New study sharpens focus on problems of obesity

(Alexander Raths)

Cardiovascular disease linked to obesity may be worse than thought while health problems associated with being underweight may have been overstated, according to a study published by the British Medical Journal (BMJ) on Wednesday.

The paper, written by doctors in Britain and Sweden, seeks to finetune a well-known tool - the body mass index (BMI) - which is used to measure obesity and ill-health.

BMI entails taking one's weight in kilos and dividing it by the square of one's height, in metres. A BMI of 25-30 is generally considered overweight. while a figure of above 30 indicates obesity.

Previous studies have already found a big link between BMI and higher risk of death from cardiovascular disease, diabetes and some cancers.

In contrast, a low BMI - in other words, people who are very skinny - is associated with increased mortality from respiratory disease and lung cancer.

But many experts have questioned the findings about low BMI, arguing that the figures could be skewed by something called reverse causality.

For instance, diseases such as lung cancer, which cause weight loss, are being factored in as low BMI, they argue. And smoking and poor socioeconomic circumstances may also cause bias.

Seeking to find out more, specialists from Britain's University of Bristol and the Karolinska Institute in Sweden set out to get a wider view of the BMI benchmark.

They measured BMI and mortality among more than a million pairs of Swedes, comprising mother-and-son pairs and father-and-son pairs, over 50 years.

They found strong patterns of mortality and BMI.

The link was especially noticeable in cardiovascular disease - for which prevailing estimates may be "substantially underestimated" - as well as diabetes and kidney cancer.

But there was no evidence of an association between low BMI and an increased risk of respiratory disease and lung cancer mortality.

The findings are important, because they imply that obesity programmes could yield major health benefits, says the study.

"Suggestions to the contrary are probably misguided," it adds.

ri/ach