A study of the prevalence and risk factors leading to HIV infection among a sample of street children and youth of Kathmandu

The true prevalence of HIV and other sexually transmitted diseases among street children in Nepal is virtually unknown while information on related behavioural risk factors in this population is non-existent. The risk of HIV infection among street children and adolescents may be especially high due to their marginalized social and economic conditions. This study was conducted to determine the prevalence of HIV infection among a sample of street children and youth of Kathmandu and to identify risk factors associated with HIV infection in this group.A sample of street children and youth was recruited based on the purposive sampling of ten streets in Kathmandu, Nepal, known to have a high density of street children and youth. A total of 251 street children (aged 11-16 years) and youth (aged 17-24 years) were enrolled, with informed consent, from November, 2008 through June, 2009. Most of the participants (95%) were male. Case status was determined by serological assessment of HIV status; data on risk factors were obtained using structured survey interviews. HIV prevalence and rates of a number of behavioural risk factors suspected to play a role in HIV transmission among street children and youth were determined, including unprotected sex, intravenous drug use, and other risky sex and substance use behaviours. Among the 251 children and youth, we found an overall HIV prevalence of 7.6%. As the sample size of females was small (n = 13) and the behavioural risk factors are likely to be quite different for boys and girls, we conducted separate analyses by gender. As our small sample of females is unlikely to be representative and lacks power for statistical testing, our report focuses on the results for the males surveyed.The strongest behavioural risk factor to emerge from this study was intravenous drug use; 30% of the male subjects were injecting drug users and 20% of those were HIV positive. Furthermore, frequency of drug injection was a highly significant predictor with a dose-response relationship; males reporting occasional injection drug use were nearly 9 times more likely to be HIV positive than never users, while weekly drug injectors had over 46 times the risk of non-users, controlling for exposure to group sex, the only other significant risk factor in the multivariate model. This sample of street children and youth of Kathmandu has a nearly 20-fold higher prevalence of HIV infection than the general population of Nepal (0.39%). The children and yout