The Guardian
London, UK
28 November, 2006



It's long been accepted as the most accurate indication of good health, but now the efficiency of the body mass index is being questioned. Peta Bee reports

What do Brad Pitt and Russell Crowe have in common with most members of the England rugby and football teams? Those perfectly honed, toned bodies, of course. Except if you rely on the most widely used measurement of fatness, the body mass index (BMI) - which classifies this collection of supremely fit-looking individuals as either overweight or obese. So concerned have some researchers become with such inaccuracies, that they are calling for the BMI to be replaced with more accurate measures of health status.

Devised by the Belgian statistician Adolphe Quetelet, the BMI has been used to define weight for more than 100 years. It is used by hospitals and doctors' surgeries, insurance companies, university researchers, drugs companies and slimming clubs; and professions such as the fire and police services require prospective employees to have a BMI that does not exceed healthy recommendations.

Part of its appeal is its simplicity of calculation: weight in kilograms is divided by height in metres squared - someone with a BMI of less than 18.5 is considered underweight, between 18.5 and 24.9 is "normal", 25 to 29.9 is "overweight" and 30 or greater is clinically obese.

But the measurement's downfall is that it does not take into account body composition - whether or not excess weight is fat or muscle - which is why fit people often find themselves in the fat category of the BMI rating system. Among those leading the call for the BMI to be replaced is Dr Margaret Ashwell, a visiting research fellow in nutrition at Oxford Brookes University and a former member of the government's Food Advisory Committee. "The important thing to consider is how body fat is distributed around the body, as the real problems occur when fat accumulates in the central abdominal region," Ashwell says.

Now scientists from the renowned Mayo Clinic in Rochester are questioning the accuracy and usefulness of the BMI. Reviewing data from 40 studies involving 250,000 people with heart disease, they found that while severely obese patients had a higher risk of death, overweight people had fewer heart problems than those with a normal BMI.

Because muscle weighs more than fat, many physically fit people are mistakenly classified as "overweight", while they are actually less likely to die young than a "normal" weight individual whose excess weight is mostly fat.

Dr David McCarthy, a senior lecturer in human nutrition at London Metropolitan University, looked at waist circumference as a predictor of obesity in a study of children published last year in the International Journal of Obesity. It was found that the average waistband of two-year-old girls has increased by more than 5% in a decade, while that of boys has grown by 4%. The measurements represented a far greater increase in fat than had been suggested by studies using the BMI.

But if BMI is outdated and inaccurate, what else might be used in its place? Tim Cole, a professor of medical statistics at the Institute of Child Health at Great Ormond Street Hospital, says that "body mass index is a cheap and cheerful way of measuring fatness but is not terribly good in detail". Cole suggests that a waist circumference measurement "is more informative, in that it is a direct measure of the part of the body that tends to accumulate fat". Having a waistband of more than 88cm (35in) in women and 102cm (40in) in men indicates the highest risk of cardiovascular and metabolic disease. There is an increased risk of the diseases for women with measurements of more than 80cm (32in) and men whose measurement is over 94cm (37in).

"Measuring waist circumference can help identify patients at higher risk of these problems," says Dr David Haslam, a GP and chair of the National Obesity Forum. "Yet surveys we have carried out show that many people, including some doctors, are unaware that this measurement is a good health predictor."

Ashwell recommends a waist:height ratio "that is applicable to everyone, even children". She has developed her own chart that is already being used by some practice nurses, heart charities and private medical companies such as BUPA. "Interestingly, when you work out the number of people who are obese using the waist circumference or waist:height ratio, there are more UK men at risk than women. This underlines the problems with the BMI, which suggests the opposite to be true."

Getting out a tape measure is a simple way to keep tabs on your fatness at home, but what else is available? Dr Beckie Lang, spokesperson for the Obesity Resource and Information Centre (ORIC), says that skin caliper tests, such as those performed at gyms, can give an accurate measurement of body fat "provided the test is done by someone experienced as it is prone to error". Otherwise, body-fat scales, such as Tanita, are "fairly accurate", she says. "They work by sending a very small current through the body when you stand on the scales," Lang explains. "Muscles conduct the current, fat resists it and a computer inside the scales converts these figures to arrive at your body fat percentage."

Ashwell argues that GPs, insurance companies and the general public need to become used to a different way of thinking. "What we want is for people to understand that a fat middle is bad news," she says. "It is not how much you weigh, but where the excess fat gathers that causes problems."




http://www.guardian.co.uk/medicine/...1958687,00.html

i'm glad to see someone fighting (or at least, the media NOTING that someone's fighting) to kill the BMI standard. personally, i wish they'd go by overall bodyfat percentage rather than choosing one specific part of the body to focus on. i've got a lot of fat around my midsection, yes, but i've also got a lot of excess skin that was damaged from repeated weight loss and gain, not to mention a twin pregnancy. that's going to throw off any measurements anyone takes

oh well... at least it's a start.

There is a correlation of fat around the middle with metabolic diseases though, so that's probably why they're going for that particular area.

Weight and height are easy to measure and very repeatable. Waist measurements very depending upon when the tape is put and how much the person being measured tenses the abs.

I don't know that things will change anytime soon. It seems like simplistic ideas in our society are very hard to kill. Ideas like "fat makes you fat".

The thing about BMI is it's based on a simple calculation, based on two easily available variables: height and weight. So the number crunchers can easily whip up statistics for large segment of the population and make sweeping statements. And who knows, maybe used on a large scale is it somewhat useful.

So I don't see them replacing the standard with body fat measurement as that is more difficult to obtain and less easily available.

Previous to BMI, the gold standard was the charts made by the insurance companies. That was a pretty poor standard as well. I don't know if BMI is an improvement or not over it.

The problem I see with straight measures of waist size is that it too has flaws - a man or woman who is taller than average could be tagged at "increased risk" when there is no real increase in risk - they're taller and their waistline is going to be larger than someone shorter.....and I'm not talking about an inch or two above average, but women who are 6' like me definitely can have a 32" waist without excess fat at their waistline - in tall clothing, a 32" waist is either a 12 or 14 depending on brand....and when you're six feet tall a size 12 looks completely different than someone who is five feet tall. That's the big problem I see with this - you can't use the same measure on someone who is five feet tall and expect it to work on someone six feet tall as accurately!

I really think it comes down to the stronger measure of waist-hip ratio, not the straight measure of waistline if we're talking a population-wide measure to use for risk.....in women at least.