Semi-parametric modeling of SARS-CoV-2 transmission in Orange County, California using tests, cases, deaths, and seroprevalence data.

Mechanistic modeling of SARS-CoV-2 transmission dynamics and frequently estimating model parameters using streaming surveillance data are important components of the pandemic response toolbox. However, transmission model parameter estimation can be imprecise, and sometimes even impossible, because surveillance data are noisy and not informative about all aspects of the mechanistic model. To partially overcome this obstacle, we propose a Bayesian modeling framework that integrates multiple surveillance data streams. Our model uses both SARS-CoV-2 diagnostics test and mortality time series to estimate our model parameters, while also explicitly integrating seroprevalence data from cross-sectional studies. Importantly, our data generating model for incidence data takes into account changes in the total number of tests performed. We model transmission rate, infection-to-fatality ratio, and a parameter controlling a functional relationship between the true case incidence and the fraction of positive tests as time-varying quantities and estimate changes of these parameters nonparameterically. We apply our Bayesian data integration method to COVID-19 surveillance data collected in Orange County, California between March, 2020 and March, 2021 and find that 33-62% of the Orange County residents experienced SARS-CoV-2 infection by the end of February, 2021. Despite this high number of infections, our results show that the abrupt end of the winter surge in January, 2021, was due to both behavioral changes and a high level of accumulated natural immunity.