To the Editor.
—In a partner study on HIV transmission among heterosexual couples, Padian et al1 observed only one case (1.4%) of male-to-female transmission of HIV in their study of 72 male partners of HIV-infected women. The observed case was in a couple described as unique in that vaginal and penile bleeding during intercourse was noted. However, since nearly half (32/72) of the couples studied used condoms, the risk of transmission may have been underestimated.On the other hand, male-to-female transmission was estimated to be 17.5 times more effective than female-to-male transmission (crude OR). Apart from the fact that the 90% CI for the OR given in the article is large (3.5 to 353), and the usual 95% CI would be still larger, the crude measurement of the relative efficiency of HIV transmission in both directions may be invalid for at least one reason: the authors did not take