Avoid the chook shed

In terms of where the next influenza pandemic might come from, the bird flu strain H5N1 continues to be a big concern, according to Kelso. "It has spread so far and there have been many human cases. It continues to infect poultry flocks and has spread throughout many wild bird populations around the world, across Asia through into the Middle East and Europe and into Africa. We are very fortunate that it has not yet spread into populations in the Americas, North or South, or into Australasia."

Although migratory birds contribute to the virus spread, the poultry trade is seen as the biggest threat - live birds and uncooked meat being transported around the world. It is impossible to predict, however, whether H5N1 is more likely than another influenza virus to mutate and give rise to a pandemic, and the WHO is closely monitoring several other avian viruses.

This is where the work of the influenza surveillance network becomes so vital. "Certainly, any human infections that turn out to be not caused by a standard seasonal human influenza virus require particular attention," Kelso says.

The issue of transmission from birds to human and potential transmission between humans is scientifically challenging and the subject of intensive research in the area. One possibility for an influenza virus striking the right variation for human-to-human spread is by co-infection of a human cell by a human virus and an avian virus like H5N1, Kelso explains.

The influenza virus has a segmented RNA genome in eight strands, and thus can undergo re-assortment. Potentially, then, if two viruses were present at the same time in the same cell, a new virus might be created with RNA material from both parent strains.

"In that scenario, you could get a human virus in which only the haemagglutinin molecule has changed. So, it would be easily transmissible between humans, but we would have no immunity against it."

This scenario was proven in principle by a group in Atlanta, who used reverse genetics to create a viable, hybrid virus by mixing the genomes of H5N1 and various human viruses. Kelso reassuringly highlights that multiple changes must have to happen because, if only one or two mutations had to happen, it probably would have by now.