Search

Background The relationship between HIV and poverty is complex and recent studies reveal an urban--rural divide that is not well understood. This paper examines the urban--rural disparity in the relationship between poverty and HIV infection in Kenya, with particular reference to possible explanations relating to social cohesion/capital and other moderating factors. Methods Multilevel logistic regression models are applied to nationally-representative samples of 13 094 men and women of reproductive age from recent Kenya Demographic and Health Surveys. Results The results confirm a disproportionate higher risk of HIV infection among the urban poor, despite a general negative association between poverty and HIV infection among rural residents. Estimates of intra-community correlations suggest lower social cohesion in urban than rural communities. This, combined with marked socio-economic inequalities in urban areas is likely to result in the urban poor being particularly vulnerable. The results further reveal interesting cultural variations and trends. In particular, recent declines in HIV prevalence among urban residents in Kenya have been predominantly confined to those of higher socio-economic status. Conclusion With current rapid urbanization patterns and increasing urban poverty, these trends have important implications for the future of the HIV epidemic in Kenya and similar settings across the sub-Saharan Africa region. [ABSTRACT FROM AUTHOR]

Using data from the Nairobi Urban Health and Demographic Surveillance System (NUHDSS), this paper seeks to understand the characteristics of adolescent girls who are circumcised in Kenya. The paper discusses the determinants of female circumcision in two Nairobi informal settlements. It is based on detailed information collected from young persons aged 12-24 targeting major transition events such as first sex, childbirth, marriage, and circumcision. Out of 4058 adolescents and young people interviewed 2010 were adolescent girls and young women. Out of the total number of interviewees, the 527 girls and young women on whom this paper is based are from ethnic communities that practice circumcision. We used the life-table technique to estimate the median age at circumcision and logistic regression to analyse the relationship between female circumcision and adolescent sexuality, controlling for other characteristics. The type of stay within the demographic surveillance area (DSA), religion, ethnicity, residential location and mother's education were associated with being circumcised. Current school attendance was not associated with being circumcised but if one had never attended school then they were more likely to be circumcised. For policy making, it is imperative to explore the nature of social networks within which circumcision decisions are taken and enforced. [ABSTRACT FROM AUTHOR]