Uncertainty and error in SARS-CoV-2 epidemiological parameters inferred from population-level epidemic models

• DOI: 10.1016/j.jtbi.2022.111337
• Publication Date: 2022-11-06T22:45:43Z
• Authors: Dominic G. Whittaker; Alejandra D. Herr...; Maurice Hendrix; Markus R. Owen; Leah R. Band; Gary R. Mirams; Kirsty J. Bolton; Simon P. Preston
• Keywords: Pandemic; Epidemic model; Basic reproduction number

During the SARS-CoV-2 pandemic, epidemic models have been central to policy-making. Public health responses shaped by model-based projections and inferences, especially related impact of various non-pharmaceutical interventions. Accompanying this has increased scrutiny over model performance, assumptions, way that uncertainty is incorporated presented. Here we consider a population-level model, focusing on how distributions representing host infectiousness infection-to-death times are modelled, particularly inferred characteristics if these mis-specified. We introduce an SIR-type with infected population structured 'infected age', i.e. number days since first being infected, formulation enables be consistent clinical data. show inference based simpler without age, which implicitly mis-specify distributions, leads substantial errors in quantities relevant policy-making, such as reproduction quantification via Bayesian approach, implementing for both synthetic real data UK period 15 Feb-14 Jul 2020, emphasising circumstances where it misleading neglect uncertainty. This manuscript was submitted part theme issue "Modelling COVID-19 Preparedness Future Pandemics".