Modeling the impact of racial and ethnic disparities on COVID-19 epidemic dynamics | Life where S, E, I, R refer to the number of people in susceptible, latently infected, infectious, and recovered compartments, respectively. Given a mean incubation period and mean serial interval of 5 days as suggested by empirical studies (Nishiura et al., 2020; Lauer et al., 2020), we set the mean latent period 1 / r to be 3 days to allow for pre-symptomatic transmission and the mean infectious ...
elifesciences.org
The idea will be to repurpose and extend that model to address new research questions. Ideas for these extended uses include:
Calibration of the model to alternate sets of data that describe COVID outcomes that disproportionately affected a certain group or groups
Explore the effect of assuming that general transmission mitigation efforts had different effectiveness for different groups (this was not one of the possible mechanisms explored in the publication)
Use the model to explore the effects of interventions designed to reduce inequities, especially for vaccination
Emma/Damon will code the model and extensions with a differential equation (DE) numerical solver and other DE analysis techniques similar to the publication.
Additional idea (to be discussed with George): in parallel with the DE solver, we could perhaps code an agent-based model that initially uses the same assumptions as the DE model but simulates transmissions stochastically among the agents. This would give us a stochastic version to play with, and would then be extendable to adding additional features that are difficult to represent in a DE model, such as adding household structure to the population.
Here is part of the email with the com