Advertisement

SKIP ADVERTISEMENT

Comments

The Pandemic’s Hidden Toll: Half a Million DeathsSkip to Comments
The comments section is closed. To submit a letter to the editor for publication, write to letters@nytimes.com.

The Pandemic’s Hidden Toll: Half a Million Deaths

U.S.
79,000 weekly deaths
Jan.
Dec.
Deaths in 2020
Expected deaths in 2020
U.K.
24,000 weekly deaths
Jan.
Dec.
Italy
No data
86,000 monthly deaths
Jan.
Dec.
Spain
19,000 weekly deaths
Jan.
Dec.
France
19,000 weekly deaths
Jan.
Dec.
Netherlands
5,000 weekly deaths
Jan.
Dec.
Belgium
4,000 weekly deaths
Jan.
Dec.
Sweden
2,000 weekly deaths
Jan.
Dec.
Switzerland
2,000 weekly deaths
Jan.
Dec.
Poland
64,000 monthly deaths
Jan.
Dec.
Austria
2,000 weekly deaths
Jan.
Dec.
Portugal
2,000 weekly deaths
Jan.
Dec.
Czech Republic
4,000 weekly deaths
Jan.
Dec.
Hungary
3,000 weekly deaths
Jan.
Dec.
Germany
15,000 weekly deaths
Jan.
Dec.
Norway
500 weekly deaths
Jan.
Dec.
Denmark
500 weekly deaths
Jan.
Dec.
Finland
1,000 weekly deaths
Jan.
Dec.
Ireland
500 weekly deaths
Jan.
Dec.
Moscow, Russia
15,000 monthly deaths
Jan.
Dec.
Israel
1,000 weekly deaths
Jan.
Dec.
Mexico
28,000 weekly deaths
Jan.
Dec.
Brazil
33,000 weekly deaths
Jan.
Dec.
Bolivia
17,000 monthly deaths
Jan.
Dec.
Chile
3,000 weekly deaths
Jan.
Dec.
Colombia
8,000 monthly deaths
Jan.
Dec.
Ecuador
20,000 monthly deaths
Jan.
Dec.
Peru
29,000 monthly deaths
Jan.
Dec.
South Africa
5,000 weekly deaths
Jan.
Dec.
Istanbul, Turkey
2,000 weekly deaths
Jan.
Dec.
Jakarta, Indonesia
5,000 monthly burials
Jan.
Dec.
Japan
132,000 monthly deaths
Jan.
Dec.
South Korea
28,000 weekly deaths
Jan.
Dec.
Mumbai, India
13,000 monthly deaths
Jan.
Dec.
Thailand
47,000 monthly deaths
Jan.
Dec.
Note: Bolivia’s Civil Registry recorded almost no deaths in April due to the closure of government offices during a lockdown. Officials said least some of the deaths that occurred in April could have been registered in later months.

At least 496,000 more people died last year during the coronavirus pandemic than the official Covid-19 death counts report, a review of mortality data last year in 35 countries shows — providing a clearer, if still incomplete, picture of the toll of the crisis.

Far more people died in most of these countries than in previous years, The New York Times found. The totals include deaths from Covid-19 as well as those from other causes, likely including people who could not be treated as hospitals became overwhelmed. These numbers undermine the notion that many people who have died from the virus may soon have died anyway.

