The influenza virus is a protean beast, its multiple strains mutating at lightning speed and giving each succeeding winter a different pathogenic complexion. To keep up, scientists spend the summer studying illness patterns in the southern hemisphere as a guide to what is to come – and developing that year's flu vaccine accordingly.

The approach appeared to be working fine. Indeed, until now, the official line was that vaccination is highly effective and, if only enough people could be persuaded to have the jab, the annual scourge would be less virulent.

Cracks are appearing in the conventional wisdom, however, with warnings from a number of top scientists that flu vaccines are not the guarantee against infection they are purported to be. More troublingly, official unwillingness to acknowledge the jab's shortcomings comes with dangerous implications of its own.

The most immediate problem is one of trust. The furore over MMR is an object lesson in public sensitivity over vaccinations. Government support for the combined vaccine was ultimately proved valid; but suggestions of a link with autism led to a 10-year panic, thousands of unprotected children, and a spike in infection. Overselling the efficacy of the flu jab risks a similar loss of public confidence. Of equal concern, it also takes the pressure off drug companies to invest in developing a newer, better vaccine.

The dilemma is a tricky one. Public health is not an arena that lends itself to nuance, and suggestions that the jab is not 100 per cent effective might result in fewer people choosing to subject themselves to it. But the risks of overstating the benefits are too great to be dismissed.

It is time, then, for a subtler message from the Government. The vaccine is still worth having; but it will neither guarantee the avoidance of illness, nor is it good enough to stop research into finding a better one. Flu is a sophisticated virus; it requires a sophisticated response.