Do Flu Shots Save Lives?
One Study Shows a Large Impact, but Another Study Shows None
Feb. 14, 2005 - It is widely believed that getting a flu shot reduces deaths this time of year among the elderly by half or even more, but new research finds the lifesaving benefits of vaccination to be greatly overestimated.
Researchers from the National Institute of Allergy and Infectious Diseases (NIAID) found no correlation between an increase in flu vaccine coverage over the past two decades and a decrease in influenza-related deaths among the elderly. Their conclusions contradict most other reported studies, including one published in the same Feb. 14 issue of the journal Archives of Internal Medicine.
In that study, researchers from the Netherlands reported that influenza vaccination was associated with a 50% reduction in deaths among people aged 65 and over. Vaccination was also credited with reducing deaths by 78% among younger, high-risk adults, such as those with chronic heart or lung diseases.
Coverage Has Increased
So which study is right? Noted flu vaccine expert William Schaffner, MD, says the truth probably lies somewhere in the middle.
The NIAID study was the first to examine flu deaths for the entire U.S. population rather than focusing on a selected group, as has been done in previous studies. Researcher Lone Simonsen, PhD, and colleagues used statistical models that tracked deaths due to pneumonia, influenza, and all other causes among the elderly from 1968 until 2001.
The percentage of elderly Americans who got annual flu shots rose steadily from around 15% before 1980 to 65% in 2001. Simonsen tells WebMD that the dramatic increase in coverage should have led to a dramatic drop in flu deaths.
"This is not what we found," she says. "Certainly if this intervention really does reduce winter deaths in the elderly by 50% we would expect to see it. So the mortality benefits are probably very much overestimated."
Simonsen says the 1997-1998 flu season perfectly illustrates the point. That year, the vaccine that was manufactured ended up being totally mismatched with the circulating flu strain, she says. So even though 63% of the eligible elderly got their flu shots, coverage was effectively zero.
Despite this, there were approximately 5,000 fewer excess deaths in this age group than there were the next flu season, when the same percentage of people were vaccinated and the vaccine did match the flu strain.
But Schaffner says the national data Simonsen and colleagues were forced to rely on were not comprehensive enough to give a true picture of the impact of vaccination on mortality.
"National record keeping is not precise enough in tracking illness, vaccination, and influenza diagnosis in the individual patient," he says. "We have not been able to show a mortality impact with vaccination, and that has been a source of frustration."