Great question, Kylie!! I started down that track a few years ago and questioned a few prominent body composition experts. It was recommended to locate a NIH document from a meeting (in the mid-90's) when that was discussed by a panel of experts -- I'll try to find that email. Anyway, I think the fact that we have to look beyond PubMed, Google Scholar, etc; -- and that leading experts don't even know -- indicates that the current classification system needs to be revisited. It is my professional stance that 40 should be "obese" and 30-40 = overweight; 20-30 = normal weight; and < 20 = underweight. That said, BMI (or even % fat-free mass) should not be a single assessment of health.

Dr. Bob Kuczmarski (NIDDK) kindly forwarded one of his papers on the topic.

Thanks for your answers. I tracked the BMI and the categories all the way back - it was very interesting. I basically found what Bob did. Starting with the NIH paper I worked my way back through the reference lists.

It started with the Life Tables in 1912 and over the following 70 years a lot of research went into getting more accurate and mathematically superior indicators. I found the 1972 paper that first argues the BMI is superior, then a 1981 book (J. Garrow) book that had the categories first proposed, and a 1985 WHO paper from a meeting in Warsaw that stated that BMI should be the indicator of choice, and that they accepted the Garrow categories. The cut offs have been further developed, but essentially a BMI of 30 is where the relative risk curve changes.

If anyone is after any of the references let me know, I'm happy to share.

Have a lovely day.

Very nice question, We had ask about the begin of the tables and issues...

maybe you would also like the following

Critical analysis of long term weight loss following gastric bypass. Reinhold RB.

http://www.ncbi.nlm.nih.gov/pubmed/7051382

Criteria for definition of overweight in transition: background and recommendations for the United States. Am J Clin Nutr 2000;72:1074–81 Kuczmarski RJ, Flagel KM.

http://www.ncbi.nlm.nih.gov/pubmed/11063431

Keys A, Fidanza F, Karvonen K, Chimaera N, Taylor HL. Indices of

relative weight and obesity. J Chronic Dis 1972;25:329–43.

Revicki DA, Israel RG. Relationship between body mass indices and

measures of body adiposity. Am J Public Health 1986;76:992–4.

Flegal KM. Ratio of actual to predicted weight as an alternative to a

power-type weight-height index (Benn index). Am J Clin Nutr 1990;

51:540–7.

Cole TJ. Weight-stature indices to measure underweight, overweight,

and obesity. In: Himes JH, ed. Anthropometric assessment

of nutritional status. New York: Wiley-Liss, 1991:83–111.

Billewicz WE, Kemsley WF, Thomson AM. Indices of adiposity. Br

J Prev Soc Med 1962;16:183–8.

Khosla T, Lowe CR. Indices of obesity derived from body weight

and height. Br J Prev Soc Med 1967;21:122–8.

Florey CD. The use and interpretation of ponderal index and other

weight-height ratios in epidemiological studies. J Chronic Dis

1970;23:93–103.

BMI in fact is the Quetelet index, which appeared by the end of 19th century. The association with cut-off values for defining obesity is more recent, from the beginning of last century, with the Metropolitan weight table. However, the association of BMI and obesity has become somewhat misleading because the concept of obesity has evolved from someone who is "heavy" (what the BMI tells us) to fat accumulation and excessive adiposity, which is revealed by body composition assessment techniques and BMI does not tell us about body composition. Besides, it is a single direct association of an index with a complex multifactor disease such as obesity.

This is a pretty cool topic; tracking down our error in using BMI.

BMI was developed pretty early; around 1850. The equation was developed by Adolphe Quetlet, sociologist to describe populations.

It is used primarily as tool to describe populations, and does not take into count the individual body composition of the person

Ancel Keys the 1970's published findings in journal of chronic disease that BMI is a good proxy for body fat composition determination/more convenient.

Hope this helps.

Hello Kylie,

I'm looking for the history of the 'underweight' category. On the internet I did find a differentiation between men (