Much of medicine involves weighing risks and benefits. The calculation is often a fine one. When questions are rarely asked, and reviews conducted infrequently, such assessments are all the harder.

For this analysis of two widely used medicines, Tamiflu and Relenza, in one of the most common infections, seasonal flu, researchers trawled more than 500 studies going back a decade to assess effectiveness in children.

Even then, only four trials, involving 1,243 children with flu who received antivirals, could be found. A further three trials, involving 863 children given the drugs preventatively were also identified, but are arguably of less relevance to the current pandemic. The conclusions researchers came to — that there is no clear evidence antivirals prevent complications in children with seasonal flu — is a crucial contribution to strategy.

While the Department of Health contends the extrapolation to swine flu is “questionable”, the researchers conclude that the effectiveness of widespread antiviral use is more questionable.

They assert, on more than one occasion, that the Government must re-examine its swine flu policy, which has included the distribution of more than 300,000 packs of Tamiflu in the past three weeks alone. While it is a costly exercise — with low-strength Tamiflu retailing at £8 a pack — the greater burden will be if it is being distributed with little or no clinical benefit, while heightening the risk of the H1N1 virus developing antiviral resistance. Most GPs are exercising caution, recommending Tamiflu only when clinically required. The National Pandemic Flu Service, working on a swine flu symptoms algorithm by phone and website, is less discerning. Worse still, there are now occasions of health professionals seeing flu-like symptoms, sending for the Tamiflu, and leaving other conditions such as meningitis and measles dangerously misdiagnosed.

Concerns raised in yesterday’s BMJ paper demand review, as do those published in The Lancet last week, suggesting a similar lack of benefit. Children are known to be at high risk, with more than 40 per cent of pre-school children and 30 per cent of school-age children expected to catch the virus. School children are also the main agent spreading flu in households.

As the researchers observe, the last review of antiviral medication use was in 2005, and “a more up to date assessment of benefits and harms is now required in the light of the widespread use of these drugs in the current pandemic”. This must incorporate all available data — including the highly guarded findings held by the pharmaceuticals’ manufacturers.

This is now an imperative for the Government, because the concerns raised by the BMJ findings are far outweighed by the fact that, after Britain’s much trumpeted years of pandemic planning, these researchers were the first people in far too long to do such calculations.

Times Recommends

• The chilling lie behind two webcam suicides
• Crochet and crafts at the hen party
• Want to be a genius? Practise, says geneticist