Research at odds over impact of flu vaccine

How much difference does the flu shot really make?

Various studies offer wide-ranging conclusions. One, for instance, found that seasonal influenza vaccine halves elderly deaths from any cause in the winter months. But another found it has no impact at all on this age group's health. These and other conflicting findings are pushing public health officials and scientists to look for better ways to quantify the preventive's impact.

"Flu vaccine works in elderly people, but what we don't have is a complete picture of what the true effect might be and how it might vary from season to season," said David Shay, MD, MPH, a medical officer in the Centers for Disease Control and Prevention's influenza division.

Consider the following results.

A meta-analysis in the March 15, 2002, Vaccine reported that inactivated influenza vaccine in those older than 65 who did not live in institutions cut influenza-like illness by 35% and all-cause mortality by 50%. Another paper, in the Oct. 4, 2007, New England Journal of Medicine, pooled data on 18 cohorts of 713,872 elderly people and found that vaccination was linked to a 27% cut in hospitalization from pneumonia or flu and a 48% reduction in the risk of death.

Other research offers an alternate perspective.

A study in the Feb. 14, 2005, Archives of Internal Medicine suggested that the decline in influenza-related mortality in the 1970s among those ages 65-74 came from immunity acquired during the 1968 pandemic. Also, vaccine coverage increased from between 15% to 20% in 1980 to 65% in 2001, but this increase could not be connected to declines in mortality. Another study in the Aug. 2, 2008, Lancet examined 1,173 elderly people with community-acquired pneumonia and 2,346 controls. The study found no link between flu vaccination and a reduced risk of this illness.

All agree: Vaccine has value

The authors of studies questioning the effectiveness of the flu vaccine say, however, that the take-home message from their work is not that this immunization doesn't work. Rather, the data saying it does are not as strong as they would like.

"The true benefit is somewhat unknown," said Lisa Jackson, MD, MPH, one of the authors of the Lancet paper and a senior researcher at Group Health Center for Health Studies in Seattle. "For the individual, it's reasonable to continue getting influenza vaccine, but we need a more realistic message about the benefits they might expect. [A] 50% reduction in all-cause mortality is just not plausible in any sense. Influenza only causes 5% of winter deaths at most."

But the kind of placebo-controlled randomized trials that might quantify the benefit once and for all are unlikely. Numerous such investigations already have been conducted, but they were carried out in young and middle-aged adults rather than those older than 65 -- the demographic that makes up the biggest target of vaccination efforts and suffers the greatest ill effects from this respiratory infection. A trial focused on this age group would be considered unethical. And the observational studies researchers have used to help answer the question may be confounded by other factors. For instance, according to those who question the more positive papers, results may be inflated by the possibility that the elderly who get the vaccine may be fitter than those who do not.

"The seniors who are healthy and are able to go get a flu shot are less likely to get pneumonia or die. The benefits [of the influenza vaccine] are probably real but rather small," said Michael L. Jackson, MPH, PhD, lead author of the Lancet paper. He wrote it as a postdoctoral fellow at Group Health.

In addition, the impact may not be big enough to detect in smaller studies. And because the evaluated endpoints vary widely, it can be tough to compare one to another.

"We have to be very careful about making sure that we understand what the study populations are and what the outcomes are," said Kristin Nichol, MD, MPH, the lead author of numerous papers finding benefits from influenza vaccine, including the 2007 NEJM article. She also is associate chief of staff for research at the Minneapolis Veterans Affairs Medical Center. "It's really important to look at the entire body of evidence."

But these investigations are complicated by the nature of influenza vaccine, which aims at a constantly moving target. The viral strains it includes change every year. Circulating viruses shift constantly. And every season is different in regard to severity and spread.

A study in the Jan. 15 Journal of Infectious Diseases found the vaccine's effect in preventing an infection that requires medical attention ranged from 5% during the 2004-05 season to 37% in 2006-07. These rates stemmed from the differences between what was in the vaccine and what was moving around in the community.

Public health officials are trying to get a better handle on the vaccine's actual impact, and advocates say the effort is still a good idea. The CDC is running several projects seeking an answer. And according to the CDC Web site, when the vaccine is well-matched to circulating strains, immunization reduces the risk of getting influenza by 70% to 90% among healthy adults. It may be less effective in elderly people and very young children, although it still provides protection against severe complications.

"Influenza is a pretty good vaccine. It's far from perfect, but there is not any doubt that there's some benefit," said William Schaffner, MD, president-elect of the National Foundation for Infectious Diseases and chair of the Dept. of Preventive Medicine at Vanderbilt University School of Medicine in Nashville, Tenn.

More than 143 million doses were delivered this season. The National Influenza Vaccine Summit, which is sponsored by the American Medical Association and the CDC, will meet in Dallas May 19-20 to discuss this season and lay plans for the next.