Joseph Whitmeyer
$599,995
Benjamin Lopman
Samuel Jenness
Mark A Schmidt
Trustees of Boston University
Massachusetts
Mathematical and Physical Sciences (MPS)
In this project, the estimation of epidemic quantities and the assessment of intervention impacts are improved by incorporating realistic behavioral feedback into mechanistic models. Mechanistic mathematical models are powerful tools to understand the epidemiology of infectious diseases, evaluate the impact of various control measures, and forecast the trajectory of cases and deaths. Lack of empirical data on social contact patterns and their temporal variation, however, has hindered progress and application of these models. Specifically, there is a dearth of social contact data from individuals with acute infections and their close contacts, such as household members, as behavior data have traditionally been collected from healthy individuals. Consequently, transmission models for acute infections often rely on simplistic assumptions that infected individuals perfectly isolate themselves or that contact patterns remain unchanged throughout the infectious period. These unrealistic assumptions in model structure and parameterization can introduce bias into the results. To address this issue, in this project the temporal changes in contact patterns among individuals with acute infection and their close contacts is measured. Long-term benefits of the project include the building of capacity in infectious disease modeling, thus providing decision makers and public health officials with more informed decision-making tools to develop interventions. To achieve research objectives, clinic-based approaches are employed to recruit U.S. cases with acute respiratory infections and acute gastroenteritis and household members of these cases. Daily changes in their contact patterns are captured over a two-week period using an online contact diary and how these patterns vary by disease severity is assessed. Household members of these cases are recruited to evaluate how close contacts modify their behaviors following exposure. This enables development of transmission dynamic models (susceptible-infectious-recovered models and appropriate elaborations), which are structured and parameterized based on the behavioral realism investigated in this study. SARS-CoV-2 is used as an example of acute respiratory infection and rotavirus is used as an example of acute gastroenteritis. How the estimation of key model parameters as well as the impact of interventions differ (i.e., improve) after incorporating behavioral feedback into these mechanistic models is quantified. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.