Speaker: Nathan Duarte, McGill University

Date and Time: Wednesday, March 29, 2023 - 2:00pm to 3:00pm

Abstract

Serosurveillance allows accurate measurement of the extent of infection during a pandemic, but it is not a timely indicator because SARS-CoV-2 seroconversion only occurs ~2 weeks after infection and serosurvey results are often released months after data collection. The concentration of SARS-CoV-2 in wastewater is a more timely metric, which has been used to identify regions with a relatively high prevalence of infectious individuals. Given their complementary characteristics, wastewater and serological data could be jointly analysed to develop timely and unbiased estimates of the incidence of infection within a population. We fit a Bayesian model in which the anti-nucleocapsid (anti-N) SARS-CoV-2 seroprevalence and the concentration of SARS-CoV-2 in wastewater are generated from an underlying, unobserved infections process. We estimated that up to ~10% of the population was infected per week during the first Omicron wave. The inclusion of wastewater data increased confidence in our estimates. The statistical relationship between anti-N SARS-CoV-2 seroprevalence and the concentration of SARS-CoV-2 in wastewater was stable until after the first Omicron wave, after which the nature of the relationship appeared to change.