| dc.description.abstract | Background: In designing HIY-l vaccine efficacy trials, it will be important to accounrfor changes in HIY -1 incidence, especially changes over time and while participating in riskreduction programs.
Methods: Data from an open cohort of female prostitutes in Mombasa, Kenya were analyzed. Individuals were censored after 3 years of follow-up or at HIY -J seroconversion. Models of HIY -J and sexually transmitted disease incidence and sexual risk behaviors over time were constructed using generalized estimating equations. Annual HIV -1 incidence for the entire cohort was calculated.
Results: The risk of HIY -J infection declined I O-fold during 3 years of follow-up (from 17.4 to 1.7 cases per 100 person-years for the first and last six months of follow-up, respectively, p<O.OO I). The incidence of gonorrhea, chlamydia, and genital ulcer disease each fell by half, and significant reductions in high-risk sexual behavior were observed. Fifty-seven percent of I-ITV-J seroconversions occurred within 6 months of enrollment, and 73% occurred within I year. Despite on-going recruitment and evidence of increasing HIV -1 transmission in the community, accumulation of low-risk individuals led to a significant decline in HIY -1 incidence in the cohort as a whole during the first 4 years of the study (17.1, 11.5. 13.8, and 9.1 cases per 100 person-years during years 1.2. 3. and 4. respectively. p<O.OO I).
Conclusions: This study documents a dramatic decline in the risk HIV-l infection while participating in a prospective observational cohort. with most seroconversions occurring within one year of enrollment. Failure to anticipate variations in HIY-J incidence within high-risk populations will result in HIY-l vaccine trials with insufficient numbers of seroconversions to demonstrate protection from HIY -1 infection by efficacious vaccines. | en |
