Controversial flu vaccine study finally published, but questions raised linger

A set of controversial studies that suggested getting a previous seasonal flu shot doubled a person's risk of catching pandemic H1N1 last spring has finally been published -- though the mystery of whether the finding is real remains.

Provinces and territories scrambled to rethink their flu shot programs last fall when word of the then-unpublished findings spread like wildfire through Canadian research and public health communities. Most decided to delay their seasonal flu vaccine programs until distribution of pandemic vaccine was completed.

But outside of Canada, where the effect wasn't seen, scientists scratched their heads and dismissed the inexplicable and troubling findings as "the Canadian problem."

Publication of the research in a top flight journal -- PLoS Medicine -- may make some skeptics consider whether they should revise their thinking.

But the fact remains research elsewhere so far doesn't corroborate these findings and there isn't an easy way to explain why getting vaccinated against seasonal flu would make a person more susceptible to catching the new flu virus.

"I do think that they did the best they could with the data they had," said Dr. Mark Loeb, an infectious diseases expert at McMaster University in Hamilton who was not involved in the study and who has been a skeptic about the findings.

"The question is: OK, is it real? And the answer is: It's not clear."

The studies are presented as a single paper by a team of over 40 researchers including many of Canada's top influenza experts. The data are from four studies that draw cases from British Columbia, Alberta, Ontario and Quebec.

The message is consistent, across studies and provincial borders. It makes it appear that Canadians who got flu shots for the 2008-09 winter were somewhere between 1.4 and 2.5 times more likely to catch the pandemic virus last spring than people who had not been vaccinated against seasonal flu.

The lead author and person who first observed that unexpected finding admitted the results were "heretical."

Dr. Danuta Skowronski of the British Columbia Centre for Disease Control said for her, the story of this research was one of "finding results that you don't like and that others like even less."

"But that doesn't mean that we as scientists can turn a blind eye to results that we don't like," Skowronski, who has taken considerable heat professionally over the past year, said in an interview Tuesday.

She insisted she and her co-authors remain proponents of flu shots, saying the risk "if it's real" would no longer be an issue because next fall's flu shot will protect against the pandemic H1N1 virus.

If real, the effect appeared to relate to the small window when humans were initially exposed to the first pandemic virus to emerge in 41 years.

"We don't believe that the results we found in the spring, summer of 2009 should deter people from getting seasonal vaccine in subsequent seasons," Skowronski said.

Skowronski and her co-investigators were castigated in some circles for not releasing their data last fall so that others could study them and try to determine if people who had previous flu shots were at elevated risk.

They did, however, verbally outline the findings at length to public health officials federally and provincially in Canada as well as to the U.S. Centers for Disease Control and the World Health Organization.

Still, making public health policy during a pandemic based on data most people hadn't seen was far from ideal, said Dr. Ross Upshur, head of the University of Toronto's Joint Centre for Bioethics.

"There was a whole pile of issues on process that were less than optimal," he said, adding he believes there still needs to be discussion about when public health priorities trump the rules of scientific publication.

"Putting the imprimatur of a high impact peer review journal first is I think what we need to have the discussion about," Upshur said.

Journals generally won't publish material that has already been placed in the public domain.

Skowronski said the researchers had no choice but to hold off until publication, so that the findings could become part of the scientific record, to be explored by others.

"We thought it was our duty to see this through as scientists," she said.

Skowronski admitted one journal turned the work down -- a turn of events that added to the delay in publication. Meanwhile studies of varying quality from Mexico, the U.S., Australia and Britain have failed to find the effect seen in Canada.

Skowronski now hopes other scientists -- influenza immunologists especially -- will mount studies to see if there was something about the body's response to the new virus that could explain this unexpected effect.

Loeb said some researchers have data sets that can be explored to try to shed more light on the issue. He has one -- a prospective study of vaccinated and non-vaccinated Hutterite communities.

Dr. Ed Belongia, director of the epidemiology research centre at the Marshfield Clinic Research Institute, in Marshfield, Wis., also has data. He's looked at it -- and it does not corroborate the Canadian findings.

Belongia said based on the Marshfield data, previous seasonal flu shots neither raised the risk of catching pandemic H1N1, nor offered any protection against it.

Still, he praised the Canadian work, saying the researchers evidently tried hard to find flaws in their studies that might have led to inaccurate results.

"Having read the actual paper as opposed to just hearing rumours about it, I'm impressed with the level of scrutiny they gave their data in terms of looking for potential bias and confounding. So in that sense I can't easily dismiss it," he said.

But the fact it hasn't been seen elsewhere still troubles Belongia. "I think it still falls under the rubric `the Canadian problem' -- even though we don't know what that means yet."