Combining models to generate consensus nowcasts for the COVID-19 epidemic status in England #

Harrison Manley, Josie Park, Luke Bevan, Alberto Sanchez-Marroquin, Gabriel Danelian, Thomas Bayley, Veronica Bowman, Thomas Maishman, Thomas Finnie, Andre Charlett, Nicholas A Watkins, Johanna Hutchinson, Steven Riley, Jasmina Panovska-Griffiths, Sebastian Funk, Paul Birrell, Daniela De Angelis, Matt Keeling, Lorenzo Pellis, Marc Baguelin, Graeme Ackland, Jonathan Read, Christopher Jewell, Robert Challen

14:30 Tuesday in 2Q50/51.

Part of the Modelling and inference in public health session.

Abstract #

The effective reproduction number R has been widely accepted as an indicator for the status of the COVID-19 epidemic. In the UK, the R value published on the UK Government Dashboard has been generated as a combined value from an ensemble of fourteen epidemiological models via a collaborative initiative between academia and government. In this talk we outline this collaborative modelling approach and illustrate how, by using an established combination method, a combined R estimate can be generated from an ensemble of epidemiological models. We show that this R is robust to different model weighting methods and ensemble size, and that using heterogeneous data sources for validation increases its robustness and reduces the biases and limitations associated with a single source of data. We also show that R is correlated with the rate of change in COVID-19 cases, hospital admissions and deaths, albeit with a delay, suggesting R can be good indicator of future epidemic trends and a proxy for epidemic status.