What is considered a death by COVID-19 in Spain?
Deaths confirmed by COVID-19 include those who have been diagnosed prior to death. The diagnostic criteria is what has changed during the epidemic. Confirmed PCR turns the suspect into a case.
The clinical diagnosis (without PCR, but with other elements of judgment, such as the symptoms plus the typical bilateral pneumonia) escapes this statistic. Also, since April 24, positives that are not exclusively by PCR, such as serological tests, are left out.
In all cases, the communities must report deaths regardless of whether they occur in a hospital or not, since April 17, something that Catalonia began to do in its reports two days before.
Are there deaths attributable to the pandemic that escape from regional and health statistics?
Before the Ministerial Order of April 17, there was no uniform criterion, so that deaths not occurring in health centers could be left out.
Likewise, there are deaths that foreseeably would not have occurred without the emergence of the new virus and its pandemic spread. Although not easily measurable, during the peak that pushed hospitals to the limit for admissions with COVID-19, the quality of care for all pathologies could be compromised.
Likewise, the delay in appointments and scheduled surgeries could also lead to deaths from other causes, which, in part, reflects the All-Cause Mortality Monitoring System (MoMo / ISCIII).
Is MoMo a reliable measure of COVID-19 deaths?
The MoMo was created to assess the impact of heat waves on the country’s overall mortality. Powered by Fernando Simón in 2004, it became popular with the epidemiological community by giving a measure of the severity of each flu season.
Never in the history of the MoMo has there been a peak of excess mortality like that experienced during March and April 2020. In itself, there is no (or contrary) attribution to COVID-19 in absolute numbers.
It seems obvious that the two elements mentioned in the previous point are reflected in these excesses: Deaths of people with SARS-CoV-2 in their bodies. And collateral deaths from other pathologies that, under normal circumstances, would not have been fatal.
What deaths does the MoMo account for?
MoMo mainly feeds on daily death data from the National Statistics Institute (INE) and, above all, from computerized civil registries. It reaches 93% of the population. And it compares them with those expected for the same period, under normal circumstances. The technical details of its methodology can be seen here, applied to the 2012 seasonal flu.
As they explain from the National Center for Epidemiology, which manages the MoMo, “the data has a delay in the notification date. For example, deaths occurred on day 1 are reported on the following days (on 2, 3, etc.). Typically, and in normal periods, on day 7, practically all the deaths produced on day 1 are already available ».
The coronavirus epidemic has caused much longer delays, exceeding 12 days in April, they explain, hence the statistics are being corrected and completed. In the case of Barcelona, there was a data blackout between March 12 and May 10.
MoMo does not directly transfer civil registry deaths. “The model that corrects the observed deaths is conservative by construction and tends to complete downward the unobserved deaths to avoid the generation of artificial alerts,” they explain.
Are there other mathematical models to measure the mortality of the epidemic in Spain?
Recently, a team from the Polytechnic University of Madrid published a work (not published in a scientific journal) in which they estimated that April would have closed with some 40,000 deaths from COVID-19 accumulated in Spain, based on the interpretation of the MoMo.
As reported in the daily press conferences Fernando Simón, the National Epidemiology Center will process the data to prepare mortality studies that will determine the most realistic figures on the impact of COVID-19 in Spain.
Why do we distinguish between mortality and lethality?
The lethality of a virus is measured from the deaths it causes among the number of people infected. If we measure the lethality of the disease, we would only consider those people who end up developing it, not only those who are HIV positive. With preliminary seroprevalence data in Spain, the case fatality rate for SARS-CoV-2 would be around 1%.
Mortality measures the impact on the total population (or on a number of this, such as 100,000 inhabitants). And it influences future expectations. Life expectancy in Spain has been cut almost 9 months (-0.7 years), according to this preliminary work from the University of California-Los Angeles.
Carlos III Health Institute, MoMo methodological documentation, National Center for Epidemiology, report from May 10, Fernando Simón, director of the CCAES, at a press conference on April 23 and May 7, Rafael Cascón Porres, Miguel Berzal Rubio and Francisco Santos Olalla, Excess mortality due to COVID-19 Patrick Heuveline, Michael Tzen, Beyond deaths per capita, COVID-19 mortality, in MedRxiv, May 20, 2020, Ministerial Order SND / 352/2020 of April 16 . The Autonomous Communities are given a template to send the notifications and the cases are redefined.