Multicentre study on factors determining differences in rate of spread of HIV in sub-Saharan Africa: methods and prevalence of HIV infection

Objective: The objective of this study was to explore whether the differenc es in rate of spread of HIV in different regions in sub-Saharan Africa coul d be explained by differences in sexual behaviour and/or factors influencin g the probability of HIV transmission during sexual intercourse. Methods: A cross-sectional, population-based study was conducted in two cit ies with a high HIV prevalence (Kisumu in Kenya and Ndola in Zambia) and tw o cities with a relatively low HIV prevalence (Cotonou in Benin and Yaounde in Cameroon). In each of these cities, approximately 1000 men and 1000 wom en, aged 15-49 years, were randomly selected from the general population. C onsenting men and women were interviewed and were tested for HIV, syphilis, herpes simplex virus type 2 (HSV-2), gonorrhoea, chlamydial infection and trichomoniasis (the latter for women only). In addition, a survey was condu cted on a random sample of 300 sex workers in each city. The research instr uments, including the questionnaires and the laboratory procedures, were st andardized to permit comparison of results. Results: The numbers of men interviewed were 1021 in Cotonou, 973 in Yaound 6, 829 in Kisumu, and 720 in Ndola. The corresponding figures for women wer e 1095, 1116, 1060 and 1130. In YaouncI6, Kisumu and Ndola, the response ra tes for men were lower than for women due to failure to make contact with e ligible men. The proportion of eligible women who were interviewed was 86% in Yaounde, and 89% in Kisumu and Ndola. In Yaound6, 76% of eligible men we re interviewed, along with 82% in Kisumu and 75% in Ndola. The prevalence of HIV infection in men was 3.3% in Cotonou, 4.1% in Yaound6 , 19.8% in Kisumu and 23.2% in Ndola. For women, the respective figures wer e 3.4, 7.8, 30.1 and 31.9%. The prevalence of HIV infection among women age d 15-19 years was 23.0% in Kisumu and 15.4% in Ndola. Among women in Kisumu who had their sexual debut 5 years before the interview, the prevalence of HIV infection was 46%; in Ndola, it was 59%. Among sex workers, the preval ence of HIV infection was 57.5% in Cotonou, 34.4% in Yaounde, 74.7% in Kisu mu and 68.7% in Ndola. Conclusions: The HIV prevalence rates in the general population confirmed o ur preliminary assessment of the level of HIV infection in the four cities, which was based on estimates of HIV prevalence from sentinel surveillance among pregnant women.