Professor of Sociology and Statistics
University of Washington
EpiModel 2.0 is an interdisciplinary project to develop the statistical methods and software tools for the network-based mathematical modeling of HIV and other sexually transmitted infections (STIs) among men who have sex with men (MSM) and heterosexuals in the United States. As a MIDAS project, our goal is to build and apply a comprehensive computational epidemiology platform for investigating critical issues in HIV/STI prevention science now and over the next decade. This work will build on nearly two decades NIH and CDC funded research that generated modeling tools now being used to support the development of HIV policy by the CDC and local health departments. The major substantive problem that this project addresses is the isolated nature of many mathematical models with respect to modeling diseases or populations independently, limited forms of connection to realistic behavioral or clinical data for parameterization, and with disjointed connections between the epidemic simulation and economic analyses of interventions. EpiModel 2.0 will achieve major advances in our robust modeling infrastructure to achieve integration across all these domains in a comprehensive open-source software platform that may be used by our research group and many others to address the fast-evolving questions of HIV/STI prevention science. Under the framework of integrated models for HIV/STI prevention science, our three specific aims are as follows. 1) Integrated Networks: We will extend EpiModel to accommodate new data forms of bio-behavioral data, including innovative sociometric network data platforms, to allow for modeling MSM and heterosexual networks with greater detail relevant to intervention targeting and scale-up; 2) Integrated Epidemiology: We will design new EpiModel modules to handle the co-circulation of HIV and related bacterial and viral STIs, including gonorrhea, chlamydia, syphilis, and herpes simplex virus, including within-host pathogen evolution to address issues of drug resistance; 3) Integrated Economics: We will develop a new cost-effectiveness analysis software that works seamlessly with EpiModel, specializing in economic analyses of prevention interventions under conditions of uncertainty in defining costs, health utilities, and information valuations. Our major infrastructural components include: 1) comprehensive methods for model calibration and validation using a flexible Bayesian framework; 2) rigorous, open-source software development, testing, and dissemination; and 3) ongoing software and methods documentation and in-person training resources for both epidemiological researchers and public health policymakers. The completion of EpiModel 2.0 will represent the advancement of our robust epidemic modeling software platform for new user bases to investigate complex modeling questions and to make data-driven public health HIV/STI prevention policy.