4.4 HIV incidence modelling simulations for Kampala women and validation to Kampala ANC sentinel surveillance trends.
The simulation model for urban Kampala was compared to ANC trends
overall and by age, in both the baseline and the incidence reduction scenarios.
The reduction in incidence was started in the model in 1990, and was maintained
for a period of 5 years, resulting in slightly over 50% reduction in HIV
incidence. In 1995 no further incidence reduction was simulated. HIV prevalence
trends under the incidence reduction scenario are compared to overall Kampala
ANC sentinel site trends in Figure 19, and to age-specific prevalence trends for
the same site from 1992-96, in Figure 20.
Simulations of HIV
prevalence (%) in pregnant women in Kampala Uganda under baseline and incidence
reduction scenarios from 1985-2005, compared to HIV prevalence at the Nsambya
(Kampala) antenatal clinic sentinel surveillance site, 1985-96.
Simulation of HIV
prevalence (%) by age groups from 15 to 34, among '·
pregnant women in
Kampala Uganda under the incidence reduction scenarios from 1992-96. The
simulation (bars) is compared to age-specific HIV prevalence rates at the
Nsambya (Kampala) antenatal clinic sentinel surveillance site from 1992-96.
(Error bars represent 95% confidence intervals around the mean
prevalence of the empirical data).
The incidence reduction simulations show a very good
fit with the observed data and provide substantial validation of the finding
that major declines in HIV incidence underlie the observed HIV prevalence trends
in young adults, and that the epidemiological model incorporates reasonably
valid HIV incidence dynamics for