NATIONWIDE COMMUNITY-BASED SEROLOGICAL SURVEY OF HIV-1 AND OTHER HUMAN RETROVIRUS INFECTIONS IN A CENTRAL AFRICAN COUNTRY

In December, 1986, a nationwide serological survey of human immunodeficiency virus type 1 (HIV-1) infection in the general population of Rwanda was done in two parts—one in the rural and the other in the urban population. The sampling method was a modification of the cluster sampling technique developed for monitoring immunisation coverage. Antibodies to HIV-1 (and to HIV-2 and human T-cell leukaemia/lymphoma virus type I [HTLV-I]) were detected by immunoenzymatic assays and confirmed by western blot. The number of clusters surveyed was 30 in each setting, urban and rural. HIV-1 seroprevalence was 17·8% (95% confidence interval 14·3-21·2%) in the urban sample (n = 1870) and 1·3% (0·5-2·2%) in the rural sample (n = 742). In the urban sample, females were more frequently HIV-1 seropositive than males (21·0% vs 14·6%). Age-specific peaks of HIV-1 seroprevalence were identifed at 0 to 5 years of age (10·1%) and at 26-40 years (30·0%). No differences in seroprevalence were observed in terms of age and sex in the rural sample. None of the sera were seropositive for HIV-2 and HTLV-I seroprevalence was 0·2% in the urban sample and 0·3% in the rural. Nationwide serological surveys could be effective in evaluating the spread of HIV infection and the efficacy of public health interventions against AIDS in developing countries.