How excess deaths compare with reported Covid-19 deaths in 2020

Area PCT. above normal Excess
deaths
Reported Covid-19 deaths = Difference
Mexico
March 2 - Nov. 15
46% 230,800 98,259 = 132,600
U.S.
March 1 - Dec. 19
17% 385,100 316,370 = 68,700
Peru
March - December
118% 103,600 37,680 = 65,900
South Africa
March 4 - Nov. 10
14% 57,700 25,657 = 32,000
Poland
March - November
16% 48,700 17,150 = 31,600
Italy
March - November
19% 85,600 55,535 = 30,000
Ecuador
March - Oct. 25
79% 36,800 12,553 = 24,200
Brazil
March 2 - Nov. 21
21% 190,300 169,016 = 21,300
Spain
March 2 - Dec. 27
23% 71,200 50,046 = 21,100
Bolivia
March - December
71% 28,900 9,165 = 19,700
Jakarta
March - December
60% 17,300 3,287 = 14,000
Istanbul
March 2 - Dec. 26
26% 15,500 8,714 = 6,800
Colombia
March 2 - Dec. 27
23% 46,700 42,171 = 4,600
Portugal
March 2 - Dec. 13
12% 10,100 5,559 = 4,600
Netherlands
March 2 - Dec. 20
12% 14,700 10,491 = 4,200
Czech Republic
March 2 - Nov. 29
15% 11,900 8,307 = 3,600
Austria
March 2 - Dec. 13
12% 7,300 4,473 = 2,800
Mumbai, India
March - July
25% 8,900 6,350 = 2,600
Hungary
March 2 - Nov. 29
7% 6,900 4,672 = 2,200
Moscow
March - November
27% 23,600 21,795 = 1,800
Finland
March 2 - Dec. 20
4% 1,500 489 = 1,000
South Korea
February - October
Normal 600 464 = 100
Switzerland
March 2 - Dec. 27
13% 7,200 7,210 = <100
Belgium
March 2 - Nov. 22
21% 15,600 18,545 =
U.K.
March 6 - Dec. 18
17% 79,700 82,620 =
Chile
March 2 - Dec. 27
15% 14,100 16,443 =
France
March 2 - Dec. 6
12% 53,100 55,381 =
Sweden
March 2 - Dec. 20
12% 8,100 8,582 =
Israel
March 2 - Dec. 7
9% 2,800 2,924 =
Germany
March 2 - Dec. 13
3% 19,300 22,406 =
Ireland
March 2 - Sept. 27
2% 300 1,569 =
Japan
March - October
Normal <0 1,749 =
Denmark
March 2 - Dec. 13
Normal <0 941 =
Norway
March 2 - Dec. 20
Normal <0 404 =
Thailand
March - April
Normal <0 54 =
Note: Excess deaths are estimates that include deaths from Covid-19 and other causes. Reported Covid-19 deaths reflect official coronavirus deaths during the period when all-cause mortality data is available, including figures that were later revised. Covid-19 deaths in Istanbul were estimated based on an October 2020 report from the Ministry of Health showing that reported coronavirus deaths accounted for roughly three-fifths of excess deaths in Istanbul.

Mortality data in the middle of a pandemic is not perfect. In most places, the disparities between the official death counts and the total rise in deaths reflect limited testing for the virus rather than intentional undercounting. Officially, nearly 2 million people have died of the coronavirus worldwide as of Jan. 12.

But the total death numbers offer a more complete portrait of the pandemic, researchers say, especially because many countries report only those Covid-19 deaths that occur in hospitals.

“Whatever number is reported on a given day is going to be a gross underestimate,” said Tim Riffe, a demographer at the Max Planck Institute for Demographic Research in Germany.

Excess deaths spiked across Europe

After large surges in excess mortality in the spring, most countries across Europe returned to normal levels in the summer. But a second wave of serious illness returned in the fall and continued through the winter, leading to an uptick in mortality once again.

In Belgium and Switzerland, the peaks in mortality in the second half of the year were even greater than in the spring.

Czech Republic
11,900+ excess deaths from March 2 to Nov. 29
4,000 weekly deaths
2,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Hungary
6,900+ excess deaths from March 2 to Nov. 29
3,000 weekly deaths
1,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Italy
85,600+ excess deaths from March to November
86,000 monthly deaths
43,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Poland
48,700+ excess deaths from March to November
64,000 monthly deaths
32,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Belgium
15,600+ excess deaths from March 2 to Nov. 22
4,000 weekly deaths
2,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Switzerland
7,200+ excess deaths from March 2 to Dec. 27
2,000 weekly deaths
1,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Netherlands
14,700+ excess deaths from March 2 to Dec. 20
5,000 weekly deaths
2,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Austria
7,300+ excess deaths from March 2 to Dec. 13
2,000 weekly deaths
1,000
Jan.
March
Dec.
2020
Expected deaths in 2020
France
53,100+ excess deaths from March 2 to Dec. 6
19,000 weekly deaths
9,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Spain
71,200+ excess deaths from March 2 to Dec. 27
19,000 weekly deaths
9,000
Jan.
March
Dec.
2020
Expected deaths in 2020
U.K.
79,700+ excess deaths from March 6 to Dec. 18
24,000 weekly deaths
12,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Portugal
10,100+ excess deaths from March 2 to Dec. 13
3,000 weekly deaths
1,000
Jan.
March
Dec.
2020
Expected deaths in 2020

These estimates were made for each country by comparing the total number of people who died this year to the number of deaths that would be expected given local mortality trends, adjusted to account for changes over time. The Economist is also tracking these deaths, known as excess deaths, in a similar way.

