COVID-19 killed roughly 7.5 million people in its first two years alone, based on WHO excess mortality estimates, and continued to cause hundreds of thousands of deaths annually even after vaccines became widely available. The toll has dropped sharply since the pandemic’s peak, but COVID still kills far more people each year than seasonal flu.
Global Deaths by Year
Counting COVID deaths precisely is harder than it sounds. Governments officially reported fewer than 6 million deaths worldwide through the end of 2021, but WHO estimated that actual excess deaths (the number of people who died above what would normally be expected) reached 14.9 million during 2020 and 2021 combined. Independent analyses from the Institute for Health Metrics and Evaluation put the figure even higher, at 18.2 million for the same period. The gap exists because many countries lacked testing capacity, and deaths in rural or low-income areas often went uncounted.
Official WHO-reported figures, which represent a floor rather than a ceiling, break down roughly as follows: about 1.9 million reported deaths in 2020, over 3.5 million in 2021 (the deadliest year), around 1.2 million in 2022, and smaller but still significant tolls in 2023 and 2024 as immunity from vaccination and prior infection accumulated. By early 2025, the cumulative official global count exceeded 7 million, though the true number, accounting for undercounting, is likely more than double that.
How COVID Compares to the Flu
One way to put COVID’s toll in perspective: through February 2023, three years of COVID deaths in the United States equaled what 17 years of flu and pneumonia deaths would produce. In every U.S. state except Hawaii, it took at least nine years’ worth of flu mortality to match what COVID did in three. For four states, even 21 years of flu deaths weren’t enough to reach COVID’s three-year total.
Seasonal influenza typically kills between 290,000 and 650,000 people globally per year. Even in its mildest recent years, COVID has exceeded the upper end of that range. The comparison matters because “it’s just like the flu” became a common framing early in the pandemic. The data show that framing was never accurate, and while the gap has narrowed, COVID remains the more lethal respiratory virus.
Who Is Most at Risk
Age is the single strongest predictor of dying from COVID. In the U.S. general population, 60% of all adult COVID deaths have occurred among people 75 and older. Data from Texas during the pre-vaccine period showed that the fatality rate for people 85 and older was 24%, compared to 0.01% for children aged 1 to 14. That age gradient has remained consistent throughout the pandemic, even as overall death rates dropped.
Men have consistently faced higher hospitalization and fatality rates than women, at roughly 6.2% versus 5.2% for hospitalization and 1.9% versus 1.5% for fatality during the pre-vaccine period. Racial disparities also appeared early: Black Americans had the highest hospitalization rates (14%, age-adjusted), while Asian and Pacific Islander Americans had the highest case fatality rates (4.4%). These gaps reflect differences in occupational exposure, access to healthcare, and rates of underlying conditions rather than any biological predisposition.
How Vaccines Changed the Numbers
Vaccination dramatically reduced the death rate. CDC data from late 2022 showed that unvaccinated people were 14 times more likely to die from COVID than those who had received an updated booster dose. Among adults aged 65 to 79, that gap was even wider: unvaccinated individuals died at nearly 24 times the rate of boosted ones.
The protection isn’t static, though. During the Delta wave, unvaccinated people died at roughly 50 times the rate of recently boosted individuals. By the time Omicron subvariants BA.4 and BA.5 were circulating, that ratio had dropped to about 14 to 1, reflecting both viral evolution and waning immunity over time. This is why updated boosters remain relevant: they restore protection that fades in the months after a previous dose.
What Death Certificates Actually Show
Some skepticism about the death count centers on whether people died “from” COVID or “with” COVID. U.S. death certificate data from 2020 provides a clear picture. Of the 378,048 death certificates listing COVID that year, 87% placed it in the chain of events directly causing death, not as a side note. Overall, 97.3% of certificates with additional diagnoses listed conditions that logically follow a severe COVID infection: pneumonia, respiratory failure, cardiac arrest, or sepsis. About 5.5% of certificates listed COVID as the only diagnosis, with no other conditions mentioned at all.
Nearly 75% of death certificates showed a chain-of-event condition, meaning COVID triggered a cascade (like respiratory failure) that directly caused death. Another 18% listed a contributing condition only, such as diabetes or hypertension, which made a fatal outcome more likely but didn’t replace COVID as the primary cause. In short, the vast majority of counted COVID deaths reflect cases where the virus was genuinely responsible.
Why Exact Counts Are Difficult
The WHO’s excess mortality estimate of 14.9 million deaths in 2020 and 2021 was 2.74 times higher than the 5.42 million officially reported for the same period. That ratio tells you how much undercounting occurred, particularly in countries without robust vital registration systems. India, for example, likely experienced millions of deaths that never appeared in official tallies.
Even in wealthy nations with good record-keeping, counts can lag by months as death certificates are processed and coded. Testing availability also affects the numbers: if someone dies at home without ever being tested, their death may not be attributed to COVID even if the virus caused it. Excess mortality estimates sidestep this problem by simply comparing how many people died in a given period to how many would have been expected to die based on historical trends. The gap between those two numbers captures COVID deaths, including ones that were never officially diagnosed.