The recent article in the Sunday Express (May 30th 2010) suggesting that being overweight is not a health risk, is interesting; in one sense it is highly misleading, but in another it draws attention – albeit inadvertently – to a very important point.
The research to which the article refers was carried out by Jarret et al and is published in the International Journal of Obesity, a highly reputable scientific publication1. The investigators studied the relationship between body mass index (BMI), age, gender and health status in 9071 women and 8880 men, collected between 1988 and 2006. The measure of health status was defined by the use of current prescribed medication. Based on these measures, the researchers found little relationship in current health status among normal versus overweight individuals. Thus, a cursory interpretation of the evidence might suggest that being overweight is not a predictor of poor health. However, this is not what the study says.
In fact the amount of medication being used was significantly greater in the obese groups when compared with normal weight individuals, though the effect was mainly in those aged 40-years or over. This is to be expected, since the adverse impact of weight becomes greater with age. There is also the question of whether current use of medication is a reliable indicator of health. But the more telling point made by the investigators, is that body mass index (BMI), though widely used, is a poor discriminator of risk. Put simply, the researchers failed to find a strong association between overweight and poor health not because there is no association, but because they used the wrong instrument.
There is nothing new in any of this. The inadequacy of BMI as a predictor or risk has long been recognized by those of us who work in the field2. BMI – a measure of relative weight for height – was invented by Adolphe Quetelet (1796-1874) a Belgian mathematician. Whilst useful as a simple measure of overweight in epidemiological studies, it is a poor indicator of risk at the individual level. It discriminates unfairly on the basis of gender, race, age and fitness and does not reflect the degree or distribution of body fat. In fact, in the same edition of the International Journal of Obesity, a study by Okorodudu et al shows that commonly used BMI cut-off values to define overweight and obesity fail to identify half of the people with excess body fat!3
The problem is that the National Institute of Health and Clinical Excellence (NIHCE) Guidelines use BMI as the criteria for deciding who is eligible for obesity surgery. In order to qualify, an individual must have a BMI of 35kg/m2 or greater in the presence of co-morbidities (type II diabetes, hypertension etc) or a BMI of 40kg/m2 or above. But this leads to absurdities. For example, a woman in generally good health with a BMI of 41kg/m2 would be eligible for surgery, whereas an Asian gentleman with established type 2 diabetes, hypertension, raised cholesterol levels and a BMI of 34.8kg/m2, would be denied, despite the fact that surgery in his case could be life-saving.
The continued use of BMI as the main eligibility criterion for weight loss surgery in the UK, is outdated and discriminatory. It should be abandoned in favour of a model which more accurately reflects risk at the individual level. Given scarce resources, it is essential that those who are most at risk from their obesity should be at the front of the queue for weight loss surgery.
Dr David Ashton
27th May 2010
___________________________________________
References
1. Jarrett B et al. The influence of body mass index, age and gender on current illness: a cross-sectional study. In J Obes 2010; 34:429-36
2. Pories WJ et al. Beyond the BMI: The Search for Better Guidelines for Bariatric Surgery. Obesity 2010; E-Pub
3. Okorodudu DO et al. Diagnostic performance of body mass index to identify obesity as defined by body adiposity: a systematic review and meta-analysis. Int J Obes 2010; 34:791-9
