Nick Trefethen's new BMI
-------------------------------
The limitations of the body mass index (BMI) proposed by Adolphe Quetelet (see article page 14) and popularized by the World Health Organization have been known for a long time. How many great athletes, with their sleek build, have a BMI exceeding 25, the threshold for being classified as overweight? The two main reasons for this are the substantial muscle mass, whose density is much higher than that of fatty tissue, and a tall stature.
According to Professor Nick Trefethen of the University of Oxford, the BMI does indeed markedly overestimate the overweight status of tall people. He therefore proposes the formula I = 1.3 x P/T2.5, where P denotes the weight expressed in kilograms and T the height measured in meters. Normal values under this index fall within the interval \[18.5; 25\].
What should we make of this formula? The factor 1.3 obviously plays no role. Its sole purpose is to ensure that this new index produces values similar to those of the old one. The exponent 2.5, on the other hand, does change things. Quetelet had indeed noted that "if man grew equally in all his dimensions, weights at different ages would be as the cubes of heights", but he had realized that this bore no relation to reality whatsoever. The exponent 2 that he proposes, however, is certainly too low.
If it turned out that the index proposed by Nick Trefethen is more accurate, certain WHO recommendations would become obsolete. The consequences are easy to imagine, given the number of preventive treatments based on BMI. Nick Trefethen's index, however, likewise fails to account for characteristics related to muscle mass.
