Sir, – William Reville's discussion of the limitations of the Quetelet or body mass index (BMI) as a mechanism to assess the adverse effects of fatty tissue on health is an excellent and timely overview ("Doctors weigh in on importance of BMI in health outcomes", Science, April 1st).
The last year has been one of unprecedented concerns about wellbeing, and many have taken the opportunity afforded by lockdown to tackle aspects of their lifestyle that worried them.
There are two other points worth recognising about BMI that make it undesirable as a goal in this endeavour, particularly for some individuals in society.
One is ethnicity. The methods underpinning the origin of BMI, where life assurance companies looked at pooled data for patterns of mortality, did not represent diverse ethnic groups appropriately.
This is important, because some populations can be obese, incurring the associated risk of disease, at much lower levels than others.
For example, in the Lancet in 2004 a World Health Organisation expert group stated that in defining people as overweight the “cut-off for different Asian populations varied from 22 to 25”.
In certain groups according to their data, obesity is diagnosed at a BMI of 26. Because the world’s most populous countries are in Asia, this is no small matter when setting the ideal BMI standards for humanity.
A second situation in which BMI carries significant caveats is the setting of critical illness. There is a well described, but poorly understood, phenomenon whereby the overweight, and indeed obese, have better outcomes when admitted to intensive-care units than those of ideal body weight.
This “obesity paradox”, as it’s often known, may offer researchers an insight into how our bodies can cope better with the challenge of severe illness, and is the topic of extensive research in the domain of critical-care medicine. – Yours, etc,
BRIAN O’BRIEN,
Kinsale,
Co Cork.