AEGiS-14IAC: Can population differences in Mwanza, Rakai and Masaka explain the contrasting outcomes of the intervention trials? A modelling study.

14th International AIDS Conference

Barcelona, Spain - July 7-12, 2002

Can population differences in Mwanza, Rakai and Masaka explain the contrasting outcomes of the intervention trials? A modelling study.

Int Conf AIDS 2002 Jul 7-12; 14:(abstract no. MoOrD1086)

White RG, Orroth KK, Korenromp EL, Bakker R, Wambura M, Sewankambo NK, Wawer MJ, Kamali A, Grosskurth H, Habbema JD, Hayes RJ
London School of Hygiene and Tropical Medicine, London, United Kingdom

BACKGROUND: Differences in sexual behaviour and curable STD prevalence have been hypothesized as an explanation for the differential impact of the Mwanza, Rakai and Masaka HIV/STD intervention trials. This study tests this hypothesis using microsimulation modelling.

METHODS: The STDSIM microsimulation model was used to simulate the transmission of gonorrhoea, chlamydia, chancroid, syphilis, trichomoniasis, HSV2 and HIV in the three trial populations. Simulated STD natural history, cofactor values, transmission probabilities, and the demography, sexual behaviour and HIV/STD epidemiology of the trial populations were based on trial data and literature review. Scenarios were modelled for each site, with and without the intervention (100 iterations).

RESULTS: A good fit of the model to baseline data on demography and sexual behaviour was achieved (not shown). By assuming earlier HIV introduction and reduction in risk behaviour starting in 1986 in Uganda, the model outcomes fitted the lower rates of curable (short-duration) STD and higher HIV prevalence observed in Rakai as compared to Mwanza at the start of the respective trials, as illustrated by HIV incidence in the figure. Simulated reductions in cumulative HIV incidence over 2 years were 13% in Rakai and 37% in Mwanza, which lie within the 95%CI of observed impacts in the two trials (-16% to +19% and 15% to 55%). [Image not included]

CONCLUSIONS: The model results support the hypothesis that relatively high risk behaviour in Uganda prior to the late 1980s, and subsequent reductions since, can explain the lower impact of STD treatment on HIV incidence observed in Rakai compared to that in Mwanza. The differing impacts are a consequence of the lower fraction of HIV incidence attributable to curable STD in Rakai, which is, in turn a consequence of the lower prevalence of curable STD and the more generalised HIV epidemic in Rakai. These are preliminary results, final results, including model outputs for the Masaka trial, will be presented.

Keywords: AEGIS, Prevalence, Incidence, HIV Infections, Sex Behavior, Risk-Taking, Research Design, Evaluation Studies, Population, Gonorrhea, Uganda, HIV Seropositivity, Syphilis, Chlamydia Infections, Demography

020707
MoOrD1086