Across Britain, pharmacists are being marshalled as the front line troops in the battle of the bulge. And to reinforce their role, battalions of them have been taking the opportunity to measure up their customers during National Obesity Week, given its theme, inspired by the National Obesity Forum, is to get people thinking about their Body Mass Index or BMI.

Pharmacists – including some of the big high street names – manage to flog many pills and potions without a thought for the extent to which they are complicit in misleading their customers. A lot of these products have no genuine scientific testing, but may soon be forced to come up with proof or drop the bogus claims.

Talking to a leading pharmacist recently, he was in no doubt that the pharmacists' profession needs to clean up its act in respect of the dud stuff masquerading as diet and weight control treatments on the shelves. When someone walks through the chemist's door to get a BMI check, they need a trusted professional hand to find out not only if they would genuinely benefit from a weight-loss regime, but to warn them that spending quite extravagant sums of money on these dubious alternatives is the quickest way to lose pounds – sterling, that is.

Strangely, BMI, this once quirky statistical curiosity (invented by a 19th-century Belgian boffin, Alphonse Quetelet), has become a touchstone in discussion of overweight and obesity. Many years ago, I was involved with an international working group preparing to tell everyone that you should "Know Your BMI". It was already a growing fad in the US in the 90s and, like all American fads, it quickly crept over here as the marketeers discovered that if you call snake oil a weight-loss product, people will willingly part with their hard-earned cash and blame themselves if it doesn't work. And if the placebo factor and a good measure of diet and activity does achieve a result, then the snake oil did work, after all.

The real snag about "knowing" your BMI is that you need to understand its significance. It isn't a straightforward number like a speed limit; and just like driving, because you happen to be below the maximum limit doesn't mean you aren't at risk of a very unhealthy crunch if you hit a barrier.

BMI 30 was thought up as a convenience to make it easier to count up how many of us are really fat or obese. It was primarily intended to let latter-day Quetelets crunch the population figures to produce health statistics. The idea that body mass index should be used in a more personal way has a long history, and for more than half a century, insurance companies have been using their own height and weight tables to work out how much profit they can make out of your life policy premiums. The fatter you are, the sooner you are likely to die – and the sooner company has to shell out. It works in reverse for the pensions companies, which love clients to have high BMIs because it means they will pay out far fewer monthly cheques before having to deal with a claim for death benefit.

It wasn't insurance actuaries (although they weren't far off the mark), but a group of WHO technical advisers who decided, in 1995, that BMI 25 was a critical point for overweight – a decision conveyed in a report on anthropometry, which is just another way of talking about standards for measuring people. Before that, there was no fixed point, so that different countries and ethnic groups gathered their overweight statistics in different ways, making it hard to compare one country with another. The experts landed on the nice round figure of BMI 25 to mark the turning point for even rounder figures, and then plumped for BMI 30 as a convenient benchmark – which was at first described as another phase of overweight but after two years became clearly defined as obesity.

The BMI issue wasn't just about overweight either. A low BMI may indicate something's amiss, so a BMI of 18.5 was chosen to mark the low point for health, below which emaciation is a risk. Why the UK ignored this and maintained that BMI 20 should be the bottom line, so to speak, for the British remains unclear. It serves no real purpose except to worry exceedingly healthy people with a BMI of 19.

So, during National Obesity Week, we have all been exhorted to consider our BMI – and may even have been persuaded to have a check-up in the local chemist's. Beware. You aren't off the hook if your BMI is below 25; and if you are Asian, you are likely to be in trouble even if your BMI is down to 23. It depends on your percentage of body fat, which you can't get from a BMI check.

Sometimes, a simpler way of doing things is overlooked. Years ago, the aptly-named Scottish nutrition guru, Professor Mike Lean, came up with a tape measure that makes it comparatively easy to get an accurate measurement of your waist. The bad news is that BMI disguises the fact that many more of us have bulging waistlines – the real symptom of "killer fat". And the BMI statistics have done us a disservice in masking just how bad the fat problem really is. According to the last trend analysis in the Health Survey of England 2007 – an update is due next month – the proportion of men with a "raised" waist circumference (more than 102cm) rose from 20% in 1993 to 33% in 2007; while for women, the proportion with a raised waist circumference (more than 88cm) rose from 26% to 41%. There is no escaping that these represent the true obesity prevalence figures – 33% of men and 41% of women – far worse and far more serious than many of us, even health professionals, realise.

So, when checking your BMI, the lithe may take comfort. If your waist circumference is below 94cm for a man, or 80cm for a woman, then a BMI up to 30 does not mean increased risk to health. Sadly, the likelihood of anyone falling into that category that appears to be slim. If you are below BMI 25 but have a "high" waist circumference, there is, according to the WHO's expert advisers, also no increased risk. I suspect very few of us will be able to walk out the GP's surgery or the chemist's door without a worried frown and an uncomfortable tightness around the belt.