AEGiS-14IAC: Applying statistical methods in sexual networks to understand the dynamics of HIV/STDs and dropouts in mobile and non-mobile cohorts of rural South Africa.

14th International AIDS Conference

Barcelona, Spain - July 7-12, 2002

Applying statistical methods in sexual networks to understand the dynamics of HIV/STDs and dropouts in mobile and non-mobile cohorts of rural South Africa.

Int Conf AIDS 2002 Jul 7-12; 14:(abstract no. C10704)

Zuma K, Garnett G, Williams B, Karim SS, Mkaya D, Lurie M
Medical Research Council, Durban, South Africa

BACKGROUND: Mobility is associated with HIV. To investigate the direction of HIV/STD infection in clusters of sexual networks among migrants, non-migrants and their rural sexual partners. Identify factors associated with dropout in longitudinal studies of HIV/STDs. Design and

METHODS: Between October 1998 and October 2001, 631 migrant men, non-migrant men and their rural partners from one South African district were recruited into a community based longitudinal study. There were 100 individuals, 187 dyads, 40 triads, 4 quadriads and 1 pentad. HIV, other STDs and sexual behaviour were measured every four monthly visit. We used generalized linear mixed model approach to account for interdependence of units within and random variation between sexual networks.

RESULTS: Baseline HIV infection was higher among migrant men (29.1%) than their partners (22.6%), p=0.127 and non-migrant men were slightly more likely to be infected than their partners, 19.2% and 18.2%, p=0.818. Cluster adjusted odds of being infected with at least one STD at one particular time were higher in a migrant relationships than in a non-migrant relationships (OR=1.23, p=0.058). Estimated dropout at each visit was 40%. HIV infection was consistently higher among dropouts than non-dropouts with average infection rate of 41.4% and 21.3%. Among dropouts, 85% knew their HIV negative status and 67% knew they were infected with HIV, p=0.014.

CONCLUSION: Migrants and their partners are at particular high risk of HIV infection and other STDs. The trend of higher STD infections in migrant relationship persists in all visits. High dropout is likely to underestimate incidence of HIV and other STDs. Knowledge of HIV status gives information about participation in the study. This has important implications for intervention programmes aimed at controlling HIV infection since both infected and uninfected individuals have to be counselled about an ongoing risk of being infected with and transmitting HIV.

Keywords: AEGIS, HIV Infections, Sexual Partners, HIV Seropositivity, Sex Behavior, Patient Dropouts, Transients and Migrants, Risk Factors, Longitudinal Studies, South Africa, Human, Male, methods, Epidemiology, statistics & numerical data

020707
C10704