Adjusted age-specific case fatality ratio during the COVID-19 epidemic in Hubei, China, January and February 2020

Julien Riou; Anthony Hauser; Michel J Counotte; Christian L Althaus

The crude case fatality ratio (CFR), the number of reported deaths divided by the number of reported cases at a specific time point, is not a reliable indicator of overall mortality [8, 2] . Other reported measures such as the confirmed, symptomatic or hospitalized CFR, based on different denominators, are sometimes used, adding confusion. The crude CFR can be misleading if used to assess the overall mortality because of two opposing biases [9, 10] . First, because of the delay of several weeks between disease onset and death, the number of confirmed and reported Covid-19 deaths at a certain time point does not consider the total number of deaths that will occur among already infected individuals (right-censoring). Second, surveillance-based Covid-19 case reports underestimate the total number of SARS-CoV-2-infected patients, because testing focuses on symptomatic individuals, and, among symptomatic cases, on patients with more severe manifestations (preferential ascertainment). It is often assumed that the crude CFR overestimates true mortality [11] , but it depends on the relative influence of the two opposing biases, so it could differ in different situations.