British Columbia and Prince Edward Island have joined a growing list of provinces that have announced they will delay part of their seasonal flu shot programs this year, decisions which are partially fuelled by concerns raised by controversial and unpublished Canadian research.
The study, by scientists in British Columbia, Quebec and Ontario, suggests people who got a flu shot last year were twice as likely as those who didn't get a flu shot to catch swine flu.
The unconfirmed data have been a source of worry for provincial and territorial public health authorities who are getting ready to roll out two massive vaccination programs this fall and winter.
And they have caused consternation, both abroad and at home, among experts who worry the work will undermine confidence in flu shots and potentially vaccines in general.
"I think it really is unfortunate because either way it's going to set back legitimate public health goals," said Dr. Ross Upshur, director of the University of Toronto's Joint Centre for Bioethics and a primary care physician who has been heavily involved in pandemic planning over the years.
"And as a consequence, we're going to have fractured and inconsistent vaccine policies in all the provinces, which doesn't serve the aim of understanding influenza or better control."
The research - led by Dr. Danuta Skowronski of the British Columbia Centre for Disease Control and Dr. Gaston De Serres of Laval University - is being analyzed by a medical journal and by an independent panel set up at the behest of the Public Health Agency of Canada. Researchers in the U.S., Australia and Britain have looked for and haven't seen the same effect.
Eventually scientists may come to a conclusion that the work is likely right or wrong, but by then the public may have drawn its own conclusions, Upshur said.
"Scientists are used to debating and dealing with the uncertainties," he suggested. "And for the public, it's just: 'Can't trust any of this."'
But Dr. David Patrick, director of communicable disease epidemiology at the BCCDC - and a colleague of Skowronski's - said that while it is difficult to come out with "inconvenient findings" at this time, it is important to put the work to the scientific test.
"These studies are under review right now. We want that process to run properly with the independent international scientific review in terms of getting a good idea about what these studies mean in the long run," he said at a news conference in Victoria where B.C. officials revealed their vaccination plans.
"But we think that we've taken the middle path here, which is rational based on the epidemiology alone, let alone considering these studies."
Besides B.C. and P.E.I., Ontario, Nova Scotia, Quebec and Saskatchewan had announced they will only give seasonal flu vaccine early in the fall to people 65 and older and residents of long-term care facilities. The rationale is that older adults don't appear to be at much risk of catching the pandemic virus, but they are susceptible to seasonal flu viruses.
Pandemic vaccine programs will be rolled out when the H1N1 vaccine becomes available in early November. Following the delivery of that vaccine, these provinces say they will resume giving seasonal flu shots, though some have left open the door to further modifications depending on the state of the science at the time.
New Brunswick, Newfoundland and Labrador and Manitoba are currently proceeding with their normal seasonal flu vaccine programs before the pandemic vaccine rollout. Alberta and the three territories have not yet declared what they plan to do.
Dr. Perry Kendall, B.C.'s chief medical officer of health, said the three-step approach protects against the theoretical risk identified by the study and simplifies the logistical challenges of administering these large vaccination programs over a short period of time.
"Should the perplexing findings of the studies on prior seasonal influenza vaccination be replicated and stand up to critical review and analysis, we will have averted a potential risk scenario," Kendall said as he and other provincial officials announced the province's plan in Victoria.
Prince Edward Island announced it too would go this route.
"This was a complex decision, and we wanted to ensure that our decision for P.E.I. was informed and evidence based," Dr. Heather Morrison, chief health officer for the department of health said in a release.
"This year is a different flu season, but our overall goal is the same; we need to target those most at risk for the flu strains that are in the community this season, to ensure the right people receive the right vaccine at the most appropriate time."
Influenza and public health experts outside this country have dubbed these findings "the Canadian problem," often preferring not to talk about a study they haven't seen suggesting link they hope isn't real.
Several, though, have gone on the record to say they think it will turn out that the work is flawed in some way. A commonly mooted suggestion is that there is a selection bias at work, meaning the type of people studied are not representative of the population in general and therefore the findings can't be generalized.
"There are obviously in any kind of retrospective study many, many options for bias and quirks of methodology to influence the kinds of associations you're looking at - which in this case are of relatively small magnitude," said Dr. John Treanor, an influenza vaccine expert at the University of Rochester in New York State.
Treanor said the lack of a plausible biological explanation for why a flu shot might make one more susceptible to a virus not contained in it makes him skeptical of the findings, though he acknowledges he hasn't seen the data.
"Right now I'd say: 'Gee, you've got this sort of observation that's interesting but kind of small in magnitude and it doesn't really make any biological sense.' So it would take some convincing to make me think it's real," he said.
"Not that it's impossible. But I'd say: 'Gee, you have to be really careful about interpreting that."'