Study Criticizes Portugal’s Health Directorate Covid-19 Data Used in Scientific Analysis
RIO DE JANEIRO, BRAZIL – Missing victims, too many or too few infections that do not match the numbers released publicly by Portugal‘s General Health Directorate (DGS) itself, and thousands of cases incorrectly entered into the National Epidemiological Observation System.
A study signed by 12 researchers from the Faculty of Medicine of the University of Porto, recently published in MedRxiv international scientific journal, concludes that the National Epidemiological Observation System (SINAVE) databases that have been provided to the scientific community in recent months on Covid-19 cases are of poor quality, contain inaccuracies, inconsistencies and a great deal of missing information.
To aggravate the matter, past problems have led to other studies, conducted based on these data, to produce conclusions that may be incorrect, namely in terms of the higher or lower risk for certain patients, with preexisting chronic diseases, to suffer additional complications should they contract the novel coronavirus.
The article warns that, consequently, the data provided by the DGS to academic researchers can only be of very limited use and unhelpful to help curb the pandemic.
The list of flaws is long and includes ludicrous details and others that uncover deeper structural shortcomings in the part of the SINAVE database – the section that is filled in by doctors – the only one that the DGS agreed to send to researchers.
For instance, cases of a 134-year-old patient and three men classified as “pregnant” as well as 19 patients who were reported to have had the disease before the first known case was diagnosed in Portugal.
In addition, there are months with far fewer patients than those reported in the DGS daily updates and months with far more… but also months (May) with half of the deaths publicly reported and others (June) with zero victims – when it is known from the daily reports that the pandemic killed 155 people that month.
On the other hand, 90 percent of the cases in the DGS database do not provide the date for the positive test and a large part fail to mention whether or not the infected person needed intensive care.
Among the two databases submitted to researchers, one in April and another in August, there was a great deal of differing information for the same patients in thousands of cases, particularly in terms of preexisting conditions (8,902 cases).
In the first database, 40 percent of infected patients were identified as not having preexisting chronic diseases, but in the second, the very same people were reported as patients with undetermined preexisting diseases.
Cristina Costa Santos, one of the 12 researchers at the Center for Research in Technologies and Health Services (CINTESIS) and the Community Medical Department of the University of Porto, explains that they were even more concerned when they found out that there are at least three scientific articles already published that use data provided by the DGS, and in at least one of them the results were clearly “biased due to poor data quality”.
The researcher understands that the DGS has decided to provide the academic community with data from only one of the databases with information on Covid-19 infections and patients, but this does not explain so many flaws, a concern that is growing, given, as she recalls, that it is this type of study that often helps policymakers to proceed with decisions to control the pandemic.
If the data are analyzed “without taking into account the quality of the data, this is dangerous and can produce biased scientific articles, as has already occurred, but it is much more serious than this because it is these studies that lead to decisions on how to manage the pandemic,” says Cristina Costa Santos.
Graça Freitas, DGS Director, concedes that “the data deal with epidemiological monitoring” and, therefore, are not perfect, while pointing out that there are other priorities at this time. “The main priority now is to detect patients, treat patients, and isolate contacts,” she said.
Source: TSF
More: Brazil news in English, every day from The Rio Times.
This article was produced by The Rio Times’ automated AI newsroom. Editorial responsibility: Matthias Camenzind, Publisher and Editor-in-Chief. The Rio Times also runs Ask Rio Times, which answers from 100,000+ of our own articles, the daily Latin America Power Map, and prediction-market odds on elections and markets.. How we use AI · Report an error
Brazil runoff, Sunday, October 25: Brazil runoff: date, poll closing times and when results start. Latest: Brazil runoff polls, Flávio 51%, Lula 49%.