TORONTO - The World Health Organization's current definition of a flu pandemic is too broad and should be revised to avoid potential false alarms, a journal editorial published Monday argues.

The editorial, by infectious diseases specialist Dr. Peter Gross, said changes made to the definition in the past few years could capture some seasonal virus changes that don't deserve to be classified as pandemics.

"My concern is going forward that if there are minor changes in a drifted strain, the WHO's current definition could include that as a new virus and declare that as a pandemic," said Gross, who works at Hackensack University Medical Center in New Jersey.

"And if they do, they could be declaring pandemics every one to three years for H1N1s or H3N2s or whatever Hs and Ns spill over to the human population and spread."

Drift is a term that refers to the small mutations that flu viruses undergo on an regular basis. Those changes require frequent updating of seasonal flu vaccine, but don't qualify as a pandemic. That's because the changes are too small to allow the drifted viruses to get past the immune systems of people who've had recent exposure to similar viruses.

The term pandemic is used in flu science to describe the emergence and spread of a new virus to which most people have no immunity.

It used to be thought that for a pandemic to occur, a virus of a new subtype would have to emerge. That means a virus with a new hemagglutinin protein on its exterior -- the H number in a virus's name.

The current H1N1 virus has rewritten that rule. That's because while H1N1 viruses remotely related to this virus have been circulating for decades, the new H1N1 is so far apart from them genetically that most people under age 60 have little immunity to it and it has been able to spread rapidly.

In the editorial in BMJ Clinical Evidence, Gross objected to the fact that the revised WHO pandemic phases say a pandemic could be caused by an "animal or human-animal influenza reassortant virus." Reassortants are hybrids, viruses made up of genes from two or more flu strains or subtypes.

The previous definition, published in 2005, defined a pandemic virus as a new subtype, meaning a virus with a new hemagglutinin.

The change broadens the definition and could see viruses that actually aren't much different from previously circulating viruses qualify as pandemic viruses, Gross said in an interview.

He said the definition should be rewritten to specify that pandemics are caused when a virus of a new hemagglutinin subtype or when a virus of an existing subtype that is substantially different genetically starts to spread among people.

Under that latter stipulation, the new H1N1 virus would qualify as a pandemic. Study has shown it shares only about 72 to 75 per cent of the amino acid profile of seasonal H1N1 viruses, which is a huge distance in flu virus terms. Seasonal viruses are generally between 97 and 98 per cent similar to one another genetically.

A spokesperson for the World Health Organization said the global health agency may revisit the pandemic definition, but not while the pandemic is underway.

"There's never any reason to exclude going back and looking at things, if a revision is justified," Gregory Hartl said, adding the agency's top flu expert, Dr. Keiji Fukuda, has spoken about the possibility of refining the language of the pandemic phases once the current crisis has passed.