BSE cases 'underestimated'

However, statisticians at Imperial College London say that there should be no extra threat to human health, despite almost twice as many infected carcasses entering the food chain.

Eating BSE-infected meat is firmly thought to be the cause of the fatal human disease vCJD.

So far, well over 100 people have been diagnosed with the illness, and the majority of these have already died.

The government put in place a programme of culling in order to stop older cattle - more likely to carry BSE - from being eaten.

Earlier research into the epidemic suggested that approximately one million cattle had been infected, 870,000 of which entered the food chain.

Screening clue

The Imperial research, published in a Royal Society journal, used the results of a recent mass screening of cattle to check original estimates.

Its authors now say that it is likely that as many as 1.9 million were infected, and that 1.6 million were turned into human food.

This means individual exposure to infected meat was more or less double previous estimates.

There are concerns that vCJD may incubate for an unknown number of years before causing illness - and this is making it difficult for experts to predict roughly how many humans will develop the disease.

However, Professor Christl Donnelly, who led the latest research, said the extra exposure did not mean that there were now likely to be many more cases of vCJD.

She said: "If cigarette companies suddenly announced that cigarettes actually contained twice as much tar as they had been saying, it would not mean that many more people would die from tobacco-related diseases.

"What it would mean is that tar was much less dangerous than previously thought."

Under-reporting

She said that, in the same way, her findings meant that BSE-infected meat was likely to be less infectious to humans than previously thought.

She said there were two possible explanations for the large disparity between the two figures.

One was that farmers were deliberately not reporting cases of BSE developing in their herds, reducing the official number of infected animals.

The other was that cattle at the very earliest stages of BSE illness might not obviously have "mad cow disease", but have other symptoms - such as a drop in fertility or milk production - which would mark them out for early slaughter as part of normal farming practice.

She said that a combination of both factors seemed probable.

However, the results do suggest that the BSE epidemic in the UK is well and truly over.

Falling away

The team's estimates for the number of infected animals entering the food supply has been falling since a peak in 1989, and this year, it was likely that no infected animal had been slaughtered for meat.

However, Professor Donnelly's analysis of the prospects for other EU countries makes slightly less comfortable reading.

Italy and Belgium, for example, have slightly higher BSE infection rates among their herds, data suggest.

Professor Donnelly said that this did not yet constitute a major problem, but should serve as a "wake-up call" to keep on the look-out for the disease.

She said: "After all, our intention is to eradicate BSE across the EU."

A spokesman for the Food Standards Agency said that the study suggested that measures taken in 1996 - after the emergence of vCJD - had proved effective.

There is an ongoing review of the policy of slaughtering cattle over 30 months of age, and the agency said that the findings would be considered as part of this.