to the fears of the protagonists of abstinence-only education, not one of the trials found that teenagers behaved in a riskier fashion in either the long or the short term after receiving abstinence-plus instruction.

Comment: I would characterize this as "Duh", except that the Bush administration does not seem to have thought of this obviously likely possibility. Of course, it is important to have meta-analysis of a body of research on so important an issue to be sure. But Duh! JAD