Covid-19: 'soft' measures can avoid second peak


Masks and physical distancing can effectively replace the lockdown, averting a second wave of the epidemic, in Lombardy as in London.

Two Italian economists have come to this conclusion by considering both the official data of infections, recoveries and deaths, and the numbers, more difficult to estimate, of unobserved cases (at least double those surveyed) and of undetected Covid-19 deaths (35% more than the official figure in Lombardy, 17% more in London). The study was published in recent days in Covid Economics, a special publication by the Centre for Economic Policy Research.

“The epidemic model we propose is appreciated in Lombardy and London, two regions particularly affected by the virus - explains Dario Palumbo, “Carlo Giannini” Fellow at the Department of Economics of the Ca’ Foscari University of Venice and co-author of the study with Salvatore Lattanzio of the University of Cambridge - also takes into account unobserved states and mobility policies and predicts the evolution of the disease according to different policies. We show how mitigating the probability of contagion with ‘soft’ measures, reducing it by up to 20/40% compared to a scenario without measures, has positive effects comparable to those of an extension of the lockdown".

For the researchers, it is clear that official statistics have underestimated cases and deaths. For this, they developed a mathematical model that predicts four possible states of people with respect to the epidemic (susceptible, exposed, infected and deceased), but introducing two types for infected, cured and deceased: observed and unobserved.

At the end of the period on which the model was tested (9 April in Lombardy and 15 April for London), they estimate that 5.7% of Lombards and 2% of Londoners had been infected. It means that the ‘unobserved’ would have been double the cases reported by the statistics, that the recovered would have been between 20 and 26 times more than those surveyed and that the number of deaths from Covid-19 has been underestimated by 35% in Lombardy and 17% in London.

Thanks to the model, then, the economists have calculated scenarios of a gradual restart of mobility, assuming a recovery in the movement of people up to 75% of the pre-pandemic level.

“Without any containment measure, we see a second peak of the epidemic and a resumption of deaths as inevitable - says Palumbo - however, by acting on the probability of contagion, the second peak becomes less likely. In particular, by reducing this probability by 40% in Lombardy and between 20 and 30% in London, the death toll is back in line with that of a permanent lockdown".

The removal of the lockdown restrictions, the research shows, does not imply a resumption of the epidemic curve in the presence of active policies that promote the reduction of the probability of infection such as physical distancing, masks, better hygiene and isolation of infected cases.

“Not adopting these mitigation measures - concludes Palumbo - means risking a second peak even in scenarios where the lockdown is eased very gradually, as happened in Italy”.

The novelty of the study also lies in the methodology developed by the two economists, which applies a statistical technique called Kalman Filtering to a SEIRD (Susceptible-Infected-Exposed-Recovered-Dead) epidemiological model, which dynamically adapts the estimates by correcting them for each point in time even for unobserved cases. The more common alternative is to assume that the 'submerged' cases are a certain fixed proportion of the observed. Adjusting this calculation over time reduces the uncertainty of the results.