1. Skip to content
  2. Skip to main menu
  3. Skip to more DW sites

COVID deaths may be three times higher than we think

Alexandria Williams
March 18, 2022

A new study estimates deaths caused by COVID are much higher than previously thought. The results help experts paint a clearer picture of the pandemic's wider global impact.

People carrying a coffin
A new study tracks estimated deaths due to COVID infection as well as the impact of stressed hospital systemsImage: Agustin Marcarian/REUTERS

Some  18 million people across the globe may have died in response to the COVID-19 pandemic — three times the number previously reported by the WHO, according to a new Lancet study. 

Researchers at the University of Washington’s Institute for Health Metrics and Evaluation (IHME) set out to measure something called "excess mortality" — the difference between the number of deaths during the first wave of the COVID pandemic and the number of deaths that would have occurred in its absence.  

Their study says its estimates indicate that "the full impact of the pandemic has been much greater than what is suggested by official statistics." 

Not enough data 

IHME researchers surveyed 191 national and 252 subnational locations to get a picture of excess deaths between January 2020 and December 2021. 

The study revealed that the global excess mortality rate across all age groups was 120.3 deaths per 100,000 people. 

The countries of India (4.7 million), the USA (1.13 million) and Russia (1.07 million) saw the highest numbers of excess deaths. They were followed by Mexico, Brazil, Indonesia and Pakistan. 

Even in wealthy industrialized countries, COVID deaths have been hard to keep track of. For example, when a person in a nursing home died of COVID near the start of the pandemic, sometimes they were recorded as dying of other causes than COVID in official statistics.  

But they're nearly impossible to quantify in countries where the government hasn’t kept track of deaths at all.  

"Around 50% of countries in the world either did not have registration systems set in place or did not log information," said Dr. Haidong Wang, the IHME professor who led the study.


Person in India wearing a mask
India saw some of the highest numbers of COVID-19 infections in the worldImage: DW

Wang and his team used the weighted average of six different predictive statistical models designed to control for oddities like heath waves and lags in data registration to obtain estimates for these countries. This method produced staggering results.  

Certain states in the US, like Mississippi, had an estimated excess death rate two times higher than India, one of the nations hit hardest by the COVID pandemic. 

But the rate of excess deaths caused by COVID accounts for the number of deaths across all age groups and the total size of country's population. India is a young country with a large population relative to Mississippi. This contributes to a comparatively low excess death rate.  

 "So, when we compare that with some national level estimates elsewhere, India's excess mortality rate is not as high," said Wang. 

Understanding COVID's impact 

Studies like the IHME's help us understand "the wider impact of the pandemic," Nikos Kapitsinis told DW. He studies the underlying factors of excess mortality at the University of Copenhagen.

When interpreted alongside COVID response methods, the data show that among all factors impacting excess mortality, a country's capacity to "test, trace and isolate" — test for COVID cases, trace infections and isolate those who come in contact or are infected with COVID — was the most critical, said Kapitsinis. 

South Korea's excess mortality rate was 4.4 per 100,000 people in Wang's study, well below the global average. This may be because South Korea committed to the test, trace and isolate strategy early on.  

After a super spreader event in February 2020, South Korean health authorities employed contact tracers, who used CCTV footage to rigorously track the movements of people who had tested positive. An €7,234 ($8,000) fine was imposed for those who did not isolate.


COVID-19 in South Korea

New Zealand and Australia fared even better, with negative estimated excess mortality rates of −9.3 and −37.6, respectively. This means those countries saw fewer overall deaths over the course of the pandemic than if it had never happened.  

As islands, these two countries were able to isolate and control borders more easily, Kapitsinis said. 

 New Zealand, for example, closed its borders to everyone except travelers from Australia, which also closed its borders  for a similar period. 

Social distancing reduced incidences of severe flu, which may have also contributed to the lower-than-average mortality rates. 

The study's results also help combat the misconception that some countries were more susceptible to COVID than others. 

"What appeared to happen in real time is that high income countries were being affected most by COVID, when the reality was those countries had testing capacity to show the impact of COVID in real time," said Dr. Oliver Watson, a visiting researcher at the Imperial College London who works on modeling COVID transmission.

Accurate estimates 

Watson said it is still important to consider that figures like those published in the IHME study are estimates.

As the report notes, there was little accessible data for countries in regions throughout Africa. For example, the study notes large, seemingly inexplicable differences in estimated mortality rates across sub-Saharan Africa, with the exception of Eswatini, Lesotho, Botswana and Namibia.

Dr. Wang said low rates in certain regions could be explained by young populations, but could not account for the difference between low rates across most of Africa and higher rates in some southern African countries. However, he said the lack of available data should not have a significant impact on results.

Although the research team's methods may result in slight overpredictions, its estimates are "close to being true," said Ariel Karlinsky, a statistician who measures excess mortality estimates for the World Health Organization.

Edited by: Clare Roth