Waist size 'predicts early death'

Tuesday August 10 2010

“Men and women with large waists are at increased risk of dying young,” BBC News has reported. The article said that people with a large gut all have the same risk of early death, regardless of whether their body mass index (BMI) is healthy.

The research behind the news tracked more than 100,000 men and women aged 50 and over for nine years. It found that people with very large waists – 47 inches (120cm) or more for men and 42 inches (110cm) or more in women – were about twice as likely to die during the follow-up period compared to thinner people. These deaths were not just from weight-related problems.

It has long been thought that fat collecting around the tummy puts people at greater risk than weight put on elsewhere. The strengths of this study lie in its large size and the collection data on the participants over time rather than their consulting records alone. It may eventually lead to guidance that both BMI and waist circumference need monitoring by people trying to maintain or regain a healthy weight. Regardless of overall weight, people should avoid becoming too big around the middle.

Where did the story come from?

The study was carried out by researchers from the Epidemiology Research Programme at the American Cancer Society in Atlanta, Georgia. No funding sources were reported. The study was published in the peer-reviewed medical journal, Archives of Internal Medicine.

BBC News reported this study fairly and included a quote from the National Obesity Forum saying that the research “underlines the message that fat inside the belly is dangerous”.

What kind of research was this?

The data for this study came from a large US cohort study designed to identify causes of cancer and prevent cancer that ran from 1997 and 2006. This subsequent study looked at waist circumference (WC), a measure of abdominal obesity and linked to higher rates of death independently of other measures of obesity such as body mass index (BMI). The researchers wanted to explore this link further as they say few studies have looked at WC in relation to sub-categories of BMI or the effect of  very high waist circumferences, which are becoming increasingly common.

The researchers were able to sub-divide the large population into men and women and eight different waist circumference categories, with sufficient data in each to examine links between weight circumference, BMI and risk of mortality in a meaningful way. This well-designed study also corrected for many of the other factors that could have been influencing the results, such as race, educational level, marital status, smoking status, alcohol use, height and physical activity.

What did the research involve?

The researchers explain that a large waist circumference (WC) is linked to higher rates of type 2 diabetes, abnormal lipid levels, measures of inflammation and heart disease. The theory is that WC may be associated with these conditions because it indicates “visceral fat”, a collection of fatty tissue surrounding the organs within the abdominal cavity. A person does not necessarily have to be obese or overweight to have high levels of visceral fat, which means that a simple waist measurement might be a better predictor of death risk than BMI alone.

The data was analysed for 48,500 men and 56,343 women who were all over 50 and enrolled in the Cancer Prevention Study II Nutrition Cohort. In the original cancer study participants had completed a 10-page questionnaire sent to them during 1992/3. This included information on things such as age, sex, medical history and behavioural lifestyle information on smoking and physical activity. WC was first measured by the participants themselves in 1997: they were sent a tape measure and asked to measure their WC just above the navel to the nearest quarter inch while standing, and to avoid measuring over bulky clothing.

The BMI of each participant was calculated from weight reported in the 1997 survey and height reported in the 1982 survey. They excluded participants with missing, extreme or implausible values for BMI or WC, or plus those with unknown smoking status. They also excluded a relatively large number of people (7,997 men and 7,482 women) who had shown a weight loss of 4.5kg or more between 1992 and 1997, on the basis that their weight loss may have been related to unknown or unrecorded disease. Deaths were recorded in the usual way on death certificates and the researchers were automatically notified of any deaths among the participants. The researchers then looked at how WC related to specific causes of death..
Results were analysed appropriately and presented as the relative risk of death adjusted for age, race, educational level, marital status, smoking status, alcohol use, height, and physical activity. Analyses of women were also adjusted for hormone therapy. An additional separate analysis was done adjusting for BMI.

What were the basic results?

Over a nine-year follow-up period, 9,315 men (19.2%) and 5,332 (9.4%) women died. When the researchers analysed the data with the adjustments for BMI and other risk factors they found that in both men and women a very high WC carried about twice the risk of mortality than those in the lowest WC group:

  • 2.02 times greater for male WC of 120cm or more compared to WC of less than 90cm (Relative Risk [RR] 2.02, 95% confidence interval [CI], 1.71 to 2.39)
  • 2.36 times greater for female WC of 110cm or more compared to WC of less than 75cm (Relative Risk [RR] 2.36, 95% CI 1.98 to 2.82)

Increasing waist circumference was associated with increasing mortality across all categories of BMI (normal, overweight, and obese). In contrast, when analysed by categories of WC, mortality rates did not increase with increasing levels of BMI.

Increased WC was associated with increased risk of mortality for all specific causes of death they analysed; cancer, cardiovascular, respiratory and all other causes. The strength of the association was strongest for respiratory disease and all other causes of death, followed by cardiovascular disease and then cancer as a cause of death.

How did the researchers interpret the results?

The researchers conclude that their results emphasise the importance of waist circumference as a risk factor for mortality in older adults, regardless of BMI.


This large, prospective cohort study was well-conducted and analysed. The new finding is that, after making appropriate adjustments for BMI, increasing levels of WC were associated with higher risk of mortality in men and women. Although this was suspected before, this is the first study that has analysed such a large number or participants in so many sub-categories of waist size and within the three standard clinical categories of BMI (normal, overweight, and obese).

Other points worth noting are that:

  • Waist circumference was measured and reported by the participants, which may have introduced some error.
  • As this is an observational study, the link between WC and mortality might be overestimated because of confounding by unmeasured or unknown factors associated with both larger WC and higher mortality.
  • As all study participants were 50 years or older and nearly all white. The results may not apply to younger populations or those of other racial or ethnic backgrounds.

Overall, this large, well-conducted study emphasises the importance of considering obesity in different ways. In any of the standard clinical groups of BMI, (normal, overweight and obese) it appears that avoiding excessive abdominal obesity can reduce risk of premature death.

Analysis by Bazian
Edited by NHS Choices