Four months in, Brazil’s coronavirus data still doesn’t explain the outbreak

. Jul 13, 2020
Four months in, Brazil's coronavirus data still doesn't explain the outbreak Image: Elenabsl/Shutterstock

Since the beginning of the pandemic, we at The Brazilian Report have warned readers that official data on the coronavirus outbreak in the country must be taken with a grain of salt. Doubts around the reliability of testing data have persisted — and the more we know about them, the more reason we have to be skeptical.

The case of Brazil’s highest-profile coronavirus patient helps us understand why.

On July 6, President Jair Bolsonaro took a Covid-19 test after feeling ill for a couple of days — he announced on the following day that he had contracted the virus.

This was Mr. Bolsonaro&#8217;s fourth test, with the first three were taken under aliases in March, after dozens of his aides caught the coronavirus after a trip to meet U.S. President Donald Trump in Florida.</p> <p>According to the Health Ministry&#8217;s latest coronavirus update, the country ran 1,147,408 RT-PCR tests — molecular tests that diagnose <a href="">active infections</a> — until July 4. But the numbers don&#8217;t reveal exactly how many individuals took these tests. There is no way of knowing how many overzealous patients took multiple exams, as Mr. Bolsonaro did. While they could be in the small minority, not having any clear data on this point further clouds the view on Brazil&#8217;s true testing capacity. We know that several families in higher-income cities rushed to get themselves tested as fears of the outbreak rose —&nbsp;while poorer communities have less access to healthcare.</p> <p>Official data shows an exceptionally high proportion of tests coming back positive: 38 percent. That is a testament to how deficient the country has been in monitoring the spread of the virus.</p> <div class="flourish-embed flourish-chart" data-src="visualisation/3164903" data-url=""><script src=""></script></div> <p>Since the start of the pandemic, Brazilian authorities made a decision to prioritize testing severe patients, due to the scarcity of materials. &#8220;Focusing on hospitalized patients — or those already showing symptoms of acute respiratory syndrome (ARDS) — reduces the sample and increases the rate of positive tests,&#8221; warns Marcelo Gomes, coordinator of the Infogripe system, which tracks <a href="">ARDS data in Brazil</a>. &#8220;Mass testing is what brings us closer to knowing what is really going on,&#8221; he told <strong>The Brazilian Report</strong>.</p> <p>Another problem that diminishes the accuracy of Brazil&#8217;s testing data was the government&#8217;s decision to recommend people not to seek medical attention unless they were experiencing severe symptoms. The RT-PCR test — which the World Health Organization (WHO) considers to be the most definitive —&nbsp;has a higher chance of a negative result as days go by. Viral tests do not indicate whether someone has been infected in the past.</p> <h2>Lifting quarantine measures</h2> <p>One of the WHO criteria governments should follow before reopening the economy is a low rate of positive test results. &#8220;Less than 5 percent of samples positive for Covid-19, at least for the past 2 weeks, assuming that surveillance for suspected cases is comprehensive,&#8221; says the organization guidelines. In Brazil&#8217;s case, surveillance is far from being comprehensive.</p> <p>&#8220;We need to set the record straight: the total number of reported cases is not the total number of cases within the population — but rather the number of <em>severe</em> cases. Thinking otherwise could give a false sensation that the problem is less severe,&#8221; says Mr. Gomes.</p> <p>When testing is restricted to severe cases, this impacts the reliability of mathematical models to project the evolution of the pandemic in the country. Mr. Gomes says that skewed data is even more worrisome as states lift restrictions on non-essential activities. On Sunday, Brazil&#8217;s largest newspaper <a href="">published</a> an article on its homepage entitled &#8220;Covid-19 curve suggests higher [collective] immunity and low probability of second wave.&#8221; But it might be too soon for a victory lap.</p> <p>Cities such as São Paulo, Manaus, Rio de Janeiro, or Recife are seemingly experiencing a stabilization of new cases — but the outbreak is spreading fast in the interior of the country.&nbsp;</p> <div class="flourish-embed flourish-map" data-src="visualisation/3373211" data-url=""><script src=""></script></div> <h2>Coronavirus testing in other South American countries</h2> <p>Argentina and Colombia, South America&#8217;s most populous nations after Brazil, have also posted elevated rates of positive tests according to data aggregation platform <a href="">Our World in Data</a>. Those two countries, however, have implemented much stricter social isolation measures within their territories.</p> <p>Early in July, the Buenos Aires Metropolitan Area (AMBA) — home to 94 percent of new cases recorded in Argentina on July 12 — has once again closed non-essential services and restricted the use of public transport after cases rose once again in June, after quarantine measures had been relaxed.&nbsp;</p> <p>In Colombia, President Ivan Duque confirmed that quarantine will be enforced at least until August 1 —&nbsp;leaving open, however, the possibility of more leniency to less-affected areas. Today, capital city Bogota was once again placed under strict isolation.</p> <p>That is in stark contrast to São Paulo, where <a href="">bars, restaurants</a>, and gyms have been reopened.

Aline Gatto Boueri

Aline Gatto Boueri is a data journalist. She has had her work published by Gênero e Número, Universa UOL, Marie Claire, Projeto Colabora, among others.

Our content is protected by copyright. Want to republish The Brazilian Report? Email us at