Abstract
Sexual behaviour and HIV prevalence trends reported in overall study populations may mask different sexual behaviour and HIV prevalence changes in specific sub-groups. One group of particular interest are adult migrants from rural households. In this paper, we compare and contrast trends in sexual behaviour and HIV prevalence by residential status in a rural South African community 2005-2011. We use repeated cross-sectional sexual behaviour and HIV data available in the Africa Centre Demographic Information System (ACDIS). Multiple imputation methods and inverse probability weighting are used to adjust for missing data and selection bias respectively. Generally, trends in sexual behaviour indicators are in the same direction for residents and non-residents although sexual risk behaviours remain higher in non-residents compared to residents after adjusting for age. Marked differences are observed in risk behaviours by gender. Non-resident women are more likely than resident women to report condom use at last sex with their regular partner; the opposite is observed in men. There is no significant difference in HIV prevalence by residency status for men or women. In this paper we also discuss the extent to which current residential status may mask complex histories of migration.
confirm funding
Event ID
17
Paper presenter
21 090
Type of Submissions
Regular session only
Language of Presentation
English
Weight in Programme
2
Status in Programme
1
Submitted by Victoria.Hosegood on