A0603
Title: Using wastewater data for COVID-19 surveillance in the post-pandemic era: A data integration approach
Authors: Guangquan Li - Northumbria University (United Kingdom) [presenting]
Peter Diggle - Lancaster University and University of Liverpool (United Kingdom)
Marta Blangiardo - Imperial College London (United Kingdom)
Abstract: During the COVID-19 pandemic, wastewater-based epidemiology (WBE), a suite of methods to detect and measure viral contents in wastewater, has been recognised as an efficient surveillance tool to monitor the disease. WBE is used in the post-pandemic setting, where data collection via national randomised surveys is run at a reduced scale, but wastewater data, a spatially refined and low-cost metric, can be used to complement the reduced health data for cost-effective disease monitoring. Using data collected from a network of sewage treatment works, a geostatistical model is constructed to predict wastewater viral load at a fine space-time scale for the whole of England. A data integration framework is developed to combine these viral predictions with prevalence, estimated using data collected through randomised surveys and community testing. The data integration framework aims to produce prevalence nowcast at a fine spatial scale when prevalence estimates can only be derived at a coarse spatial level due to the scaled-down health data collection. The results from the cross-validation demonstrate the added values of wastewater data, not only improving the accuracy of the prevalence nowcast but also reducing the nowcast uncertainty. The investigation also highlights the critical role of the coarse-level prevalence estimates in anchoring the wastewater data, thus calling for the need to maintain some form of reduced-scale national prevalence survey in the non-pandemic periods.
