Cumulated burden of Covid-19 in Spain from a Bayesian perspective

TL;DR

This study uses a Bayesian model to estimate the true number of Covid-19 cases in Spain during early 2020, revealing the actual burden was much higher than reported and emphasizing data quality's importance for pandemic response.

Contribution

It adapts a hierarchical Bayesian model to estimate unreported Covid-19 cases in Spain, providing more accurate infection counts and lethality estimates during the initial pandemic wave.

Findings

Estimated 2.4 million cases in Spain by June 2020

Actual cases were significantly higher than registered cases

Model estimates closely match seroprevalence study results

Abstract

The main goal of this work is to estimate the actual number of cases of Covid-19 in Spain in the period 01-31-2020 / 06-01-2020 by Autonomous Communities. Based on these estimates, this work allows us to accurately re-estimate the lethality of the disease in Spain, taking into account unreported cases. A hierarchical Bayesian model recently proposed in the literature has been adapted to model the actual number of Covid-19 cases in Spain. The results of this work show that the real load of Covid-19 in Spain in the period considered is well above the data registered by the public health system. Specifically, the model estimates show that, cumulatively until June 1st, 2020, there were 2,425,930 cases of Covid-19 in Spain with characteristics similar to those reported (95\% credibility interval: 2,148,261 - 2,813,864), from which were actually registered only 518,664. Considering the results obtained from the second wave of the Spanish seroprevalence study, which estimates 2,350,324 cases of Covid-19 produced in Spain, in the period of time considered, it can be seen that the estimates provided by the model are quite good. This work clearly shows the key importance of having good quality data to optimize decision-making in the critical context of dealing with a pandemic.