A pandemic before it happens: this is how we will make increasingly accurate predictions

The pandemic has left the scientific community with the awareness that only maximum collaboration between research centers can – and will – make a difference in defeating viruses and pathogens. Forecasting is the watchword that summarizes the role of RespiCast, the first European hub capable of developing probabilistic simulations from one to four weeks, which will be managed by Isi Foundation, the Institute for Scientific Interchange based in Turin, the prestigious private research center, supported by the Crt Foundation, which deals with data science and public health.

The Hubs that simulate probabilities

Last year Isi launched Influcast, a hub that analyzes the evolution of flu-like syndromes, aggregating predictions from multiple models, with monitoring on a national and regional scale. It is this “eye” that pushed the ECDC, the European Center for Disease Prevention and Control, together with the London School of Hygiene & Tropical Medicine, to choose the center to manage the new platform that works at a continental level.

“The idea is to provide information for public health decision makers on the epidemiological situation of respiratory diseases, in particular influenza in European countries, so as to allocate adequate beds in hospitals and alert doctors”, he explains Daniela PaolottiSenior Research Scientist at the Foundation and coordinator of the con platform Nicolò Gozzi And Corrado Gioannini: “RespiCast follows what was done during the pandemic, with the Forecasting Hub, but it is a significant advancement in terms of visualizations, forecasting capabilities and data available. Our president is Alessandro Vespignanione of the founding fathers of the study of pandemics”.

Collaboration with healthcare facilities

Central to RespiCast is the collaboration of the healthcare facilities in the countries adhering to the platform. By providing updated data in real time, the center’s researchers can develop probabilistic predictions on flu-like syndromes, acute respiratory infections and Covid-19-related indicators.

“Every week the surveillance systems of the 28 European countries communicate the data to us. Obviously no model is perfect and for this reason six teams use different models, so as to reduce the margin of uncertainty”, highlights Daniela Paolotti. If for the flu monitoring coincides with its seasonal spread, for Covid there is no pause, even if “the data is declining in terms of quality, because almost no one takes swabs anymore”.

Meanwhile, another one is emerging from the RespiCast experience: it is Scenario Hub, a platform that will not only provide forecasts, but actual scenarios. “It will be possible to simulate the effectiveness of a vaccination campaign and the consequences of the different types of intervention decided by the health authorities.”

v