Correspondence (letter to the editor): Quintessence Unjustified

We would like to comment on the descriptive analysis of Lenz et al., with their conclusion that overweight – rather than obesity – is not generally associated with increased mortality and morbidity.

It is problematical to define normal weight as corresponding to a BMI between 18.5 and 25 kg/m², as it has long been known that a very low BMI is linked to increased risks of mortality and morbidity. It would be better to take the lower limit of the normal range as being 20 kg/m², even though this is not implemented in the current recommendations of the professional societies or the WHO.

The statement about mortality is evidently derived from the study of the EPIC cohort, including evaluated data from 359 387 subjects (1). It must first be said that there is in fact an increase in the overall mortality in the EPIC cohort between a BMI of 23.5–25 kg/m² and a BMI of 28–30 kg/m². This corresponds to 11% for women and 8% for men. Even though these changes sound small, they correspond to a considerable loss in life expectancy for a cohort with an average age on entry of 51.5 years. Moreover, the BMI is greatly influenced by smoking and consumptive diseases. In almost all studies, smokers have a lower BMI than nonsmokers. The formulation “Whether it is smoking alone or the associated lower weight that contributes to the increased risk [of bronchial carcinoma]” implies that obesity may protect smokers from bronchial carcinoma.

As it is generally accepted that smoking is the predominant cause of bronchial carcinoma, it is difficult to establish whether there is also an effect of BMI. Consumptive diseases also regularly lead to low BMI values. It follows that a very low number of patients with consumptive diseases could enormously distort the results – particularly in age groups with low mortality.

We therefore conclude that the authors’ quintessence on overweight and mortality is not justified on the basis of the available data and is even dangerous.