Get best of the week

How the Ebola virus taught us how to deal with infection data, and we forgot all his lessons

My name is Roman Nester , I am a professor at HSE. I have been doing data for the past 10 years. For example, my partners and I made a startup Segmento, which uses big data in advertising, and Sberbank bought it from us. Today I am developing a master's program in product management and marketing based on data in the communications department of the HSE. I am worried about countless articles with figures, conclusions, and decisions related to coronavirus. Do not forget - we often make mistakes due to incorrect data. An example of this is the fight against the Ebola virus in 2014. I look at the struggle with COVID-19, and it seems to me that we have not learned any of those lessons.

I want to cite a series of quotes from my favorite book, Factfulness, by the Swedish statistician and doctor, Ph.D. Hans Rosling. According to it, we at the HSE, in my opinion, made a cool course on data-based solutions. And now it dawned on me - because what is happening now is very similar to what was already happening 6 years ago. Rosling was a direct participant in those events. The feeling intensified when I read about Moscow's “ Tactics of Intimidation, ” when frightening numbers and headlines about infected people are published.

UFO Care Minute


The pandemic COVID-19, a potentially severe acute respiratory infection caused by the SARS-CoV-2 coronavirus (2019-nCoV), has officially been announced in the world. There is a lot of information on Habré on this topic - always remember that it can be both reliable / useful, and vice versa.

We urge you to be critical of any published information.


Official sources

, .

Wash your hands, take care of your loved ones, stay at home whenever possible and work remotely.

Read publications about: coronavirus | remote work

image

Bill Gates called Rosling's book one of his favorites

Ebola


In 2014 , an Ebola epidemic erupted in Liberia . It is a highly mortal contagious disease that kills every second. From January to March, time was lost - no one took the figures seriously. In March, it became clear that the epidemic could not be stopped. International professionals flew to Africa. Everyone then just talked about the “exponent” (now, thanks to covid-19, any inhabitant learned about it) - this is how the infection schedules and, most importantly, mortality charts looked like.

They introduced quarantine, closed institutions, and introduced unprecedented restrictions and measures to improve hygiene in West Africa. Despite everything, the charts for the number of infected grew. They were published in regular releases and the whole world began to follow them. It seems that the measures taken did not work.

Rosling decided to figure out what was going on. The fight against Ebola involved the US federal agency CDC (Centers for Disease Control and Prevention) and WHO (WHO). It was they who published regular data on “infected people,” calling them “suspected cases”.

The charts of those who died from the Ebola virus, for example, included those patients who were brought with suspicion of Ebola, but then they died from other causes. The more the fear of illness increased, the more people were admitted to the hospital “suspected” and taken into account in the main statistics of “infected” (total cases).
“The farther, the more these growing curves of“ sick people ”scared us and the whole world, and the less we understood the real number of confirmed cases. If you cannot measure progress, you cannot even understand if your containment measures are working. "Rosling recalled.
image

Data


The doctor took confirmed cases from the Liberian Ministry of Health. It turned out that patients who received “with suspicion” were taken blood and sent to 4 laboratories. The results of their analysis came in Excel tables. But then there was no one responsible for compiling these tables! Everyone was only interested in how many new “suspects” continued to enter hospitals. Rosling took this data, cleaned it of takes and brought it together.
“ , “!”. . , , 2 . !”
People in Liberia have radically changed their behavior. They stopped shaking hands, avoided contact, and shops and public places closed. And it turned out that as a result, real infections and deaths from Ebola began to decline! But at the same time, frightened people continued to run to hospitals at the first fears. And they got into the statistics of all the new “suspects”!

“The strategy worked, but nobody knew about it before my data analysis. It gave us strength and really inspired us! “, Recalls Rosling.

Fear


He sent his schedule to WHO and the CDC. To Rosling's surprise, the CDC refused to publish it in the next release! The bureau insisted on publishing the schedule using the previous erroneous methodology with a frighteningly growing number of “hospitalized with suspected Ebola.” And then Rosling realized - the whole point is that the CDC believed that they should maintain a sense of urgency among those who were responsible for the allocation of resources.
“ , . , ! , . , , , — . — . “ ” , ( — CDC — 10 ). - !” —
image

The data helped to understand how dangerous the Ebola virus is. It was thanks to the exponential graphs that he received that Rosling dropped all business and came to Liberia. However, as soon as there was evidence that the virus could be defeated, they simply did not publish them!

Such selective use and publication of data deprives the request for the accuracy of their collection and processing of any meaning! The CDC reports on neighboring Sierra Leone were then also claimed by WHO itself, seeing crazy peaks in statistics due to a combination of heterogeneous databases. CDC have changed the accounting methodology. However, in Liberia, so far all sick and suspected of the disease are counted “in a heap” - a huge mistake is still in sight.

image
The CDC Report on Liberia - it still counts all together - both confirmed cases and those with suspicion.
The credibility of data and those who publish it must be protected. Data should tell the truth, not call to action, no matter how noble intentions are. Instinct “We must urgently do something! Analysis - then, act now! ” “This is one of the worst instincts.”

Lesson not learned


“We need to create fear” are the words of former US Vice President Al Gore. He told them to Rosling when he asked him to illustrate with his schedules the danger of CO2 emissions for his next workshop. At the same time, Gore refused the alternative schedules proposed by Rosling, which showed positive forecasts and improvements. The Swedish professor insisted on a more objective picture, but only an exaggeration of the danger could give Al Gore the expected effect. Rosling went on the principle and refused to help one of the strongest in this world.

Unfortunately, we will never know what Rosling would say, looking at media frenzy today - when every second journalist spins panic more and more, attracting new unverified data and drawing new scary graphics. Despite all the reservations of scientists and institutions that we can be very wrong now when we look at these numbers! And despite their reminders that many data has not yet been cleared and confuse us - this does not stop anyone. Panic continues to spin the world and data only feeds it.

The danger is that a multimillion-dollar audience is studying graphs based on extrapolations from inaccurate data and they have the illusion of awareness. People do not study the nature of data, do not check their sources. Charts and conclusions travel from publication to publication, without the original reservations, lists of assumptions, or alternative scenarios. The press selectively retains only vivid provocative pictures and conclusions that attract attention, removing all unnecessary and ambiguous.

This does not mean that we must ignore the loss of life and neglect the danger. But we need to be more careful and objective in the way we act. And for sure - don't let the data cause a panic.

Hans Rosling died of cancer in 2017. He left a magnificent book of Factfulness, of which I became an ardent admirer. We will use it now in our graduate program as part of our course. I want to share the main conclusions that we usually fix in the final of the course:

  • Inhale. When your urgent instinct turns on, your ability to analyze disappears. Ask for more data, check them.
  • Avoid relevant but inaccurate data.
  • Beware of the “prophets" and do not rely on unambiguous scenarios. Do not leave for your analysis only the best or only the worst option
  • Beware of radical steps - evaluate the possible harm. Less dramatic actions are often more effective.

Sources:

Factfulness: Ten Reasons We're Wrong About the World - and Why Things Are Better Than You Think, ISBN-10: 1250107814

CDC [3]. “Ebola Outbreak in West Africa — Reported Cases Graphs.” Centers for Disease Control and Prevention, 2014. gapm.io/xcdceb17.

WHO [3]. WHO Ebola Response Team. “Ebola Virus Disease in West Africa — The First 9 Months of the Epidemic and Forward Projections.” New England Journal of Medicine 371 (October 6, 2014): 1481–95. gapm.io/xeboresp.

More articles:


All Articles