The relationship between obesity and cardiovascular disease mortality may be underestimated, according to a study of the Swedish population and published in the British Medical Journal.

Numerous studies already showed that obesity have harmful effects and associated with higher rates of death from cardiovascular disease, diabetes and some cancers. By contrast, a low BMI would be associated with increased mortality from other causes such as respiratory disease or lung cancer.

However, the researchers say that these associations could be altered by other factors such as smoking or socio-economic motivations. They also underline the fact that a serious illness such as lung cancer, leads to both weight loss and higher mortality.

A team from the University of Bristol (Great Britain) and the Karolinska Institute in Sweden tried to control these potential changes by comparing the BMI of people between 17 and 25 and mortality among their parents.

Their analysis was based on more than one million Swedish parent-child pairs, measuring the height and weight of children between 1969 and 2002. According to the results of previous studies, the researchers showed strong associations between high BMI for children and parental mortality due to cardiovascular disease, diabetes or cancer.

By contrast, they found no evidence that a low BMI in children corresponded to a heightened risk for parental death from respiratory disease or lung cancer.

“These results suggest that the apparent negative consequences of low BMI on mortality from respiratory diseases and lung cancer might be overly concerned, while the adverse effects of a high BMI could be very undervalued” the researchers concluded.

It is considered that an adult is overweight if their BMI (ratio of height and weight per square meter) is above 25. Obesity means above 30 and below 20 are thinness.