Files
Abstract
Although Uganda reported large reductions in HIV/AIDS prevalence during the 1990s, recent
evidence suggests that country’s rate of new HIV infections is on the rise. This study explores
the factors that are correlated with sexual behavior and the risk of HIV infection using a unique
dataset of 19,500 individuals from the 2011 Uganda AIDS Indicator Survey. This survey tested
individuals 15-49 years of age for sexually transmitted infections, including the HIV virus. The
same survey also collected background information for all tested individuals. This information
is similar to what is collected in a typical demographic and health survey (DHS). We estimate
probit models for the correlates of risky sexual behaviors that can lead to HIV infection such as:
having concurrent sexual partners, no-condom use, and alcohol use during sex. In addition, we
examine correlates of having been tested for HIV prior to the survey. Also, we estimate models
for correlates of the risk of testing HIV positive as well as the self-assessed risks of contracting
HIV. We find that higher education attainment and access to health facilities are important for
adopting safe sexual behaviors as well as the reducing the risk of testing HIV positive. Among
HIV infected couples, we find that women have a higher rate of discordance which is at odds
with the low rates of self-reported extra marital sexual behavior.