Study: TMI a better index than BMI

Geschatte leestijd: 4 minuten

New research from the U.S. shows that TMI is a better index than BMI in determining overweight in adolescents.

BMI during growth

In the earlier article on Monday about Body Mass Index, I already mentioned the caveats of using BMI as an indicator of the ratio between lean mass and fat mass. Moreover, BMI is merely a reference to a norm and the degree of deviation from it. The implications of a certain deviation depend on various circumstances. When using BMI, besides gender, you should take into account factors such as age and ethnicity. Particularly in still-growing adolescents, BMI appears to struggle with accurately estimating body fat. Researchers from the University of Alabama, Birmingham, therefore refer to challenging methods for correcting BMI for growing children [1]. Their research shows that even with these statistical corrections (use of percentiles or “Z scores”), BMI is less accurate than TMI.

TMI Index?

I already mentioned it on Monday in the announcement of this article; TMI does not stand for a method to measure ‘corrected penis length’. I repeat it to save you the confusion of a Google search. T.M.I is the abbreviation of (Tri-)Ponderal Mass Index or also called Corpulence Index. In short: In the mid-19th century, a Belgian came up with the idea that dividing someone’s weight by their height squared could yield a good indicator of ‘relative body weight’. He named this index in 1832 after himself, the Quetelet Index[2]. Chances are that name doesn’t ring a bell to you. Quetelet’s best chance of eternal fame was ruined when in 1972 Ancel Keys decided to call this index ‘body mass index’ (BMI) [3]. Fifty years earlier, however, a Swiss thought there must be a better method. After all, what is better than dividing your weight by your height squared? Exactly, dividing by your height cubed. He also naturally named the new index after himself; Rohrer’s Index. TMI rolls off the tongue much better. So: BMI = weight in kg/m² TMI = weight in kg/m³ It should be noted that there are several variations of the formula. BMI and TMI are both formulas in a quest for the ‘normal ratio’ between height and mass. Height is fairly straightforward here, you grow in height along only one axis. BMI would be less suitable because of its two-dimensional nature since you’re working with squares. A body like a sheet of paper. TMI assumes a cube because you’re working with the third power. TMI would be more suitable for short and tall people whose body dimensions BMI errs in more often than TMI [4 to 8]. Unfortunately, people don’t grow proportionally like a cube does. The discussions about the value of BMI, TMI, and other indices are a feast for statisticians. However, the best outcome remains a statistical approach. For personal relevance, you then rely on more information about the amount of body fat and its distribution over your body.

TMI more suitable for measuring body fat in adolescents

The direct reason for this article is the findings from a new study from the U.S.[1] BMI would particularly struggle with still-growing children. In their study published in JAMA, the Britons demonstrated that TMI is a much better indication of body fat than BMI in the age range of 8 to 17 years.

BMI would frequently misclassify children into the ‘overweight’ or ‘obese’ category. In many American school districts, this is also stated in a report. It also occurs in the Netherlands that children are weighed and measured at school to assess their BMI. The researchers compared data on body composition of 2285 white youths aged 8 to 29 who participated in the National Health and Nutrition Examination Survey (NHANES) between 1999 and 2004. They compared BMI with various indices and assessed the following points

• stability with age
• accuracy in predicting percentage body fat
• accuracy in predicting overweight

TMI was found to be the best of all indices in determining the amount of body fat. Especially in young men in whom BMI struggled the most to show an accurate picture, even with the use of percentiles. (Corrected) BMI would incorrectly classify 19.4 percent of young men aged between 8 and 17 into the ‘overweight’ category. TMI did this in 8.4 percent of cases. The researchers therefore believe that TMI will ultimately replace BMI for assessing the relative weight of young people. Further research should determine whether TMI is also more accurate in other target groups. As indicated in the previous article on BMI and also here, it remains the best possible estimate based on averages. Useful in the absence of better data. If you have better data, such as body fat percentage or circumference measurements, this provides much more insight than BMI or TMI.

References

1. Courtney M. Peterson, Haiyan Su, Diana M. Thomas, Moonseong Heo, Amir H. Golnabi, Angelo Pietrobelli, Steven B. Heymsfield. Tri-Ponderal Mass Index vs Body Mass Index in Estimating Body Fat During Adolescence. JAMA Pediatrics, 2017; DOI: 10.1001/jamapediatrics.2017.0460
2. Eknoyan, Garabed (2007). “Adolphe Quetelet (1796–1874)—the average man and indices of obesity”. Nephrology Dialysis Transplantation. 23 (1): 47–51.
3. Commentary: Origins and evolution of body mass index (BMI): continuing saga
4. Babar, Sultan (March 2015). “Evaluating the Performance of 4 Indices in Determining Adiposity”. Clinical Journal of Sports Medicine. Lippincott Williams & Wilkins). 25 (2): 183. Retrieved August 2015.
5. Fayyaz, Jabeen (June 2005). “Ponderal Index”. Journal of Pakistan Medical Association.
6. Lawrence F. Ditmier: New Developments in Obesity Research. Nova Science Publishers, Hauppauge, New York 2006, ISBN 1-60021-296-4[page needed]
7. Davies, D. P. (1980). “Size at birth and growth in the first year of life of babies who are overweight and underweight at birth”. Proceedings of the Nutrition Society. 39 (1): 25–33. doi:10.1079/PNS19800005. PMID 6988835.
8. ACC/SCN NUTRITION POLICY PAPER No. 19 – Glossary by Lindsay H. Allen and Stuart R. Gillespie