Title of article:
Reference data for obesity: 85th and 95th percentiles of body mass index (wt/ht2) and triceps skinfold thickness
Authors: Must A, Dallal GE, Dietz WH
Journal: Am J Clin Nutr, Apr 1991;53(4):839-46
Abstract
Body mass index (BMI) and triceps skinfold thickness (TSF) are commonly used measures of adiposity in clinical and epidemiologic studies. The 85th and 95th percentiles of BMI and TSF are often used operationally to define obesity and superobesity, respectively. Race-specific and population-based 85th and 95th percentiles of BMI and TSF for people aged 6-74 y were generated from anthropometric data gathered in the National Health and Nutrition Examination Survey 1 (NHANES I). The complex sample design of the survey is reflected in the reference values presented. Racial differences in these extremes of the distribution do not emerge until adulthood. Researchers may choose population-based, race-specific, or age-specific criteria for obesity on the basis of assumptions underlying their specific research questions.
Comments and Key points
This data from the NHANES I study ( 1971-1974) is now out-of-date, because more recent studies, NHANES II and NHANES III have been performed. However, this was the data that was originally suggested to establish "overweight" and "obesity" thresholds, at the 85th and 95th percentiles. (However NHANES II data was actually used when the definitions became official.)
So, for historical reasons, here are those percentiles shown as data in a table:
| White | Men | Women | ||
|---|---|---|---|---|
| Age | Overweight | Obesity | Overweight | Obesity |
| 6 yrs | 16.52 | 17.75 | 16.14 | 17.49 |
| 7 yrs | 17.31 | 18.98 | 17.16 | 18.93 |
| 8 yrs | 18.10 | 20.22 | 18.19 | 30.36 |
| 9 yrs | 18.88 | 21.45 | 19.21 | 21.78 |
| 10 | 19.67 | 22.66 | 20.23 | 23.20 |
| 11 | 20.47 | 23.87 | 21.24 | 24.59 |
| 12 | 21.28 | 25.01 | 22.25 | 25.59 |
| 13 | 22.12 | 26.06 | 23.13 | 27.07 |
| 14 | 22.97 | 27.02 | 23.87 | 27.97 |
| 15 | 23.82 | 27.86 | 24.28 | 28.51 |
| 16 | 24.63 | 28.69 | 24.68 | 29.10 |
| 17 | 25.44 | 29.50 | 25.07 | 29.72 |
| 18 | 26.08 | 29.89 | 25.34 | 30.22 |
| 19 | 26.53 | 29.98 | 25.58 | 30.72 |
|
20-24 |
27.02 | 31.43 | 25.78 | 31.20 |
| 25-29 | 28.15 | 31.89 | 27.16 | 33.16 |
| 30-34 | 28.76 | 32.04 | 29.38 | 34.58 |
| 35-39 | 29.17 | 32.12 | 29.25 | 35.35 |
| 40-44 | 29.34 | 32.21 | 29.90 | 35.85 |
| 45-49 | 29.36 | 32.15 | 30.38 | 36.02 |
| 50-54 | 29.29 | 32.04 | 30.66 | 35.95 |
|
55-59 |
29.23 | 31.95 | 30.93 | 35.88 |
| 60-64 | 29.17 | 31.87 | 31.20 | 35.80 |
| 65-69 | 29.10 | 31.78 | 31.46 | 35.70 |
| 70-74 | 29.01 | 31.69 | 31.70 | 35.58 |
These thresholds were deemed suitable, once upon a time. But later, the World Health Organization decided that simpler definitions were needed, at BMI of 25 and 30, for overweight and obesity, respectively. But all such standards are arbitrary.
The chart below shows how the NHANES 1 criteria compare to the more recent CDC charts criteria. In children, they are remarkably close. But for adults, apparently the NHANES 1 values were too high to use as standards, so the CDC and WHO lowered their criteria to 25 and 30.
From this same article, I've also plotted a graph of the 15th percentile BMI values for Men and Women. I think this is important, because it represents skinny people, who remain skinny throughout their lives. Notice that their BMI increases with age. In my opinion, this indicates that it is unrealistic to expect adults to maintain the same Body Mass Index values that they had at age 18.
Review & comments by Steven B. Halls, MD, Last modified:
23-June, 2008, Copyright.
See the other Body Mass Index publications.
