Title: Integrating epidemic dynamics and statistical modelling to estimate and project HIV epidemic trends in sub-Saharan Africa
Every year, UNAIDS supports national governments to create estimates of their HIV epidemic, including estimates of HIV prevalence, new HIV infections, AIDS deaths, and the coverage of HIV treatment and prevention programmes. Estimates are generated by fitting a simple but flexible HIV epidemic model to national data sources about HIV prevalence in a Bayesian statistical framework. In this talk, I will describe the history and development of the mathematical models and statistical methods that are used for estimating HIV epidemic trends, and describe newly developed models that use stochastic processes to more flexibly model HIV transmission trends. Finally, I will discuss recent work to integrate data from survey and routine health system data to estimate small-area level HIV prevalence, ART coverage and incidence, and future directions to develop models for inferring spatio-temporal epidemic dynamics.