misleading: Posts on Facebook that claim that Johns Hopkins University conducted a study that demonstrates the low effectiveness of the lockdown as a preventive measure against the spread of Covid-19 are misleading.
One of the three authors of the research is a professor at the university, but the institution has not approved any study. The work cited in the post did not go through recommended technical steps, nor was it subjected to evaluation by other researchers.
Verified content: Posts on Facebook claim that Johns Hopkins University conducted a study that concluded that lockdowns had little effectiveness in containing Covid-19.
Posts on the Facebook pages of the Brasil Sem Medo website and by federal deputy Filipe Barros (PSL-PR) erroneously claim that a study by Johns Hopkins University would have proven that the lockdown had low effectiveness as a measure to contain Covid-19. Although one of the three authors of the research is a professor at the university, the institution did not contribute to the study.
The cited article did not go through the basic protocols to be defined as a scientific academic study, such as submission to the comments and observations of colleagues in the area.
Sought by Comprova, the federal deputy said that he and his team only published an article in a magazine. The Brasil Sem Medo page, on the other hand, did not respond to our inquiries by email until the publication of this verification.
Comprova classified the post as misleading for incorrectly attributing unproven data to a renowned institution, with the intention of distorting reality and leading the reader to a wrong conclusion.
How do we check?
The first step of the verification carried out by Comprova was to certify whether the cited study really existed, through a search for keywords on the internet. After finding it, we verified the authors’ relationships with the mentioned university and what was the content of the article. The academic and professional qualifications of the authors were also researched, as well as their publications on social networks and interviews prior to the published work.
Comprova sought out the Brazilian Society for the Progress of Science (SBPC), specialists in scientific research, to analyze the way in which the study was produced and also official data on the results of public health measures.
Verification
University did not study on lockdown
The first point that draws attention is the fact that the post attributes the study to Johns Hopkins University, which counts Covid-19 cases and deaths worldwide. The study itself and the university, contacted by e-mail, state that the material disclosed is not related to the institution.
“The opinions expressed in this document are those of the authors and not necessarily of the institutions to which the authors are affiliated”, reads the opening of the article. Of the three authors, only one, Steve Hanke, is related to the university as a professor of economics. The other two, Jonas Herby and Lars Jonung, also economists, are members of institutions in Sweden and Denmark unrelated to Johns Hopkins University.
The fact that an author works for an educational institution does not guarantee that it will support any and all material that he may produce, as explained by the president of the Brazilian Society for the Progress of Science, Renato Janine Ribeiro. “For you to say that it is a study at a particular university, you have to have the approval of that university. It is not enough just to have some relationship with this university”, says the professor.
Published material has technical flaws and distortions
In addition to having little or no experience in the health area, the three economics professors who authored the document on which the verified post is based did not comply with the technical steps for the work to be considered a scientific study, and could be described as an article only.
“You do a study, it is proposed to some journal or journal, then it is evaluated, eventually criticisms arise, they ask the author to redo or analyze and respond to the criticisms and only then, at the end of that, the study is published. And only at that point can you say that the study is finished and that it is effectively a finished and published study”, explains Professor Renato Janine.
Even describing the paper as a “work in progress”, the authors were criticized by researchers whose studies served as a basis for the research. Proposed as a meta-analysis, or literature review, where the author proposes to condense a large amount of data and information collected by other authors in order to obtain a definitive data, or at least make an analysis of the collected data in comparison to the other studies, the work cited in the post has little breadth and diversity. Of the 34 studies used as a basis, 12 have the same “unfinished work” status, 14 were also done by economists on a public health topic, and seven use the same database. The main article used in the research actually points out benefits to the lockdown.
On social media, the author of that core study said that the three “they already had their thesis ready” and who just put data and research to confirm this thesis without really going into the works.
Another problem pointed out in social networks by experts in data analysis is the way the research authors used tables and data. Economist at the University of Munich, Andreas Backhaus wrote that “gave disproportionate weight to jobs with the result they wanted” distorting the final number, information that the author himself recognizes in this document with answers to questions he published on his networks.
The exclusion of data without clear technical criteria also caught the attention of the epidemiologist at the University of Wollogong in Australia, Gideon Meyerowitz. He works with chronic diseases in Sydney and pondered that the authors “deleted all known robust works on lockdown, leaving only obscure works”which also shows a willingness on the part of the authors to use numbers that reached a determined result before the analysis.
Sanitary measures were effective against Covid
Contrary to what the checked article suggests, several studies point to measures to contain the coronavirus as successful, if well applied. In addition to the lockdown, social distancing, the use of masks and hand hygiene were identified as positive measures in the fight against the spread of the disease.
A 2020 study by Imperial College, London, already pointed out that about 120,000 lives had been saved with the restriction measures in Europe. In the same year, the World Health Organization also released a study with the benefits of the lockdown around the world.
The magazine Nature, on the other hand, published extensive material in which it points out that lockdowns have reduced an average of 80% of transmissions of the virus. Two mathematics professors at the University of Barcelona estimated a drop in the number of cases and hospitalizations as an effect of restrictive measures, including the lockdown. Produced by experts from different fields, this study calculated the impact of the lockdown depending on its intensity and duration. All publications were submitted in pairs and only after being reviewed were they published.
In a joint letter, the Brazilian Society of Infectious Diseases (SBI) and the Brazilian Medical Association (AMB) addressed the effectiveness of restrictive measures imposed in other countries and the need to repeat them in Brazil. The International Journal of Infectious Diseases, on the other hand, published a study showing the reduction in the number of cases thanks to the use of masks, social distancing and the lockdown.
Why do we investigate?
Comprova checks suspicious content related to public policies of the federal government, pandemic and elections that have gone viral on social media. The verified content, which posted incorrect data about the pandemic to try to discredit the health measures, had almost 15,000 interactions on the two Facebook posts in just three days.
On other occasions, Comprova showed how effective health measures were in combating the proliferation of the coronavirus, such as when showing that video used false information about lockdown, or when it verified a video released that falsely attributed suicide to the lockdown. We also check information on social distancing and the effectiveness of masks here, and also in this and this Comprova check on the use and efficiency of masks.
Misleading, for Comprova, is content taken from the original context and used in another so that its meaning is altered; that uses inaccurate data or that leads to an interpretation different from the author’s intention; content that confuses, with or without the deliberate intent to cause harm.
Investigated by CNN Brasil. Verified by Estadão, Jornal do Commercio, Grupo Sinos and Nexo
Source: CNN Brasil
