Facebook Twitter Tumblr Close Skip to main content
A Project of The Annenberg Public Policy Center

Trump’s Suspect Spanish Flu Claims


President Donald Trump recently said “close to 100 million people died” from the 1918 Spanish flu, and that “if you got it you had a 50/50 chance, or very close, of dying.” But 100 million is a high-end estimate of global deaths from that influenza pandemic, and we found no evidence the case fatality rate for those who had it was 50%.

Trump made those claims during a March 24 virtual town hall from the White House Rose Garden that aired on Fox News. He argued that the current coronavirus pandemic cannot be compared to the 1918 flu pandemic because of how deadly the latter was.

Trump, March 24: You can’t compare this to 1918 where close to 100 million people died. That was a flu, which — a little different. But that was a flu where if you got it you had a 50/50 chance, or very close, of dying.

I think we’re substantially under 1% because the people that get better are not reporting. So we only know people that go to doctors and go to hospitals, and we’re taking that. And we’re still a little bit above 1%. When you add all of the people — the millions of people that have it, that get better, we’re substantially less than 1%.

Trump has criticized the World Health Organization’s March 3 estimate of a 3.4% global case fatality rate for individuals with COVID-19, the pneumonia-like disease caused by the new coronavirus. The president called it “a false number” because it was based on known cases and fatalities and does not account for unreported cases. 

It’s not a false number, as we’ve explained before, but the fatality rate for COVID-19 may end up being much lower — perhaps less than 1% — when more data about the total number of people who had it is available.

Trump’s claims about deaths from the so-called Spanish flu, however, are more suspect. To start, he relied on one of the highest estimates of fatalities; others are lower.

In a 1991 report published in the Bulletin of the History of Medicine, K. David Patterson and Gerald Pyle estimated that between 24.7 million and 39.3 million people died worldwide during that pandemic — but the authors wrote that “we believe that approximately 30 million is the best estimate for the terrible demographic toll of the influenza pandemic of 1918.” 

The authors also noted that a much earlier, and much cited, estimate from 1927 put the death toll lower, at 21.6 million. 

Then in a 2002 report, also published in the Bulletin of the History of Medicine, Niall Johnson and Juergen Mueller, referring to those earlier estimates, produced their own — 50 million deaths, globally — which they said may have been an underestimate by as much as 100%. “Consequently, the real pandemic mortality may fall in the range of 50 to 100 million,” they wrote.

It appears Trump was referring to that upper estimate.

But more recently, in 2018, Peter Spreeuwenberg, Madelon Kroneman and John Paget did their own analysis — which was published in the American Journal of Epidemiology — and concluded that “100 million deaths is highly unlikely.” Instead, “the true estimate,” the authors wrote, is “probably closer to our final estimate of 17.4 million deaths (1918 and 1919 combined).”

And even Johnson and Mueller, who estimated as many as 100 million deaths, said “it would seem unlikely that a truly accurate figure can ever be calculated.”

What we also don’t know for certain is how many people around the world contracted the Spanish flu. 

But if 500 million became infected, as the Centers for Disease Control and Prevention says, then those who contracted that influenza strain had a much lower chance of dying than Trump said, based on the previously discussed estimates of deaths.

For instance, if 17.4 million of 500 million infected people died, then the global case fatality rate was about 3.5%. If there were 30 million deaths, then the rate would be 6%. And even if the number of deaths was 50 million or 100 million, then the rate would be between 10% and 20%.

None of those equals a “50/50” chance of death, or a ratio “very close” to it.