Updated estimates of condom effectiveness

Abstract

This paper reports results of a study concerning the effectiveness of condoms in preventing HIV transmission. The study expands on a 1993 meta-analysis that pooled the results of a number of studies of HIV transmission in serodiscordant couples to arrive at an overall condom effectiveness estimate of 69%. A meta-analysis of studies that compared seroconversion rates among couples who regularly use condoms and those who used them inconsistently was conducted to determine the use and/or effectiveness of condoms in preventing HIV transmission. Results of the analysis indicated that condoms are 90-95% effective when used consistently. To illustrate the impact of differential assumptions regarding the effectiveness of condoms in preventing the transmission of HIV, a community of gay men is considered in which the prevalence rate of HIV infection is 20%, supposing each man has sex once a week with a monogamous partner from the same population. The expected annual incidence of HIV infection in this community is 13% if condoms are never used, while consistent use of 95% effective condoms would reduce the incidence to about 1%. In this example, the probability of transmission for 52 acts of condom-protected intercourse is less than for a single act of unprotected intercourse. Moreover, inconsistent condom use also offers some protection against HIV in which the reduction achieved by using condoms 50% of the time is equal to almost half the reduction associated with consistent use.