Institutional members access full text with Ovid®

Stochastic simulation of the impact of antiretroviral therapy and HIV vaccines on HIV transmission; Rakai, Uganda

Gray, Ronald H; Li, Xianbin; Wawer, Maria Ja; Gange, Stephen Jb; Serwadda, Davidc; Sewankambo, Nelson Kd; Moore, Richarde; Wabwire-Mangen, Fredc; Lutalo, Tomf; Quinn, Thomas Cg; for The Rakai Project Group

Epidemiology & Social

Objective: To model the effects of antiretroviral therapy (ART) and HIV vaccines on HIV transmission using empirical data from studies in Rakai, Uganda.

Design: A stochastic simulation model estimated HIV incidence, probabilities of transmission per coital act and the reproductive number (R0) with ART and HIV vaccines. Model inputs included Rakai data on HIV transmission probabilities per coital act by HIV viral load, age and gender, and sexual behaviors. The impacts of therapy were derived from US programs, and vaccine assumptions included preventive efficacies ranging from 25 to 75%. Component projection models estimated the numbers of HIV-infected persons over 20 years.

Results: The model incidence [1.57/100 person years (PY)] closely fitted empirical data (1.5/100 PY). Simulations of ART using DHHS treatment guidelines, predicted declines in HIV incidence, but R0 remained > 1.0, and the numbers of HIV-positive persons did not change substantially over 20 years. Preventive vaccines with > 50% efficacy and > 50% population coverage could reduce R0 to < 1.0, and substantially reduce the number of HIV-infected persons over 20 years. Concurrent ART and a preventive vaccine can have substantial impact at lower levels of population coverage and would markedly reduce the HIV infected population over 20 years. However, behavioral disinhibition with increased numbers of sexual partners in either ART or vaccine recipients, increased HIV incidence and diminished intervention impact.

Conclusion: ART alone cannot control the HIV epidemic in mature epidemics such as Rakai, and persons in need of therapy will increase over time. ART in combination with a low efficacy vaccine could control the epidemic, if behavioral disinhibition is prevented.

From the Department of Population and Family Health Sciences, Johns Hopkins University, Bloomberg School of Public Health, Baltimore, Maryland, the aHeilbrun Center for Population and Family Health, Columbia University, Joseph L. Mailman School of Public Health, New York, the bDepartment of Epidemiology, Johns Hopkins University, Bloomberg School of Public Health, Baltimore, Maryland, USA, the cSchool of Public Health and the dDepartment of Medicine, Faculty of Medicine, Makerere University, Kampala, Uganda, the eDepartment of Medicine, Johns Hopkins Hospital, Baltimore, Maryland, USA, the fRakai Project, Uganda Virus Research Institute, Entebbe, Uganda, the gDepartment of Medicine, Johns Hopkins University, School of Medicine, Baltimore, and the National Institutes of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland, USA. *See Appendix I.

Correspondence to R. H. Gray, Suite 4030, Johns Hopkins University, School of Hygiene and Public Health, 615 N. Wolfe Street, Baltimore MD 21205, USA.

Received: 8 July 2002; revised: 2 November 2002; accepted: 7 March 2003.

© 2003 Lippincott Williams & Wilkins, Inc.