Detecting obesity based on skinfold thicknesses

Institute of Child Health
Centre for Paediatric Epidemiology and Biostatistics
30 Guilford Street
London WC1N 1EH
United Kingdom
E-mail: tim.cole{at}ich.ucl.ac.uk

KM Flegal

Centers for Disease Control and Prevention
Hyattsville, MD

William H Dietz

Centers for Disease Control and Prevention
Atlanta, GA

Dear Sir:

Zimmermann et al (1) showed, in 6–12-y-old Swiss children, that the 95th BMI percentile of the Centers for Disease Control and Prevention (CDC) cutoffs have a higher sensitivity than do the International Obesity Task Force (IOTF) obesity cutoffs to detect obesity based on skinfold thicknesses. They concluded that the CDC cutoffs are superior to the IOTF cutoffs, but this conclusion is misleading for 2 reasons.

It is true that sensitivity is higher with the CDC cutoffs; however, Zimmerman et al did not highlight the higher specificity of the IOTF cutoffs. In Table 5 of Zimmerman et al's article, it shows that the specificities for detecting obesity with the CDC and the IOTF cutoffs are 97.3 and 98.6, respectively, in girls, and 96.9 and 99.5, respectively, in boys. So, the false-positive rate is twice as high in girls and >6 times as high in boys with the CDC cutoffs.

This tradeoff between sensitivity and specificity is well-known, and focusing on one while neglecting the other is not the best way to compare cutoffs. The reason why the IOTF obesity cutoffs have lower sensitivity (and higher specificity) is simply because they are more extreme, as the authors point out, which leads to lower prevalence rates of obesity (2, 3).

The second concern is that the gold standard used by Zimmermann et al, percentage body fat based on skinfold thicknesses, is set at the 95th percentile of the distribution, which matches the corresponding CDC cutoffs but is lower than the IOTF cutoffs. If the definition of obesity had been based on the 99th percentile for percentage body fat, the comparative results for sensitivity and specificity would have been different, ie, the sensitivity of the IOTF cutoff would have been much higher. In this instance, as in others, it is important to compare like with like.

Given the continuing rise in the prevalence of child obesity, there is also some benefit in having a more extreme cutoff available to focus on the extreme group of children most at risk of obesity-related complications, such as those with type 2 diabetes (4).

ACKNOWLEDGMENTS

None of the authors had a conflict of interest related to the letter.

REFERENCES

1. Zimmermann MB, Gubeli C, Püntener C, Molinari L. Detection of overweight and obesity in a national sample of 6–12-y-old Swiss children: accuracy and validity of reference values for body mass index from the US Centers for Disease Control and Prevention and the International Obesity Task Force. Am J Clin Nutr 2004;79:838-43.
2. Flegal KM, Ogden CL, Wei R, Kuczmarski RL, Johnson CL. Prevalence of overweight in US children: comparison of US growth charts from the Centers for Disease Control and Prevention with other reference values for body mass index. Am J Clin Nutr 2001;73:1086-93.
3. Kain J, Uauy R, Vio F, Albala C. Trends in overweight and obesity prevalence in Chilean children: comparison of three definitions. Eur J Clin Nutr 2002;56:200-4.
4. Sinha R, Fisch G, Teague B, et al. Prevalence of impaired glucose tolerance among children and adolescents with marked obesity. N Engl J Med 2002;346:802-10.