The protective effect of condoms and nonoxynol-9 against HIV infection

Objectives. Whether or not spermicides can reduce the risk of human immunodeficiency virus (HIV) transmission remains an important question for the control of heterosexual HIV transmission. The authors provide estimates from a reanalysis of one of the few observational studies on the efficacy of condoms and spermicides, used separately and together, per vaginal contact. Methods. In this reanalysis, three different models were used to assess the efficacy of spermicides and condoms: linear (Pearl index), exponential (maximum likelihood), and monotonic (marginal likelihood). Results. Reported use of barrier methods among 27 432 contacts was as follows: condoms plus nonoxynol-9, 39%; condoms alone, 25%; nonoxynol-9 alone, 24%; and unprotected, 11%. Under all three models, the results indicate a strong protective effect for spermicidal suppositories. The Pearl index indicated that spermicide alone is apparently efficacious, but the efficacy per contact cannot be quantified with this approach. Maximum likelihood estimates for the efficacy of nonoxynol-9 alone and condoms (with or without nonoxynol-9) were 100% (95% confidence interval [CI_.95] = 43%, 100%) and 92% (95% CI_.95 = 79%, 100%), respectively. Conclusions. The data from this observational study suggest that spermicides may be efficacious in reducing the risk of HIV transmission.