COVID-19 - Global Mix / Investigation of COVID-19 Disease Parameters for Transmission Models in Low-Resource Settings

PROJECT SUMMARY - We began to quantify household- and community-level interactions in 2019 with our project, "Comprehensive Profiling of Social Mixing Patterns in Resource Poor Countries" ("GlobalMix", grant R01 HD097175-01) to investigate human-to-human interactions relevant for respiratory infection transmission. This proposal will build on existing GlobalMix study infrastructure to estimate LMIC-specific epidemiologic parameters for COVID-19. In the proposed study, we will connect field epidemiology and mathematical modeling approaches by estimating the rate of, and heterogeneity in, household-based transmission of SARS- CoV-2 through longitudinal cohort approaches. We will use this information in conjunction with highly-granular data on social interactions from GlobalMix to identify key epidemiological parameters for COVID-19, including the community-level force of infection and attack rates within households. We will then use this information to build LMIC-specific dynamic models, to evaluate the impact of key interventions to reduce transmission: vaccination and non-pharmaceutical interventions such as face masks, shelter-in-place policies and school closure. This work will be completed in three specific aims: Aim 1: Quantify COVID-19 transmission across contact networks within the household environment. We will conduct longitudinal respiratory disease surveillance in households participating in the GlobalMix study. We will collect longitudinal samples of respiratory specimens from household members for identification of COVID-19 and other respiratory pathogens such as influenza. This information will be overlaid on contact network data from GlobalMix. Aim 2: Estimate key epidemiological features of SARS-CoV-2 and other respiratory pathogens in LMIC settings. We will collect blood specimens from GlobalMix study participants and test for antibody levels (IgG) against SARS-CoV-2. We will calculate age-specific infection fatality rates (IFRs) and use antibody titers to infer time of infection and calculate community-level incidence over time. We will generate age-structured seroprevalence curves, which will provide a robust measure of exposure across the age range. Together with the contact data from GlobalMix, we will infer age-specific transmission probabilities that will be used as inputs into the network models in Aim 3. Samples will be stored for future testing, including antibody avidity and T/B cell activation. Aim 3. Estimate the impact of control measures on COVID-19 in LMIC. We will use the epidemiological parameters estimated in Aim 1 and the setting- and age-specific force of infection estimates from Aim 2 to parameterize dynamic network-based mathematical models of disease transmission. Models will incorporate social mixing data from GlobalMix to project the impact of extended shelter-in-place policies, policies concerning the use of face masks, and the introduction of a SARS-CoV-2 vaccine.