Waist circumference gain compared with waist circumference as predictors of type 2 diabetes risk

24710 Upper Trail
Carmel, CA 93923

Gary TC Ko

Department of Medicine
Alice HML Nethersole Hospital
Tai Po
Hong Kong
China

Louis J Peterson

Department of Health Sciences
San Jose State University
San Jose, CA

E-mail: goguh{at}aol.con

Dear Sir:

The findings of a recent study published in the Journal, "Comparison of abdominal adiposity and overall obesity in predicting risk of type 2 diabetes among men" (1), indicate that the relative age-adjusted risk of diabetes is best targeted by waist circumference (WC) and that those persons in the highest quintile of WC are at a much greater relative risk than are those in the lower quintiles. Nevertheless, of the 884 diagnosed cases of diabetes, 379 (45%) were found in the first 4 WC quintiles, that is, at a level below the current WC cutoff of 102 cm.

We believe that if abdominal adiposity in this cohort of 27 270 health professionals were assessed in terms of quintiles of WC gain since age 21 y, diabetes cases could be predicted even more effectively than by WC quintiles. According to another report on the same cohort (2), the questionnaires submitted by participants in the study included weight and height at age 21 y. On the basis of our research on physically active, largely normal-weight young men (average age: 31 y) (3), it is possible to estimate WC from weight, height, and age with reasonable accuracy. It can be assumed that most of the members of the Health Professionals Follow-Up Study cohort, who reached age 21 y close to the time of World War II, were lean as young adults. The WC prediction equation from our database (R² = 0.82, SEE = 3.8 cm), adjusted for age 21 y, is as follows:

For the Health Professionals Follow-Up Study cohort, who had an average weight of 72.9 kg and a body mass index (in kg/m²) of 23 at age 21 y, the above equation predicts a WC of 79.3 cm. The formula also converts the weight-gain categories in the same report (2) into WC-gain categories. At age 53 y (the average age in 1986), stable weight corresponded to a 6% gain in WC because of the 32-y age increment (32 x 0.145 = 4.6 cm). The cohort's average weight gain of 9 kg since age 21 y implies a WC gain of 14.7% (with a BMI of 25.8 and a WC of 91 cm). An excess of 20% in WC compared with young adulthood corresponded to a weight gain of close to 15 kg (with a BMI of 27.8 and a WC of 95.7 cm); the highest weight gain category, >19 kg, was associated with a WC gain of >25%. These extrapolations assume that weight gain and, hence WC gain, are evenly distributed over the initial weight and BMI categories.

If WC gain (or %WC gain), as calculated above, better predict diabetes incidence than does WC, despite the uncertainties surrounding these extrapolations, we propose that future studies based on WC gain apply a more accurate method for assessing WC gain. This method requires the determination of the skeletal width and depth of the thorax (4). Fortunately, these pressed skeletal measurements can be performed at any time convenient for ongoing research, because body build is subject to minimal changes over the adult years. On the basis of an equation derived for the same physically active young men, WC projected from thorax dimensions averaged 79.3 cm but ranged from 64 to 95 cm, a clear indication that a given WC does not reflect the same WC gain in all men. We suggest that WC gain—the difference between current WC and WC projected from thorax dimensions—is generally more applicable for assessing abdominal adiposity and related health risks than is WC. New research could then substitute a percentage WC gain cutoff for the existing WC cutoff, perhaps at a WC gain of 20% or 25%.

ACKNOWLEDGMENTS

None of the authors had a conflict of interest to report with respect to the content of this letter.

REFERENCES

1. Wang Y, Rimm EB, Stampfer MJ, Willett WC, Hu FB. Comparison of abdominal adiposity and overall obesity in predicting risk of type 2 diabetes among men. Am J Clin Nutr 2005;81:555-63.
2. Koh-Banerjee P, Wang Y, Hu FB, Spiegelman D, Willett WC, Rimm EB. Changes in body weight and body fat distribution as risk factors for clinical diabetes in US men. Am J Epidemiol 2004;159:1150-9.
3. Heinz G, Peterson LJ, Johnson RW, Kerk CJ. Exploring relationships in body dimensions. J Stat Educ [serial online] 2003;11. Internet: www.amstat.org/publications/jse/v11n2/datasets.heinz.html (accessed 30 June 2005).
4. Heinz G, Ko GTC, Peterson LJ. Waist girth normalized to body build in obesity assessment. Asia Pac J Clin Nutr 2005;14:60-8.