Background: This report describes a follow-up study conducted in 2005 to evaluate programs intended to reduce HIV/AIDS among registered and unregistered female sex workers (FSW) and their male clients in six major urban areas in Benin (Cotonou, Abomey-Bohicon, Parakou, Porto-Novo, Kandi, Malanville). The objectives of this second study combining laboratory testing and behavioral questionnaires were to estimate the prevalence of HIV and sexually transmitted diseases (STI), to determine the sociodemographic and behavioral characteristics of FSW, and to assess changes in the behavior of registered FSW in the first four cites (Cotonou, Abomey-Bohicon, Parakou, and Porto-Novo) included in the initial study in 2002.
Design and methods: A cross-sectional study was conducted in a cohort of 930 self-identified FSW. The chi-square or Fisher's exact test were used to test correlation between HIV and social, demographic and behavioral factors and the Wilcoxon test to compare the distribution of continuous variables. Correlation was measured based on prevalence odd ratios (POR) obtained by multivariate logistic regression.
Results: The overall prevalence of HIV, Neisseria gonorrhoeae (NG) and Chlamydia trachomatis (CT) were 30.5%, 3.6% and 5.7% respectively. The prevalence of HIV was highest in Parakou (48.2%) and lowest in Abomey/Bohicon (16.4%). The prevalence of NG and CT were highest in Abomey/Bohicon and Porto-Nova respectively. Multivariate logistic regression analysis showed that HIV prevalence increased significantly for FSW who had had more than 10 clients in the previous week (POR = 1.40, 95% CI: 1.02-1.94), who had no other source of income (POR = 1.47, 95% CI: 1.08-2.00), who were currently unmarried or separated (POR = 2.63, 95% CI: 1.73-4.02), and who had never been married (POR = 2.07, 95% CI: 1.43-3.00). Older age and having had no non-paying partner in the last seven days were positively correlated with HIV infection. Between 2002 and 2005 there was a significant increase in the number of FSW who reported condom use with all clients (79% vs. 90%, p < 0.0001) and with all sexual partners (78% vs. 87%, p = 0.0007) during the last 7 days as well as in the median number of visits to a health care facility in the last 12 days (3 vs. 6, p < 0.0001). Overall these follow-up data showed a large decrease in the prevalence of HIV/STI. This finding supports continuation of programs aimed at FSW especially with regard to risk awareness, condom use, and STI screening and treatment.