Disease may change to an even deadlier form

Sars resembles the fatal flu virus of 1918

Reinhard Kurth, the head of the German government's Robert Koch Institute, was not exactly encouraging on Friday night. A vaccine for Sars could be developed, he said – but it would take three to five years.

Experts worldwide are still struggling to identify precisely what is causing the illness, and how it is transmitted. Some can already see light at the end of the tunnel. "To me, the epidemic is almost certainly over," said Osman David Mansoor, a World Health Organisation (WHO) scientist who is working closely with health authorities in Singapore. "But there is always a lag time for infections. While we are speaking there may be a whole other lot of infections going on."

The possibility is that Sars will die out after failing to transmit itself, before any cure beyond acute medical care can be devised. But the lurking fear is still that it might find a way to revive itself, like the Spanish flu epidemic which spread throughout the world from a military camp in Kansas in 1918 – carried by US troops to Europe, and then spread to the Far East. Millions of people were infected; millions died. And now, we have air travel to hasten the spread of any new disease.

The similarities with the 1918 flu pandemic are quite striking – once you allow for some of the differences that medical advances have made. True, the mortality rate among those infected with Sars peaked at around 4 per cent, and doctors are getting better at treating it. The 1918 influenza strain at times killed 28 per cent of those infected, but its overall mortality rate was 2.5 per cent.

Medical experts at the Centers for Disease Control and Prevention (CDC) in Atlanta, Georgia, are still puzzling over exactly how Sars progresses. Some think that it is a single "coronavirus" – a superstrong version of the common cold. But others are suggesting that Sars is a pair of viruses, one of which lowers the body's defences while the other takes advantage and causes the rapid descent into pneumonia.

Mostly, those studying the disease have found out what it is not. "In a matter of a day and a half, we'd ruled out 100 infectious agents that might have been the cause of this outbreak," says Dr Frank Plummer, director of the National Microbiology Laboratory (NML) in Winnipeg, Canada, one of the countries that has been seriously affected by infected travellers arriving from Hong Kong.

Tests have been devised to test for the coronavirus: you can have it, yet not have Sars. And some of the people who have had Sars do not have any trace of the coronavirus, although officials at the CDC say it is a 90 per cent match. Early hopes that a simple antibody test would allow for quick testing of passengers stepping off aircraft have been dashed by the NML.

One link with the 1918 flu strain is the potential source of the disease. The theory is that last time, the presence near the First World War trenches of pigs and chickens used to feed the troops allowed potentially dangerous viruses to mutate and move back and forth between the humans and animals, until finally a swine flu virus mutated sufficiently to reinfect humans, who had no resistance to it.

As with Sars, it was not the flu as such that was the killer, but the pneumonia that developed rapidly with it. One group of women was said to have stayed up late to play bridge; the next morning three of the four were dead. So far, no such tales have emerged with Sars. But the lights will be burning long into the night in the laboratories around the world until the infection rate begins to fall.