Ratio of waist circumference to height may be better indicator of need for weight managementBMJ 1996; 312 doi: https://doi.org/10.1136/bmj.312.7027.377 (Published 10 February 1996) Cite this as: BMJ 1996;312:377
- Former science director, British Nutrition Foundation Former research assistant, British Nutrition Foundation Ashwell Associates, Ashwell Street, Ashwell, North Hertfordshire SG7 5PZ
- Professor of public health epidemiology Health Promotion Sciences Unit, Department of Public Health and Policy, London School of Hygiene and Tropical Medicine, London WC1E 7HT
EDITOR,—The metabolic consequences of obesity relate to the accumulation of visceral fat, which is seemingly reflected by the waist circumference. We add our support to the proposal that sex specific action levels based on the waist circumference could be used as a measure for managing weight.1 2 We also suggest that the ratio of waist circumference to height may be a superior measure for women as well as men.3
We took data from the 1992 health survey for England.4 For each person (1411 men and 1481 women aged 30-74) anthropometric measurements and ratios were compared with the logarithm of his or her risk of coronary heart disease, calculated from his or her sex, age, blood pressure, cholesterol concentration, and smoking and diabetic status.5
The highest coefficient of correlation with risk of coronary heart disease was with the ratio of waist circumference to height for men (r=0.38) and women (r=0.31) (table 1). Stepwise regression showed that the ratio of waist circumference to height was the two factor variable that accounted for the greatest variation in the risk of coronary heart disease for both sexes. Age standardisation reduced the correlations of all variables that included the waist circumference. This was probably because of the strong positive correlation of waist circumference with age (r=0.25 for men, r=0.23 for women; P<0.001) and because age is used in the calculation of the risk of coronary heart disease. Both the ratio of waist circumference to height and the waist circumference, but not the ratio of waist circumference to hip circumference, can also be used for monitoring the reduction in risk.
One particular advantage of using the ratio of waist circumference to height might be that “unisex” action levels could be specified. The distribution of the ratio is broadly similar in both sexes, mean values being only slightly higher in men than women (0.54 (SD 0.06) v 0.51 (0.07)). Applying action levels based on sex specific waist circumferences1 2 to our sample indicated that 22% of the men and 26% of the women would have to lose weight. Any unisex action levels based on the ratio of waist circumference to height should put more men than women in the higher risk groups.
We therefore suggest that the ratio of waist circumference to height should be used in a public health context so that relative emphasis can be put on weight management for men, who suffer greater metabolic consequences of obesity than women. However, proof of the value of any proposed simple measure for indicating weight management and the scientific validation of proposed categories for action require data from a longitudinal follow up of morbidity and mortality.
We thank J Sainsbury plc for financial support.