Impact of data accuracy on the evaluation of COVID-19 mitigation policies

Michele Starnini,Michele Tizzoni,Alberto Aleta,Yamir Moreno

doi:10.1017/dap.2021.25

Abstract

AbstractEvaluating the effectiveness of nonpharmaceutical interventions (NPIs) to mitigate the COVID-19 pandemic is crucial to maximize the epidemic containment while minimizing the social and economic impact of these measures. However, this endeavor crucially relies on surveillance data publicly released by health authorities that can hide several limitations. In this article, we quantify the impact of inaccurate data on the estimation of the time-varying reproduction number$ R(t) $, a pivotal quantity to gauge the variation of the transmissibility originated by the implementation of different NPIs. We focus on Italy and Spain, two European countries among the most severely hit by the COVID-19 pandemic. For these two countries, we highlight several biases of case-based surveillance data and temporal and spatial limitations in the data regarding the implementation of NPIs. We also demonstrate that a nonbiased estimation of$ R(t) $could have had direct consequences on the decisions taken by the Spanish and Italian governments during the first wave of the pandemic. Our study shows that extreme care should be taken when evaluating intervention policies through publicly available epidemiological data and call for an improvement in the process of COVID-19 data collection, management, storage, and release. Better data policies will allow a more precise evaluation of the effects of containment measures, empowering public health authorities to take more informed decisions.

Highlights

In response to the COVID-19 pandemic, governments worldwide adopted a wide range of intervention policies aimed at containing or mitigating the spread of SARS-CoV-2
We focus on Italy and Spain, two European countries among the most severely hit by the COVID-19 pandemic
As the adoption of nonpharmaceutical interventions (NPIs) comes with significant economic losses and social disruption, evaluating the effectiveness of such policies has been a priority for public health authorities, looking for a way to minimize the economic impact while curbing the disease spread

Summary

Introduction

In response to the COVID-19 pandemic, governments worldwide adopted a wide range of intervention policies aimed at containing or mitigating the spread of SARS-CoV-2. To quantify the effectiveness of these policies and make comparative analyses across countries, most studies rely on publicly available epidemiological data released by national or international public health authorities, such as the daily number of reported cases or the daily number of reported deaths (Flaxman et al, 2020; Brauner et al, 2021). These studies rely on data regarding the timing and strictness of NPIs implemented by different countries that have been collected in public repositories (Desvars-Larrive et al, 2020; Hale et al, 2021). We show that a biased estimation of RðtÞ might have had direct consequences on the decisions taken by the Spanish and Italian governments during the first wave of the pandemic

Surveillance Data and the Time-Varying Reproductive Number

Findings

Discussion