Hot on the heels of its headline “More young motorists are drink-driving, police warn” (August 2) readers of The Independent were told “Huge fall in deaths from drink-driving” (August 5). Two contradictory stories in the same week: any readers who are still paying attention must be thoroughly confused by now.

The first headline I have already dealt with: it came from a misleading press release from the Association of Chief Police Officers, based on statistics unworthy of the name. The second is kosher, insofar as road accident statistics ever are. It comes from data published by the Department for Transport.

Let’s stick to deaths, as they are hard to get wrong, although thanks to the Coronial system in England (Straight Statistics, passim), easy to get late. The figures show that deaths resulting from drink-drive accidents fell by a remarkable 35 per cent between 2009 and 2010, from 380 to 250. As recently as 2006 (before alcohol went “out of control”, remember) the total was more than twice as high, at 560.

Table RAS51001 from the DfT is so remarkable that its salient details are worth reprinting here. Again I have limited the figures to deaths, because casualty figures depend too much on police recording methods. The table shows the deaths caused by drink driving have fallen dramatically since 1979.

They halved in the 1980s, almost halved again in the 1990s, languished between 2000 and 2006, and have since halved again. The latest year’s fall is a bit too good to be true, and is based on provisional figures – those coroners again. The final result is likely to be a bit higher.

I wish a few more people were aware of these figures, which give the lie once more to the conviction – aired most recently in a singularly one-sided edition of Panorama – that we are drinking ourselves into oblivion. The statistics tell a different story. Every death, of course, is to be regretted. But isn’t it worth celebrating that there are so many fewer to regret than there were a generation ago?

That needs journalists to shake the scales from their eyes and start challenging the propaganda that passes for wisdom in this field. Little sign of that from The Independent, which concluded its two-paragraph story by reporting that deaths from all road accidents (not simply those in which drink was a factor) went up by 6 per cent in the first quarter of 2011.

That is true, but for the whole 12-month period between April 2010 to March 2011, deaths fell by 10 per cent, from 2,081 to 1,870. The reason for the increase in the first quarter was heavy snow in the first quarter of 2010, which cut traffic and accidents in that year (as the statistical release explains). Too difficult to read, I guess.