Why the body mass index (BMI) is virtually useless for assessing health

The standard way of gauging body size is the body mass index (BMI) – calculated by dividing an individual’s weight in kilograms by the square of their height in metres. According to conventional wisdom, a BMI of 18.5 ” 24.9 is regarded as ‘healthy’, while BMIs of 25-29.9 are considered ‘overweight’ and those of 30 or above are considered ‘obese’.

While the BMI forms the basis of the advice health professionals give to individuals about their weight, it actually has a number of significant limitations. First of all, it tells us nothing about body composition. It is entirely possible, for instance, to have a heavy, reasonably well-muscled body with relatively little fat and be classified as ‘overweight’ or even ‘obese’. Plus, it is known that if excess fat is present in the body, where it resides appears to have an important bearing on its health effects. Basically, it has emerged that it is weight that congregates around the middle of the body (known as ‘abdominal obesity’) that is linked with an increased risk of conditions like heart disease and diabetes.

Because of this, there has been a recent vogue for doctors and researchers to recommend that individuals have the ratio of their waist circumference to hip circumference (the waist-to-hip ratio) measured. Waist-to-hip ratios are increasingly being seen as a better measure of body weight and size than the BMI with regard to health.

So, I was a bit surprised to see this week a report in which researchers recommended we persist with the BMI. This suggestion came after the publication of a study in the American Journal of Cardiology which looked at the relationship between BMI and the waist-to-hip ration and risk of a condition called ‘pre-diabetes’ (in which an individual is not diabetic but appears to be heading that way) [1]. The researchers involved in this study found that the BMI performed as well as the waist-to-hip ratio in identifying those in a pre-diabetic state. The study also found that BMI correlated better with supposedly deranged cholesterol levels too. This led one of the authors of the study to comment: “why not just stick with BMI?”

While this particular study did appear to show an association between BMI measurements and pre-diabetes and cholesterol levels, it is important to bear in mind that these things are not actually conditions in themselves. Neither do they kill people. Because of this, when assessing the relationship between body measurements and health, it’s far better to focus on good hard measures such as heart attacks or death. So how well does the BMI rate when looking at genuine measures of health such as these?

One study published Last year in the Lancet spanning 52 countries assessed the relationship between BMI and heart attack risk [2]. It revealed that once other associated factors such as smoking and exercise habits were taken into consideration, BMI has no significant bearing on heart attack risk.

While research which asks questions of the conventional wisdom regarding the relationship between BMI and heart health is interesting, it is perhaps not as telling as research which seeks to establish the relationship between this measure of body weight and overall risk of death. Last year also saw the publication of a study in the Journal of the American Medical Association which found that being underweight or obese (a BMI 30 or more) seemed to confer an increased risk of death [3]. However, somewhat surprisingly, this study found that compared to those in the ‘healthy’ BMI category, those categorised as ‘overweight’ (a BMI of 25 ” less than 30) were actually at reduced risk of death, although this was not statistically significant.

To my mind, the inability of the BMI to tell us anything about body composition and the distribution of any fat, and its apparent inability to predict true health effects makes it virtually useless as a body measurement. While there may have been a recent call for us to stick with it, there is actually a strong case against this. The evidence suggests that the waist-to-hip ratio is a preferred vital statistic. For instance, the Lancet study mention above found that a higher waist-to-hip ratio was very clearly associated with increased heart attack risk. Ideally, men and women should have waist-to-hip ratios no larger than about 0.90 and 0.83 respectively.

References:

1. Farin HMF, et al. Comparison of Body Mass Index Versus Waist Circumference With the Metabolic Changes That Increase the Risk of Cardiovascular Disease in Insulin-Resistant Individuals. 2006;95(8):1053-1056

2. Yusuf S, et al. Obesity and the risk of myocardial infarction in 27,000 participants from 52 countries: a case-control study. Lancet 2005;366(9497):1640-9

3. Flegal KM, et al. Excess deaths associated with underweight, overweight, and obesity.
JAMA. 2005;293(15):1861-7.