Do overweight people live longer than those of normal weight?

Everybody knows that people living in developed countries are getting fatter. For example, as measured by body mass index (BMI), the weight in kilograms divided by the height in metres squared, Australians have got fatter: see here. To summarise:

• The 1999-2000 Australian Diabetes, Obesity and Lifestyle Study indicated over 7 million adult Australians aged 25 years+ (60%) were overweight. Of these, over 2 million (21%) were obese.
• Men were more likely than women to be overweight, with 67% of men compared with 52% of women (aged 25 years+) being overweight.
• There have been significant increases in proportions of overweight and obese Australians over the last 20 years. From 1980 to 1999/2000, for people aged 25-64 years, the proportion of overweight women increased from 27% - 47%, and the proportion of overweight men increased from 47% - 66%.
• On average, women in 1999 weighed 4.8 kg more than their counterparts in 1980, men 3.6 kg more.

Last year the Centres for Disease Control and Prevention suggested that overweight and obesity was causing 400,000 premature deaths annually in the US alone putting it second only to smoking as a preventable cause of death. The CDC later revised the figure down to 325,000 but this still dwarfed alcohol’s impact which killed 85,000 (here). The CDC’s Katherine Flegal and team subsequently changed this story to claim obesity killed ‘only’ 112,000 (here) but they made new quite astonishing claims as well:

• Underweight individuals (BMI of less than 18.5) had a higher risk of death with nearly 34,000 more deaths than expected. Most excess deaths among the underweight occurred in people age 70 or older. Among the obese, the increased risk of death was most pronounced among people younger than 70.
• Being overweight (BMI of 25-29.9) was not associated with excess mortality. Indeed the study found 87,000 fewer deaths than expected were associated with being overweight.
  

Thus being overweight is apparently not as bad for your health as doctors tell us. If you deduct from the 112,000 who die because of obesity, the 87,000 who don’t die because they are overweight, the net impact of being overweight or obese is to kill ‘only’ an extra 26,000 net.

The Harvard Medical School subsequently convened a Symposium on Overweight, Obesity and Mortality (here) which disputed the revised CDC findings by claiming that those underweight tended to be smokers, with serious wasting diseases (like emphysema) that might make them thin. Flegal counterpunched by claiming that their study had normalised for biases due to smoking or illness.

Other studies broadly support the Flegal CDC findings. Magee in the Annals of Epidemiology (here) surveys 26 studies and found, even after controlling for smoking, that being overweight was linked to a small drop in the risk of death compared to normal weight.

A recent NewScientist article (subscription only) raises qualifications to these claims:

• To be clear, being obese is very dangerous to your health – only cigarette smoking causes more premature deaths in the Western world.
• It might also be that the BMI measure is inappropriate as a measure of weight. Muscle is heavier than fat so many fit people are overweight using this definition. Similarly you may have normal weight, using the BMI definition, but have a spare tire of fat around your abdomen that exposes you to the same risk as the obese.
• Finally, observing that lower mortality is associated with high weight by itself may signify little. A causal mechanism showing how being overweight makes you healthy needs to be established. There are biological reasons for believing being fat is bad for you even if evidence does not support this.

It might also be the case that other costs are associated with being overweight. People who are overweight may be experiencing reduced quality of life because of their need to combat hypertension and high cholesterol. It might also be partly a measurement issue associated with using BMIs. Muscle weighs more than fat so high BMIs capture healthy people with lots of ‘lean mass’ as well as those with too much fat. Finally, becoming thin may have costs. Sorensen et al show here that obese people who do intend to lose weight and succeed are twice as likely to die as those who did not have this intention and do not lose weight. This might be associated with loss of lean mass. It might also be that loosing weight through exercise or by restricting carbohydrate intake preserves lean mass and enables healthy weight reduction as claimed by obesity specialists such as David Ludwig (here).

From a public health viewpoint these issues need to be sorted out. It would be foolish to pursue public health campaigns aimed at reducing the incidence of overweight people in the population if this worsens community’s health.