The authors would like to thank Dr Rothenberg and colleagues for their interesting comments  on our paper on concurrent sexual partnerships and HIV . The ‘almost perfect rank correlation’ among the three measures of concurrency and the two bacterial sexually transmitted infections (STI), gonorrhoea and chlamydial infection, is indeed striking. Such a correlation was not found either for HIV infection or for positive syphilis serology (defined as a positive plasma positive reagin and Treponema pallidum haemagglutination antibody). Using the same city order as proposed by Rothenberg and colleagues, prevalence rates for positive syphilis serology were 1.3, 6.6, 1.8, and 11.3% in Yaoundé, Kisumu, Cotonou and Ndola among men and 5.6, 4.3, 3.4, and 14.0% among women, respectively .
Dr Rothenberg and colleagues  suggested that the concurrency we measured was not a good surrogate for historical concurrency. We were indeed very much concerned that the sexual behaviour patterns at the time of our study were not the same as those that prevailed at the time of the start of the HIV epidemics in the four cities. However, we did not find convincing evidence for changes in sexual behaviour over time that could explain the lack of association at the population level, of high-risk sexual behaviour and HIV prevalence .
It is clear that there are important differences in the transmission dynamics of STI. As for HIV, the rate of spread of this virus is determined by many factors that can be classified in two categories: factors that determine the probability of sexual intercourse with an HIV-infected partner (sexual behaviour factors) and factors that influence the per sex act transmission probability (such as other STI, male circumcision, viral load of the infected partner …). In the Multicentre Study  the lack of male circumcision and herpes simplex virus type 2 seropositivity were found to be independent individual risk factors for HIV infection. These factors were also found to be more prevalent in the two high HIV prevalence cities than in the two low HIV prevalence cities . We concluded that, at the ecological level, the association between the prevalence of high-risk sexual behaviour (high rates of partner change, concurrency, contacts with sex workers …) and HIV prevalence was confounded by factors that enhance the transmission of HIV during sexual intercourse .
Dr Rothenberg and colleagues  suggested that there may be important medical transmission of HIV, which had been ignored in our study. However, if this were the case, then we should find a high prevalence of HIV infection in young men and women who deny that they ever had sexual intercourse. Among men who reported that they had never had sexual intercourse, the prevalence of HIV was 0% in all cities except in Ndola, where it was 4.8%. Among women, the HIV prevalence rate varied from 0 to 10.8%. At least 50% of these HIV infections occurred in association with another STI, questioning the validity of self-reported behaviour, especially among young women, and suggesting that most of these infections were acquired through sexual intercourse. In Kisumu, the HIV prevalence among men increased from 3.5% in the 15–19 years age group to 28.7% in the 25–29 years age group. In Ndola, it increased from 3.7 to 27.3%. This strongly suggests that most of the HIV infections we found were acquired through sexual transmission. Indeed, we found high rates of partner change to be associated with an increased risk of HIV at the individual level.
Although the estimation of the low impact of the nosocomial transmission of HIV at the population level could be updated , we believe the priority is to do more research on factors that influence the sexual transmission of HIV, in particular herpes simplex virus type 2 infection and male circumcision, and to re-design effective prevention programmes.
1. Rothenberg R, Potterat J, Gisselquist D. Concurrency and sexual transmission [Letter]. AIDS 2002, 16: 678–679.
2. Lagarde E, Auvert B, Caraël M. et al
. Concurrent partnerships and HIV epidemic in five urban communities of sub-Saharan Africa. AIDS 2001, 15: 877–884.
3. Buvé A, Weiss HA, Laga M. et al
. The epidemiology of gonorrhoea, chlamydial infection and syphilis in four African cities. AIDS 2001, 15 (Suppl. 4) : S79–S88.
4. Buvé A, Lagarde E, Caraël M. et al
. Interpreting sexual behaviour data: validity issues in the multicentre study on factors determining the differential spread of HIV in four African cities. AIDS 2001, 15 (Suppl. 4) : S117–S126.
5. Auvert B, Buvé A, Ferry B. et al
. Ecological and individual level analysis of risk factors for HIV infection in four urban populations in sub-Saharan Africa with different levels of HIV infection. AIDS 2001, 15 (Suppl. 4) : S15–S30.
6. Buvé A, Caraël M, Hayes R. et al
. The multicentre study on factors determining the differential spread of HIV in four African cities: summary and conclusions. AIDS 2001, 15 (Suppl. 4) : S127–S131.
7. Van de Perre Ph, Diakhate L, Watson-Williams J. Prevention of blood-borne transmission of HIV. AIDS 1997, 11 (Suppl. B) : S89–S98.