Deaths from H1N1 influenza in 2009 may have been 10 times higher than previously estimated, killing 123,000 to 203,000 people from respiratory illness worldwide, according to a new analysis in the journal PLOS Medicine.
The World Health Organization had listed the global death toll from the H1N1 pandemic, also known as swine flu, at 18,449, based only on laboratory-confirmed cases.
That relatively low death tally had led some to wonder if the dangers of H1N1 had been overblown, according to the analysis, conducted by 60 researchers in 26 countries and funded by the WHO.
Most people infected with the flu never got a lab test to confirm their diagnosis, according to the study, whose authors used death records from countries around the world to estimate the true number of deaths.
Some flu deaths go uncounted, because the immediate cause of death of many people is actually pneumonia or another chronic respiratory ailment, according to the study led by researchers at George Washington University School of Public Health in Washington.
Health officials around the world were on high alert after H1N1 appeared, because it was a new strain of flu virus for which, at least initially, there was no vaccine. Earlier pandemics killed millions of people worldwide, and officials were concerned that people would have no natural immunity to a new virus.
Some researchers said they expected the final estimate of flu deaths to be higher than early tallies.
"We knew all along that lab-confirmed deaths were just the tip of the iceberg," said Michael Osterholm, director of the Center for Infectious Disease Research and Policy at the University of Minnesota.
Researchers write that their analysis may underestimate the true number of flu deaths. That's because they included only deaths from respiratory illness. Researchers also covered only the first nine months of 2009 and may have missed deaths from later that year.
Mortality was highest in younger people and in Central America and South America, where death tolls were up to 20 times higher than in Europe.
That variation shows how difficult it can be to prepare for a pandemic and to respond once it's underway, said Arnold Monto, a leading flu researcher and professor at the University of Michigan School of Public Health. Policymakers can't easily predict how hard their communities will be hit. The first cases of H1N1 were found in Mexico, which had one of the highest death rates from the pandemic, Osterholm said.
The analysis confirms how H1N1 differed from seasonal flu. In a typical year, 90% of flu deaths occur in those over age 65. With H1N1, 62% to 85% of deaths occurred in those under age 65.
In the past, public health officials have judged the severity of a flu outbreak on the total number of deaths. Osterholm says it's also important to know which populations suffered the most.
"It's not just about the total number of deaths, but who is dying," Osterholm says. "In the past, we've used deaths as an indicator of severity. But a death in an 82-year-old is very different than in an healthy 21-year-old female who happened to be pregnant."