A corresponding assessment of how effective the vaccine is in Canada this winter is still in the works.
But there is good reason to believe the Canadian estimates will be at least as bad when they are published, probably later this month, experts say.
The study calculated how much protection the vaccine offered against getting sick enough to require medical care. (Most people who catch the flu tough it out without seeing a doctor, making them impossible to study.)
It found that so far this year the vaccine is, on average, lowering that risk by 23 per cent, across all age groups and all the types of circulating flu viruses.
When the scientists look specifically at H3N2 viruses — far and away the most dominant virus type so far this winter — the protection was 22 per cent.
That's not spectacular, but when the data were broken down by age groups, the picture was worse.
The study suggested the vaccine is cutting the risk for children aged six months to 17 years of age by 26 per cent. But for adults 18 to 49, the protection was only 12 per cent and for adults 50 and older, it was 14 per cent.
"It is a disappointing result," said Dr. Edward Belongia, an influenza expert at Marshfield Clinic Research Foundation in Marshfield, Wisc., and one of the authors of the study, which was a collaboration between CDC scientists and researchers in Michigan, Pennsylvania, Texas, Washington and Wisconsin.
"We do need a better vaccine. We know that. But it's not zero. And this disease can be so severe that people should take whatever protection they can get."
The study was published Thursday in the Morbidity and Mortality Weekly Report, a rapid-release journal published by the CDC.
Michael Osterholm, director of the Center for Infectious Diseases Research and Policy at the University of Minnesota, said the poor performance of this year's vaccine is further proof a better influenza vaccine is needed.
"This should be another clarion call for why we need new and better flu vaccines. Twenty per cent clearly is better than zero, but it's a long way from what we need and must eventually have," said Osterholm, who has long championed the need for completely revamped flu vaccines.
The low level of protection is being attributed to the fact that most of the H3N2 viruses in circulation are not a good match for the one in this year's vaccine.
Because flu vaccine takes months to make, experts select the strains for the following winter's shot in February each year. If a virus mutates — drifts, in the language of influenza — after production starts, sub-optimal protection can result.
The CDC report said that about two-thirds of the H3N2 viruses seen this year in the U.S. were not well matched to the vaccine virus.
"It's a pretty unusual year when we get this much drift in the predominant virus and therefore this low a vaccine effectiveness," said Dr. Joseph Bresee, head of the epidemiology and prevention branch of CDC's Influenza Division.
Bresee said vaccine effectiveness of about 50 to 60 per cent is more the ballpark for flu.
"And so year in and year out, obviously we think it's very wise for folks to get the vaccine. This year obviously we're disappointed with the vaccine effectiveness but most years we get vaccine effectiveness that is much higher."
Canada is seeing a higher percentage of H3N2 drift viruses this season than the U.S. is.
Dr. Danuta Skowronski, a leading Canadian flu researcher, said where the figure in the U.S. is about 68 per cent, in Canada nearly all — 98 per cent — of the H3N2 viruses are different from the vaccine virus.
And as it is in the U.S., H3N2 is responsible for virtually all the flu activity currently being seen in Canada. (That could change as the season progresses. Influenza B viruses often surge late in the season.)
Skowronski, who is with the British Columbia Centre for Disease Control, is working with colleagues from across the country to generate vaccine effectiveness estimates for Canada for this season.
"I would qualify these findings as showing little or no protection," Skowronski said of the U.S. data.
"And I don't think we should expect the Canadian estimates to be higher, given that we have a greater proportion of drift variants (viruses)."
She said this year's poor showing is part of an ongoing pattern of problems with H3N2 viruses.
For starters, this family of influenza A viruses don't grow well in eggs, which is the way most influenza vaccine is produced. As they adapt to eggs, the viruses acquire mutations that actually undermine the vaccine's effectiveness. That can produce a double whammy when the target shifts, as it has with this year's drifted H3N2 viruses.
All this matters because H3N2 is the nastiest of the human flu viruses currently in circulation. These viruses cause severe disease in the elderly; years where H3N2 is dominant are years when there are higher rates of influenza hospitalization and deaths.
Flu scientists who work on vaccines often talk about universal flu vaccines, a type of flu shot that would protect against all or at least many different types of flu. The goal is to develop a vaccine that one might take once in a lifetime — or perhaps every 10 or 20 years, if protection needed to be boosted from time to time.
Researchers working to find a universal vaccine claim from time to time that the science is progressing to the point where one might be possible in five to 10 years. In fact, researchers at Mount Sinai Hospital in New York reported progress on a universal flu vaccine earlier this week.
But others are skeptical, having heard these claims many times before. And it's unclear what motivation exists for manufacturers to shift from a vaccine they sell year after year to one that might last a lifetime, especially given that developing, testing and licensing a universal flu vaccine would cost hundreds of millions of dollars.
Skowronski said that as the world waits for a universal vaccine, the flu field should focus on improving the H3N2 component of the existing flu shot.
"There's a big gap between what we would like to see and what we ought to and what we are currently seeing in vaccine effectiveness," she said.
"And if this were just a one-off, then I think we'd carry on. But this is not a one-off with H3 viruses."