Summary

For older people (75 and older) an increase in BMI does not correlate with an increased risk of death, whereas the waist-to-hip measurement does.

Introduction

In recent years we've been urged to calculate the body mass index (BMI) to get a 'better' reading of overweight and obesity that just the poundage. And lately we've been persuaded that waist girth is sometimes a better predictor of trouble than either of these. Nevertheless, clinical studies usually still classify their subjects by BMI: 20-25 kg/m2 is normal, 25-30 is overweight, 30-35 is obese, and so on. For people who have a BMI of 26 and are unhappy being considered overweight, help is at hand, in the shape of a new major study. Older persons may soon be able to use the tape measure instead of the scale.

Researchers in London, UK, examined the association of BMI, waist girth, and waist-hip ratio to mortality in people aged 75 and above. They published their findings in the American Journal of Clinical Nutrition, and we summarize them here.

What was done

Almost 15,000 volunteers aged 75 or above were recruited by 53 family practioners in the UK. They were not resident in a nursing home or known to be terminally ill when they were entered the study.

The health assessments made at entry included height, weight, and waist and hip circumferences. Waist was measured midway between the iliac crest (highest point of the pelvis at the side) and the lowest rib margin. Hip circumference was measured as the widest measure over the buttocks and below the iliac crest. In addition, a wide range of physical, social, and psychological information was collected. The cause and date of any deaths among participants during the follow-up period were provided by the UK Office for National Statistics.

The participants were classified into three sets - one set each for BMI, waist circumference (WC), and waist-hip ratio (WHR). In each set, five equal-sized categories of subjects were made, called quintiles, according to their BMI, WC, or WHR. Comparisons were made between the quintiles in a set with all-cause mortality and cardiovascular mortality.

What was found

There were approximately 5,700 men and 9,100 women in the analyses; their average age was 80, and over 90% of them were non-smokers. During the 6 years of follow-up, 6,600 of them died - almost half from cardiovascular disease.

In non-smoking men and women, there was no increase in mortality with increasing BMI values; the so-called hazard ratios (i.e.the risk level) in all other categories were less than that set for the 'normal' category (BMI below 23.0 for men, 22.3 for women).

On the other hand, increasing WHR was associated with increasing hazard ratios in men and women; this association was statistically significant, i.e. it could not have occurred by chance alone.

Findings for cardiovascular mortality were the same; a clear association with WHR, no association with BMI. Waist circumference was not associated with all-cause or cardiovascular mortality.

What these results mean

This is not the first study to show that BMIs of 25 to 27 are not a risk factor for all-cause or cardiovascular mortality in older people1. But experts continue to use the BMI for ascribing risk and treatment for seniors. The results here clearly show that abdominal obesity (or rather adiposity) is responsible for increased mortality, independently from other factors. And WHR is clearly the best way of assessing this.

BMI is a poor measure of body fat. It fails to discriminate between body fat and lean mass. And measures to improve longevity should concentrate on a reduction in body fat, rather than simple weight loss. Actually, percentage body fat would be the best measurement to have, but this requires special equipment for accurate results. Some personal scales have an impedence measure included, which can give a trough estimate of body fat, and there's a calculator (see link below) which can give an approximation. Another way is to use calipers to measure the skin fold at a number of body sites.

However, the WHR is the easiest for you to measure at home; just divide your waist girth by your hip girth (see above for exactly where to measure). Ideally, women should have a waist-to-hip ratio of 0.8 or less, and men should have a waist-to-hip ratio of 0.95 or less. So get busy with the tape measure - and no cheating!

Source
• Weight, shape, and mortality risk in older persons: elevated waist-hip ratio, not high body mass index, is associated with a greater risk of death. GM. Price , R. Uauy , E. Breeze, et al. , Am J Clin Nutrit, 2006, vol. 84, pp. 449--460