It is unusual for mortality data to be released so quickly, demographers say, but many countries are working to provide more comprehensive and timely information because of the urgency of the coronavirus outbreak. The data is limited and, if anything, excess deaths are underestimated because not all deaths have been reported.

“At this stage, it’s a partial snapshot,” said Patrick Gerland, a demographer at the United Nations. “It’s one view of the problem that reflects that most acute side of the situation, primarily through the hospital-based system.”

Where excess deaths never returned to normal

The virus was last to hit Latin America, and excess deaths still had not returned to normal levels by the end of the year. Mexico recorded at least 230,800 more deaths than usual compared to the same period in previous years — about two times higher than the reported number of Covid-19 deaths during that time.

U.S.
385,100+ excess deaths from March 1 to Dec. 19
79,000 weekly deaths
39,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Mexico
230,800+ excess deaths from March 2 to Nov. 15
28,000 weekly deaths
14,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Brazil
190,300+ excess deaths from March 2 to Nov. 21
33,000 weekly deaths
16,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Peru
103,600+ excess deaths from March to December
29,000 monthly deaths
14,000
Jan.
March
Dec.
2020
Expected deaths in 2020
South Africa
57,700+ excess deaths from March 4 to Nov. 10
16,000 weekly deaths
8,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Colombia
46,700+ excess deaths from March 2 to Dec. 27
8,000 monthly deaths
4,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Bolivia
28,900+ excess deaths from March to December
17,000 monthly deaths
8,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Moscow, Russia
23,600+ excess deaths from March to November
15,000 monthly deaths
7,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Jakarta, Indonesia
17,300+ excess deaths from March to December
5,000 monthly burials
2,000
Jan.
March
Dec.
2020
Expected deaths in 2020
Notes: Bolivia’s Civil Registry recorded almost no deaths in April due to the closure of government offices during a lockdown. Officials said least some of the deaths that occurred in April could have been registered in later months; Data from weeks 1 and 53 are excluded as they may represent partial weeks.

Not all countries saw excess deaths

In a handful of countries, however, there was no clear sign of increased mortality in 2020. The reasons for this are varied and will become clearer in the months ahead as countries process and certify deaths.

In Norway, Denmark and Finland, demographers say the low mortality is due in part to a less severe flu season last winter — but also because these countries were quick to implement severe restrictions to slow the spread of the virus when their outbreaks were smaller and easier to contain.

Germany
19,300+ excess deaths from March 2 to Dec. 13
13,000
26,000 weekly deaths
Jan.
March
Dec.
2020
Expected deaths in 2020
Finland
1,500+ excess deaths from March 2 to Dec. 20
1,000 weekly deaths
Jan.
March
Dec.
2020
Expected deaths in 2020
South Korea
600+ excess deaths from February to October
15,000
31,000 weekly deaths
Jan.
March
Dec.
2020
Expected deaths in 2020
Norway
No excess deaths
1,000 weekly deaths
Jan.
March
Dec.
2020
Expected deaths in 2020
Denmark
No excess deaths
500 weekly deaths
Jan.
March
Dec.
2020
Expected deaths in 2020
Japan
No excess deaths
70,000
141,000 weekly deaths
Jan.
March
Dec.
2020
Expected deaths in 2020
Notes: Data from weeks 1 and 53 are excluded as they may represent partial weeks.

Age breakdowns in mortality data will also provide a clearer picture of the role of Covid-19 in excess deaths. Using relative age-standardized mortality rates, Britain’s Office for National Statistics found that Spain had the highest rate of all European countries during the peak of the pandemic.

Even taking into account differences in mortality by age, experts say the death toll to date could have been much worse.

“Today’s rise in all-cause mortality takes place under conditions of extraordinary measures, such as social distancing, lockdowns, closed borders and increased medical care, at least some which have positive impacts,” said Vladimir Shkolnikov, a demographer at the Max Planck Institute for Demographic Research. “It is likely that without these measures, the current death toll would be even higher.”

Read more about the methodology and download data for excess mortality from The New York Times on GitHub.

To estimate expected deaths, we fit a linear model to reported deaths in each country from 2015 to January 2020. The model has two components — a linear time trend to account for demographic changes and a smoothing spline to account for seasonal variation. For countries limited to monthly data, the model includes month as a fixed effect rather than using a smoothing spline.

Some countries have less historical data available. For countries with three or fewer years of data, the model uses a simple average of deaths in the observed years. For the United Kingdom, the model accounts for the number of bank holidays in a given week, since deaths are not registered during bank holidays.

Tracking the Coronavirus