HIV prevalence and incidence are no longer falling in southwest Uganda: evidence from a rural population cohort 1989-2005

Background: Throughout the 1990s, HIV-1 prevalence and incidence were falling in Uganda. Recently, some researchers have noticed that HIV-1 prevalence is I eve I I in g off. We examine prevalence, incidence, and sexual behaviour trends in a rural population cohort in Uganda over 16 years. Methods: We report prevalence by survey round and incidence by calendar year from a prospective general population cohort study. Using logistic regression Wald tests, we examined casual partners, condom Use, and pregnancies. We examined age at sexual debut by means of life tables. Results: HIV-1 prevalence declined from 8.511% in 1990/1991 to 6.2%, in 1999/2000, and thereafter rose to 7.7% in 2004/2005. incidence (per 1000 person-years at risk) fell from 7.5 in 1990 to 4.1 in 1998, and thereafter increased to 5.0 by 2004. The 2005 incidence estimate reached ill all-time low of 2.5, but the preliminary 2006 estimate shows a rise again. Incidence trends varied by age and sex. Some sexual behaviour indicators showed more risky behaviour in recent years compared with the 1990s, whereas others indicated that the reduction in risky behaviour that began in the 1990s continues. Conclusion: HIV-1 prevalence is rising in this cohort. Incidence is stabilizing, and shows signs of increasing among some subgroups. The extent to which changing sexual behaviour has played a role in these epidermiological trends is Unclear, but it is likely to have contributed. To solidify the S that Uganda had throughout the 1990s in controlling the HIV epidermic, the efforts in HIV prevention need to be re-strengthened, using all strategies known. (C) 2008 Wolters Kluwer Health | Lippincott Williams & Wilkins.