How do you tell members of your audience that half of them could get , and not cause panic?
A few weeks later, when there are hopeful signs that the outbreak in the UK is not spreading out of control, how do you avoid the charge of overselling the story?
If there is a right way to report H1N1 swine flu, it must involve an honest admission about the degree of uncertainty. Journalists, and in particular specialists, are often uncomfortable admitting that they don't have definitive answers.
But if someone speaks with absolute certainty on how H1N1 swine flu will develop, they are probably best avoided. There is simply too much that scientists don't know yet, such as how contagious the virus is, or how virulent it might become.
. Despite all the computer modelling and years of planning, scientists can't say how the next pandemic will play out. So if we are unsure how this will develop, we should make that clear.
Numbers can be scary and misleading. The 1918 flu pandemic killed about 50 million people, perhaps many more. The official UK estimate for deaths from the next flu pandemic is as much as 750,000.
From those two figures alone, you could panic everyone. But use some context, and it all sounds less frightening.
In 1918, we didn't have anti-virals or antibiotics to deal with the complications of flu. The 1968 pandemic killed around one million people worldwide. Bear in mind that normal seasonal flu kills between 250,000 and 500,000 people each year and it doesn't look so bad.
The official UK estimate for deaths ranges from 50,000 to 750,000 excess deaths, but takes no account of the effect that anti-virals and other containment measures may have in reducing mortality.
And remember that in winters, when there is an epidemic of seasonal flu, you can get 30,000 extra deaths. Flu, in any form, should not be underestimated.
So what happens if a pandemic is declared next week? Unless the virus is spreading in Britain, it should not directly affect the day-to-day lives of most of us.
We simply can't say when an epidemic of H1N1 will develop here; all we can predict is that it is likely at some point. Given the warmer summer temperatures, we might escape an epidemic for several months or more. The virus could return in the autumn. Previous pandemics have started mild and then become more virulent.
Several people have asked me whether this is a "good time" to catch swine flu. The theory goes that if you get a mild disease now, you will then have immunity for when the virus returns in a more serious form next winter.
Well, that's possible. But then again, it might not work out like that. It would be irresponsible to risk spreading a virus to which most of us appear to have no immunity.
The death rate is low, but you or your children might be unlucky. You might get mild symptoms, but your neighbour who already has a compromised immune system might not. So if you get an invite to a swine flu party, just tell them to keep their germs to themselves.
Annette Street Primary School in Glasgow after a five-year-old showed symptoms of the virus. However, I expect that most media reports will focus on another school - one in Berkshire. 
I'm 
