Back

Forecasting COVID-19 infection trends in the EU-27 countries, the UK and Switzerland due to SARS-CoV-2 Variant of Concern Omicron

Gerli, A. G.; Centanni, S.; Soriano, J. B.; Ancochea, J.

2021-12-16 respiratory medicine
10.1101/2021.12.16.21267785 medRxiv
Show abstract

BackgroundOn November 26, 2021, WHO designated the variant B.1.1.529 as a new SARS-CoV-2 variant of concern (VoC), named Omicron, originally identified in South Africa. Several mutations in Omicron indicate that it may have an impact on how it spreads, resistance to vaccination, or the severity of illness it causes. MethodsWe used our previous modelling algorithms to forecast the spread of Omicron aggregated in the EU-27 countries, the United Kingdom and Switzerland, and report trends in daily cases with a 7-day moving average. We followed EQUATORs TRIPOD guidance for multivariable prediction models. Modelling included a third-degree polynomial curve in existing epidemiological trends on the spread of Omicron in South Africa, a five-parameter logistic (5PL) asymmetrical sigmoidal curve following a parametric growth in Europe, and a new Gaussian curve to estimate a downward trend after a peak. ResultsUp to January 15, 2022, we estimated a background rate projection in EU-27 countries, the UK and Switzerland of about 145,000 COVID-19 daily cases without Omicron, which increases up to 440,000 COVID-19 daily cases in the worst scenario of Omicron spread, and 375,000 in the "best" scenario. Therefore, Omicron might represent a relative increase from the background daily rates of COVID-19 infection in Europe of 1.03-fold or 2.03-fold, that is up to a 200% increase. ConclusionThis warning pandemic surge due to Omicron is calling for further reinforcing of COVID-19 universal hygiene interventions (indoor ventilation, social distance, and face masks), and anticipating the need of new lockdowns in Europe.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.