W e are all in the grip of avian-flu hysteria. The World Health Organization (WHO) estimates that the virus could claim up to 7.4 million lives around the world if it successfully mutates and becomes transmissible among humans, while the UK’s chief medical officer Sir Liam Donaldson says it’s a matter of when and not if.

When it happens, it could claim the lives of up to 750,000 Britons, he estimates while, in America, it may kill up to two million people, warns President Bush.

While there are many types of avian, or bird, flu, the disease strain that is worrying health officials is the influenza A (H5N1) virus, first discovered in a flock of terns in South Africa in 1961.

While H5N1 is usually strictly for the birds, a case of infection among humans was first reported in 1997, during an outbreak in poultry in Hong Kong. It caused a severe respiratory illness in 18 people, killing six. Since then, human infections have been reported in Thailand, Vietnam and Cambodia during major H5N1 outbreaks in poultry. Around half the infected humans have died.

Outbreaks among poultry have become common, especially in Southeast Asia and, this year, cases have been reported in poultry farms in Europe. Domesticated poultry catch H5N1 from wild birds, which carry the viruses in their intestines.

As far as researchers have been able to establish, virtually all of the human cases have occurred among poultry workers who had direct contact with the birds or with infected areas. Poultry workers who work among domesticated chickens, ducks and turkeys are at risk because the virus can be passed to humans via the birds’ faeces.

Several cases have been reported where a sufferer has been able to infect another person but, so far, there is no evidence that it has been passed on beyond a limited contact group. Officials at the Ministry of Public Health in Thailand reported a typical example in 2004. It involved a poultry worker who had become infected, and who then transmitted it on to her mother and aunt who came to care for her at her hospital bedside. The mother died of pneumonia after being at her daughter’s side for 18 hours; the aunt also developed pneumonia, but survived. Both were found to have the H5N1 virus (N Engl J Med, 2005; 352: 333-40).

In both these cases, the viral infection was transmitted only after extended exposure.

Another study, carried out at the National Pediatric Hospital in Hanoi, Vietnam, showed that it is not highly infectious. Researchers at the hospital questioned 83 staff members, all of whom had been exposed to four confirmed cases of H5N1 infection; of the staff, just two, or 2.4 per cent, were possibly infected with the avian flu virus (Emerg Infect Dis, 2005; 11: 210-5).

So far, it’s been reckoned that the H5N1 virus has killed around 65 people over eight years. To put this in perspective, around half a million people die of seasonal flu every year, including 40,000 Americans and 12,000 Britons (Drugs, 2004; 64: 2031-46).

Why is an epidemic predicted? Why should scientists suddenly fear that the H5N1 virus is likely to mutate soon, and become transmissible among humans, when it has been around for at least 50 years?

There appears to be no scientific basis for the dramatic public warnings that are being issued. One reason given is the rising number of outbreaks in poultry farms around the world, and the increased exposure to humans.

But scientists also point to the cyclical nature of viruses. The world seems to be hit by a major flu pandemic three or four times a century. The major epidemic of the last century was the Spanish flu outbreak that occurred in the United States during 1918-19, killing 40 million people. There were also flu pandemics in 1957 and in 1968 - and the WHO has pencilled in 2005-06 as the date for the next outbreak.