Countries around the world are currently battling a devastating wave of the COVID-19 pandemic, fueled by the highly contagious Delta variant. But even as the death toll continues to climb, publicly available mortality statistics continue to show a yawning gap between official COVID-19 figures and the larger, more extensive real numbers.
A World Health Organization (WHO) report released in May suggests a severe undercount worldwide. The total number of global deaths attributed to the COVID-19 pandemic in 2020 exceeds three million, according to WHO estimates. This is far from the 1,813,188 figure reported by December 31, 2020 — 1.2 million more deaths than divulged by country officials largely due to differing testing capacity and reporting policy.
Dr. Ariel Karlinsky, an economist and statistician at The Hebrew University of Jerusalem’s Economics department and the Kohelet Economic Forum teamed up with Dr. Dmitry Kobak from Germany’s Tübingen University to use this metric to collect mortality data from 103 countries and territories around the world during the pandemic. The numbers have been added to a comprehensive list to create the World Mortality Dataset, a collection of data that the pair says uncovers the true rate of COVID-19 deaths.
The findings were published last month in the peer-reviewed scientific journal eLife and are believed to be the largest existing collection of overall mortality data, according to the researchers.
“I was hearing a lot of ‘whispers’ that COVID is ‘just like the flu’ since there were no excess deaths in Israel during that time,” Karlinsky tells NoCamels. “There was almost no data on all-cause and excess mortality, so I started gathering it from countries one by one, eventually obtaining the most complete and thorough dataset of its kind.”
Karlinsky began working with Kobak after he found the researcher on Twitter discussing excess mortality. Kobak, a Russian-born research scientist in Germany had already written a detailed piece showcasing excess deaths in his native country.
“Both of us also had data on Iran which showed large excess mortality there and I knew this information was hard to obtain and his method to estimate excess mortality was better than mine. So I suggested to him we collaborate and thankfully he agreed.”
The team compared the number of overall “known deaths” during the COVID-19 pandemic with the number of overall deaths from previous years. In this way, they were able to determine the likely number of excess deaths caused by the coronavirus pandemic.
They scoured official websites of National Statistics Offices, Health Ministries, Population Registries, and more, for their data. The info was hidden “in plain sight” for a lot of the countries, in obscure tables, automated reports, and through public information requests filed by US or local journalists, Karlinsky tells NoCamels. He also says they found a lot of information independently.
“Our results present a comprehensive picture of the impact of COVID-19,” Kobak said in the university statement, “We hope these findings — and their methodology — will lead to a better understanding of the pandemic and highlight the importance of open-source and fast mortality reporting.”
Undercounting COVID-19 cases
Karlinsky tells NoCamels that undercounting can occur due to a lack of capacity — “If you don’t have enough tests, you can’t certify a COVID death” — or due to malice.
He is currently working on a new paper that tries to understand these mechanisms better. “Generally, I think Latin American countries show a lack of capacity and there is an earnest attempt to monitor and report as accurately as possible.
Several Latin American countries such as Bolivia, Ecuador, Mexico, and Peru underreported their COVID-19 deaths, even though the number of excess deaths sustained during the pandemic period was over 50 percent higher than the number of expected deaths.
Peru stood out, according to the figures. They originally underreported their COVID-19 deaths—claiming only 69,000 deaths when in reality that figure was closer to 185,000. After an outcry by public health officials, Peru’s health ministry made amends.
“Peru’s an interesting case,” Karlinsky admits. “Their excess deaths were about 2.5 times higher than reported COVID deaths but in June they convened a panel of experts to audit their death certificates and they updated the number of COVID-19 deaths they reported such that it is now almost exactly the same as excess deaths”
Sign up for our free weekly newsletterSubscribe
Earlier this year, experts in Russia believed that the country has recorded some 550,000 deaths, a number that is four times higher than the 159,000 deaths published by the government’s coronavirus task force. In Belarus, the number is 14.5 times higher — 5,700 deaths, instead of 392, and in Uzbekistan 29 times higher — 21,500 deaths, instead of the 740 reported. Tajikistan “wins the underreporting prize” with a COVID-19 death rate that is whopping 100-times higher than reported — 9,000 deaths, instead of 90, the university statement said.
According to the collected data, Nicaragua’s true number of pandemic deaths is 50 times higher than reported— 7,000 coronavirus deaths instead of the 137 reported.
“In Nicaragua and some former Soviet countries, I think it’s mostly malice,” Karlinsky says. “It’s an attempt at misinformation – to project to the world and the populace that things are okay. I think they know exactly what the true toll is.”
Israel’s excess deaths during the coronavirus pandemic were actually smaller than their reported figures, which were 5,000 instead of 6,400, the study showed. This is likely due to a decrease in the overall number of deaths from non-COVID 19 respiratory infections during the winter months.
Israel is similar to other developed countries that have been counting and tracking COVID reliably, Karlinsky indicates to NoCamels. “We found no evidence of inflating or undercounting COVID deaths here.”
According to the findings, Israel fared better than its neighboring countries at 58 excess deaths per 100,000 persons. Egypt’s excess deaths were 13 times higher than reported — 196,000 instead of 15,000, Iran’s were 2.15 times higher –115,000 COVID-19 deaths instead of 54,000, and Lebanon’s figures were 1.23 times higher than reported — 9,000 deaths instead of 7,300.
“There have been leaks from Iran early in the pandemic that the regime has two separate data reports: reported COVID deaths and actual COVID deaths which were about two to three times higher and very similar to the ratio we find between excess mortality and COVID-19 deaths,” Karlinsky says.
Australia and New Zealand’s death rate during the pandemic was actually lower than in previous periods. This is likely due to their virus containment efforts, which included border closures, social distancing, and mask-wearing which decreased their overall number of deaths during the pandemic period.
“Another interesting pattern we found is that countries that had almost no COVID not only didn’t have excess deaths but had even fewer deaths than expected due to social distancing, lockdowns, reducing the number of influenza [flu] deaths, and traffic accident deaths. This is true for New Zealand and Australia and also for places where there were less restrictive lockdowns, but where they managed to keep COVID-19 at bay by border closures and the like. Places like Japan, Taiwan, and Uruguay, up to the end of 2020,” Karlinsky explains.
Among European nations, the team found that many countries faithfully reported their pandemic deaths. Per 100,000 people, the United Kingdom suffered 159 deaths, France 110, Switzerland 100. The Czech Republic suffered 320 pandemic deaths and Poland 310. Denmark and Norway were unique in that they experienced no excess mortality during the pandemic. The United States had 194 excess deaths per 100,000 persons.
Karlinsky says he hoped the data “will be a valuable asset for public health officials looking to assess the risks and benefits of a given pandemic-containment measure.”
“We are constantly adding information from more and more countries, as some of them are only now reporting deaths for 2020,” he says. Karlinsky is also working on a research project to estimate excess deaths in India, where he says there’s almost no national data, just information at the state level.
Karlinsky says he hopes the study “will lead to an international standard for reporting weekly or monthly mortality so that we could track and monitor better from now on. There’s no reason why countries should report unemployment rates, GDP, and other data, but not these vital statistics.”