Individual-based Simulation of Seasonal and Pandemic Influenza Epidemics

Project Abstract Seasonal influenza caused an estimated 48.8 million illnesses, 22.7 million health care visits, 959,000 hospitalizations, and 79,400 deaths in 2017-18. Influenza is a particularly complicated disease to prevent as the influenza virus itself changes over time, and unlike vaccinations for measles or hepatitis B, the effectiveness of annual vaccination may be limited if the particular strain that appears in an outbreak is not represented in the vaccine. In addition, the immunity produced by vaccination declines over time, and many individuals, especially the elderly, may lose protection toward the end of an influenza season. The goals of iMPH: Influenza Modeling for Public Health are to bring together four experienced modeling groups including the Public Health Dynamics Laboratory (PHDL) that has significant experience as a MIDAS Center of Excellence in the agent-based modeling of influenza, the Pittsburgh Vaccine Research Group (PittVax), vaccine policy experts who are site leads in both the inpatient and outpatient CDC influenza vaccine effectiveness (VE) networks, the DELPHI (Developing the Theory and Practice of Epidemiological Forecasting) group at CMU, the most accurate US influenza prediction group, and the current MIDAS Network Coordination Center (MCC) that brings major data and model coordination expertise. Our specific Aims are to create realistic, biologically based models of influenza, including the development and maintenance of immunity over time through either the development of the disease or vaccination so that we can test the benefit of different prevention strategies in seasonal or pandemic influenza. Specifically, we will examine strategies such as: enhanced vaccines that create high initial antibody levels, the addition of a second, mid-season vaccine, or potentially delaying vaccination in some individuals until later in the season. We will also examine the effectiveness of vaccination policies over a multi-year time, such as different vaccinations for individuals who have had previous influenza, and consideration of every-other year vaccines, Finally we will examine community-level interventions, such as school closures and working from home to impact the spread of influenza over a season. As a CDC Influenza Modeling Center, we will also collaborate with other centers to develop rapid responses to current influenza threats, to share data, models and results to provide higher confidence to the CDC that model-based recommendations can be used to formulate local, state and national influenza policy.