There has been increased interest in the implementation of combination HIV prevention and treatment trials in order to better understand their effects on the HIV epidemic. The Harvard School of Public Health has been awarded a CDC Cooperative Agreement to evaluate the Botswana Combination Prevention Project (BCPP), a matched paired community-randomized trial of an enhanced combination prevention package (CP) that includes expanded coverage of HIV counseling and testing, prevention of mother to child transmission, male circumcision, and treatment as prevention in Botswana.
We aim to simulate the BCPP using an agent-based stochastic model, the HIV Calibrated Dynamic Model (HIV-CDM) in order to inform the use of combination prevention in Botswana and southern Africa. The HIV-CDM simulates transmission of HIV from infected individuals to uninfected individuals. This expansive, detailed and innovative model will reflect the best available data from prevention studies, Botswana specific HIV epidemiology and treatment. The three specific aims are:
Aim 1 : To calibrate the HIV-CDM to Botswana population parameters, including HIV incidence and prevalence, using Bayesian melding type methods. To perform simple sensitivity analyses to understand the effects of historical antiretroviral therapy (ART) rollout and high risk sexual behavior on the epidemic dynamics between 2002 and 2012.
Aim 2 : To simulate the BCPP trial to estimate the overall impact of the CP and the contributory role of each component on the 36 month HIV cumulative incidence among 16-64 year old residents in Botswana, and the expected future timeline and the particular combinations of HIV prevention programs that may result in an AIDS-free generation in Botswana.
Aim 3 : To understand the epidemiologic and policy impact of explicitly incorporating sexual network structure and uncertainty in parameters on the results of an agent-based model. Expanding the HIV-CDM to Botswana will allow the BCPP investigators to conduct policy and sensitivity analyses to fully evaluate the impact of the CP in Botswana.
The results of the simulation will be used to help investigators estimate the effect of HIV prevention interventions alone and in combination on HIV prevalence, incidence, and life expectancy. The model will also be used to estimate the impact of sexual networks on the HIV epidemic and prevention interventions in southern African communities.
| Murray, Eleanor J; Robins, James M; Seage, George R et al. (2017) A Comparison of Agent-Based Models and the Parametric G-Formula for Causal Inference. Am J Epidemiol 186:131-142 |
| McCormick, Alethea W; Abuelezam, Nadia N; Fussell, Thomas et al. (2017) Displacement of sexual partnerships in trials of sexual behavior interventions: A model-based assessment of consequences. Epidemics 20:94-101 |
