W. Kipp et al., PREVALENCE AND RISK-FACTORS OF HIV-1 INFECTION IN 3 PARISHES IN WESTERN UGANDA, Tropical medicine and parasitology, 46(3), 1995, pp. 141-146
A population-based serosurvey was conducted to determine the HIV-1 pre
valence and to evaluate various risk factors in three parishes in west
ern Uganda. Adults over 14 years were examined in all 13 villages of K
igoyera parish far from Fort Portal, in two villages of Kyamukoka pari
sh closer to Fort Portal and in the four fishing villages of Ntoroko p
arish at Lake Albert with a high population mobility. Personal intervi
ews and HIV serology using ELISA and Western blot were performed. Sera
showing at least two envelope bands in the Western blot were consider
ed as positive. The coverage of the registered eligible population was
74% in Kigoyera, 67% in Kyamukoka and 25% in Ntoroko. The prevalence
of HIV was 4% (97 of 2267 persons examined, 95% CI 3.4-5.1%) in Kigoye
ra, a typical rural area. Whereas it was 13% (53 of 393 persons examin
ed, 95% CI 10.1-16.9%) in Kyamukoka near to the district capital and e
xceptionally high with 24% (96 of 399 persons examined, 95% CI 19.9-28
.3%) in the comparatively isolated fishing villages of Ntoroko. In a m
ultivariate model the infection risk for HIV was in Kyamukoka two and
in Ntoroko five times higher than in Kigoyera. Among the two main ethn
ic groups one had a significantly lower risk to acquire HIV infection
than the other group. Conclusions: Because of their large proportion o
f migrating persons the fishing villages presented populations with hi
gh risks for HIV infection. The ethnic composition of the village popu
lation, representing group specific sexual behaviour, was a risk facto
r for HIV infection.