July 31, 2008 -- Flu vaccines offer seniors far less pneumonia protection than was thought, a carefully designed study shows.
It's a big surprise. Earlier studies found huge decreases in illness and death among seniors who got their annual flu shots. But it's now becoming clear that flaws in these studies inflated the benefits of flu vaccination in the elderly.
A new, carefully controlled study corrects those errors. It finds that flu vaccination does not significantly protect seniors against all-cause pneumonia during flu season, says Michael L. Jackson, PhD, who worked on the study at Seattle's Group Health Center for Health Studies. He's now a medical epidemiologist with CDC.
"We don't know how the flu vaccine works in older seniors or in seniors with chronic health problems. We thought we did, but now we see we don't," Jackson tells WebMD. "We cannot say there is no benefit. But we can no longer say there is a 20% to 30% lower risk of pneumonia. It could more like 5% or 10%."
Flu Shot Still Advised
So why do Jackson and other experts continue to advise elderly people to get their flu shots every year?
It's because the Jackson study measured protection against pneumonia from all causes taken together. Flu vaccine only protects against flu-related pneumonia. And studies suggest that except during exceptionally severe flu seasons, flu accounts for no more than 15% of pneumonia cases.
This means the flu vaccine probably does what it's supposed to do -- even though elderly people who get their flu shots still risk pneumonia from non-flu causes, suggest Edward A. Belongia, MD, of Wisconsin's Marshfield Clinic Research Foundation and David Shay, MD, MPH, of the CDC's influenza branch.
"To take the positive view, the vaccine's effectiveness in preventing flu could be a lot higher," Belongia tells WebMD. Jackson and colleagues could not tell which pneumonia cases were due to flu. So the overall effectiveness against pneumonia may be quite low, but it may work quite well against flu-related pneumonia.
In an editorial accompanying Jackson and colleagues' report in the Aug. 2 issue of The Lancet, Belongia and Shay calculate that if the flu vaccine is 8% effective at preventing all-cause pneumonia, and flu is behind just 10% of pneumonia cases in a given flu season, the vaccine would actually be about 70% effective in preventing flu-related pneumonia.
But if flu were behind 30% of pneumonia cases during a flu season in which the vaccine were 8% effective against all-cause pneumonia, the flu vaccine would be about 33% effective. That's not great, but much better than no protection at all.
And flu vaccine also prevents the flu -- a miserable illness even if it does not result in pneumonia.
"Nobody suggests we stop vaccinating elderly people against the flu," Shay tells WebMD. "This study doesn't contravene any recommendations for annual vaccination for people in this age group."
Confirming Cases of Flu
The Jackson study did solve one big problem for flu researchers. It looked at the medical records for 1,173 65- to 94-year olds who had the flu during the 2000, 2001, and 2002 flu seasons and compared them to 2,346 age- and health-matched people who did not get the flu.
Because people usually get their flu shots in the fall, well ahead of the usual start of flu season, Jackson's team was able to look at whether people got pneumonia in the period after they got their flu shots but before they could have gotten the flu.
This turned up a revealing phony statistic: The flu vaccine appeared to be 40% effective in preventing all-cause pneumonia, even though it could not have had any effect during that time.
But when the researchers considered several confounding factors, they corrected this false finding. Those factors -- such as asthma, smoking, home use of oxygen, use of inhaled steroids, previous pneumonia episodes, and antibiotic prescriptions -- turned out to be responsible for the false reading. That's because people able to get their flu vaccinations tend to be healthier than those who don't.
After correction, the flu vaccine no longer prevented pneumonia before flu season. And during flu season, it was a statistically insignificant 8% effective in preventing all-cause pneumonia.
That forever changes how flu effectiveness studies should be done, says Shay. It's no longer going to be helpful to study the effects of flu vaccine without accurately identifying who truly got the flu and who did not. Measures such as flu-like illness, hospitalization, or all-cause pneumonia are just too misleading.
"If we are going to move forward, flu-vaccine efficacy studies will need to have lab-confirmed flu as an outcome," Shay says.