Preliminary research suggests the seasonal flu shot may put people at greater risk for getting swine flu, CBC News has learned.

"This is some evidence that has been floated; it hasn't been validated yet, it's very preliminary," cautioned Dr. Don Low, microbiologist in chief at Mount Sinai Hospital in Toronto.

"This is obviously important data to help guide policy decision. How can we best protect people against influenza?"

It's important to validate the information to make sure it's not just a fluke, and that the observation is confirmed elsewhere such as in the Southern Hemisphere, which just completed its seasonal flu season, Low added.

The four Canadian studies involved about 2,000 Canadians in Ontario, Quebec, British Columbia and Alberta, sources told CBC News. Researchers found people who had received the seasonal flu vaccine in the past were more likely to get sick with the H1N1 virus.

Researchers know that theoretically when people are exposed to bacteria or a virus, it can stimulate the immune system to create antibodies that facilitate the entry of another strain of the virus. Dengue fever is one example, Low said.

International influenza vaccine experts are apparently not convinced that Canadian researchers have found a true link between getting a seasonal flu shot and catching swine flu.

The consensus that emerged from a World Health Organization teleconference Friday on the controversial data seemed to be that the Canadian findings are likely due to some confounding factor or factors in the data themselves and may not reflect a real increased risk, according to a WHO official who helped pull together the meeting.

"From a WHO point of view, the fact that the findings are not replicated in other countries I think is reassuring for us that this is an outlier, if you like, the unexpected findings that are coming out of Canada," said David Wood, co-ordinator of the quality, safety and standards team of WHO's department of immunization, vaccines and biologicals.

"Most people are still looking at this as some sort of undetected confounding in the data, that for some reason is giving the results that are there."