Influenza is traditionally regarded as an illness that poses a particular risk to the very young and the very old, but young adults can also be at risk from its potentially devastating effects, as past epidemics show.
Last winter's flu was regarded as relatively innocuous, certainly as far as the numbers affected were concerned. The winter of 1999 produced an epidemic - officially recorded as more than 400 cases per 100,000 people. It is too early to predict what this year's flu will be like, although it is more than 30 years since we had a world pandemic. Many experts feel we are due another "bad one" soon.
The principal concern among doctors and scientists is that we will, at some time in the future, be exposed to a strain of flu virus similar to the infamous one of 1918-1919. Known as Spanish flu, this highly virulent disease killed up to 40 million people in less than a year - three times the fatalities recorded in the four years of the first World War.
The big difference between the 1918 flu and outbreaks before and since is the number of young healthy adults who were killed by the epidemic. The first case in a US army camp, for example, was recorded on September 14th, 1918. By the end of October, some 17,000 of the 45,000 troops in the camp, which was based west of Boston, had succumbed. Ninety young men died every day at the height of the epidemic.
The mortality was equally high in Europe. Flu killed 20,000 people in Ireland, which had a death rate similar to Scotland's. In Italy, 500,000 people perished. The virus infected both David Lloyd George, the British prime minister, and Woodrow Wilson, the president of the United States.
Cinemas and theatres closed. People routinely wore face masks. The streets were fumigated. And in the United States, a skipping song quickly became popular: "I have a little bird and its name is Enza. I opened the window and in flew Enza."
But that was then. Such an outbreak could never happen again, could it? Experts are not so sure, particularly since the 1997 near miss in Hong Kong - an episode that is both reassuring and worrying. If you contracted flu in 1997, you were most likely to be infected with H3N2, a descendant of the last pandemic strain that had emerged in Hong Kong in 1968.
On May 9th, 1977, a three-year-old boy became unwell with what seemed to be an upper-respiratory-tract infection. Five days later, he became progressively more ill and was admitted to intensive care, with respiratory failure. Within two weeks, and despite the best medical treatment, he died.
Analysis of fluid from the child's windpipe revealed a previously undetected form of influenza virus, now known as H5N1. This rang alarm bells among infectious-disease experts around the world, and an investigative team was dispatched from the Centers for Disease Control and Prevention, in Atlanta.
It soon became apparent that there was a link between the human flu outbreak and the widespread death of chickens in Hong Kong's rural territories. Almost 7,000 birds had died in March 1997 of an avian flu. The virus responsible was H5N1.
This was a worrying discovery. The natural reservoir for the influenza virus is wildfowl: ducks, geese, chickens and turkeys. And although the flu virus does not travel directly from birds to humans, there is a common link with swine. Both avian and human influenza viruses can replicate in pigs. When they mix, dramatic genetic shifts can follow, bringing with them the potential for new human epidemics.
The 1977 outbreak now began to look serious. South-East Asia is one of the most densely populated areas of the world - where hundreds of millions of people live and work in close contact with domesticated pigs and birds - so the authorities embarked on a huge cull of the chicken and duck population in Hong Kong and its New Territories.
By August 1998, no new H5N1 had shown up in Hong Kong. The 18th and final human victim of the outbreak had fallen ill on December 28th, 1997, the day the bird slaughter began. That there had been no more cases since then was no coincidence: the sharp surveillance by scientists had stopped the spread of a potential pandemic.
The last serious pandemic was in 1968. When a particularly virulent influenza hits, it is like nothing else we have grown accustomed to labelling "flu". The normal aches and pains become prostration; people are debilitated to the point at which they cannot raise their heads from the pillow.
A retired Irish GP described the rural village in which she lived and worked as being like a ghost town during the 1968 outbreak. "There was no one on the streets. Entire households were affected, with no one stirring out for weeks. Essential supplies were delivered by neighbours lucky enough not to be affected." The rate of house calls rocketed; she worked 18-hour days for weeks on end before succumbing herself.
The 1968 virus swept around the globe. It was a H3N2 version - coincidentally, labelled Hong Kong flu - which is still part of the flu jab that people will get this winter.
It is the ability of the virus to mutate and disappear that keeps flu experts alert and uneasy. The 1933 pandemic was caused by a H1N1 that then disappeared for more than 40 years, before re-emerging in 1977.
Dr Michael Oldstone, director of the viral laboratory at the Scripps Research Institute, in California, and author of Viruses, Plagues, & History, says: "The reappearance of the 'Russian' flu in 1977, a virus first isolated in 1933, raises the uneasy possibility that a return of the 1918-1919 pandemic is possible."
The possibility is strengthened by the realisation that viruses can remain inactivated in a frozen state. In August 1998, an expedition set off for Spitsbergen, an Arctic island belonging to Norway, to exhume the bodies of coal miners who had perished during the 1918 outbreak. The hope was that the bodies would be preserved beneath permafrost, so that tissue samples might help deduce the genetic code of the Spanish flu.
Although initial analysis was disappointing, because of poor preservation, researchers are still hopeful of identifying the 1918 strain from the samples.
Why bother? Because of the extreme virulence of the 1918 outbreak. Not just the numbers, but also the sort of people who were killed by that flu - the majority were young fit people who died predominately from pneumonia caused by the influenza virus - and because we know that it is possible, at least theoretically, for it to re-emerge 80 years later.
Admittedly, we now have antibiotics that will minimise the death rate from any secondary bacterial pneumonia. But because of the absence of antibiotics and detailed scientific research in 1918, we can never be sure whether the pneumonia associated with Spanish flu was viral or bacterial. Young men died so quickly after symptoms started that it must be reasonable to assume that it was the virus itself that overwhelmed their bodies.
In the meantime, the likelihood, as always, is that the flu jab will protect against whatever strain is on its way. Vaccination is up to 70 per cent effective, and it remains the mainstay of treatment for influenza. But until the first isolates of this winter's flu are analysed and compared across Europe and the United States, a frisson of anticipation will spread through the scientific world. Could this be the next big pandemic?