It was widely reported last week that there were "100,000" new A/H1N1 flu cases in England.

The news immediately caused suspicion about the true number of cases since it is virtually impossible to count that many people one by one within such a short period of time.

Where, then, did the number 100,000 come from?

The number was announced by Britain's Health Department and came from the Health Protection Agency. However, the HPA only "estimated that there were 100,000 new cases of flu in England last week (range from 60,000 to 140,000)."

"The estimate is based on the latest weekly consultation rates for flu-like illness, taking into consideration positivity rates for swine flu H1N1 from HPA's virological sampling and a range of assumptions related to the numbers of people attending their GP (General Practitioner)", the HPA said on its website.

However, it should be noted that the number was based on "flu-like illness," not laboratory confirmed cases. Britain announced in early July that the country would not test the virus for each suspect any more, in order to transfer resources to those most in need.

The methodology of the estimate was actually like this: another institution, the Royal College of General Practitioners, has a network of approximately 100 general practices covering a population of approximately 900,000 people. The practices provided data that was calculated to give the result that the consultation rate was 155 per 100,000 in the week ending July 19. The population of England is around 60 million, so there might be nearly 100,000 new suspect cases in the country.

The week before, the consultation rate was 73.4 per 100, 000, so it was reported there were around 50,000 new suspect cases.

That is the basic logic while the actual model used for estimation would be much more complex. The uncertainties make the final estimate a "range from 60,000 to 140,000." The number 100,000 that is widely discussed, is simply the median number and the most probable one.

There are other estimate systems in England, such as the "Q Surveillance" set up by the University of Nottingham and EMIS -- the main supplier of general practice computer systems within the UK -- in collaboration with the HPA.

The estimate of this system was even higher than the HPA number, reaching 221.4 consultation cases per 100,000 within the week. The rate, like the former one, also is related to all flu-like illness cases. The actual confirmed H1N1 flu cases should be smaller than the estimated figures.

(Xinhua News Agency July 26, 2009)