Australasian Science: Australia's authority on science since 1938

The Obesity Paradox

Credit: iStockphoto

Credit: iStockphoto

By Tim Olds

In the past 10 years there has been no increase in the fatness of kids, either in Australia or in many developed countries. At the other end of life, fatter adults are living longer than lean adults. What can be going on?

In August 2006 I asked in this magazine whether the increase in childhood obesity was due to gluttony (kids eating much more now than they used to) or sloth (kids being less active now than they used to be). Contrary to most of my colleagues, I argued that the answer was sloth. There appeared to have been no increase – in fact there was a decrease – in energy intake in children since the end of World War II. I still think I was right on that one, but in the meantime the obesity battlefront has changed a lot.

End of the Epidemic?

In 2007 there were dire predictions that we were losing the obesity war. Experts warned about the apparently exponential increase in the prevalence of overweight and obesity in children, both in Australia and around the world. One British study, the inaptly named Foresight Report, suggested that by 2020, 42% of 2–11-year-old boys and 48% of girls of the same age would be overweight or obese, about double the figure at the time. By 2050, the report said, 90% of kids would be overweight or obese. Another researcher told us that more than half of all Australian adolescents would be overweight or obese by 2030.

Researchers in the UK, Australia and the US were predicting that this would be the “first generation to die before their parents”. They meant that the life expectancy of that generation would be lower than that of their parents, of course, but the media conjured up images of bereaved parents at the funerals of their deceased, superobese children.

They were wrong: prevalence rates have clearly plateaued in Australia (and in many countries around the world) since the late 1990s or early 2000s, as my colleagues and I reported in 2010 and 2011 (International Journal of Obesity, 34(1): 57–66; International Journal of Pediatric Obesity, 6(5–6): 342–60). There are signs, too, that trends are beginning to flatten in adults.

Embarrassed British experts revised their 2020 projections down from 42% to 30% for boys and from 48% to 27% for girls, but even these figures now appear to be decidedly on the high side. There is almost no meaningful evidence that life expectancy will decrease. It has been going up by about 2–3 years every decade, and the trend shows every sign of continuing.

The picture is not all sunshine, however. The plateau is more marked in some demographics than others: there are still increases among the usual suspects – low income groups, migrants, and rural and remote areas. There may also be increases in fatness without increases in weight for height. For example, kids’ waist girths are getting bigger at the same weight for height. And the overall prevalence – about 20–25% of kids in Australia are overweight or obese – is still very high.

Why might this be happening? One possibility is that it is a sampling issue. Maybe fewer obese kids and adults are putting their hands up to be weighed and measured, fearing stigmatisation. Maybe, but even countries where almost every child is weighed (such as the UK, some urban areas in China and the US state of Arkansas) show the same flattening trends.

A second possibility is that we have reached a kind of “equilibrium saturation”. Proponents of this explanation argue that the social and built environments are so “obesogenic”, or fat-promoting – given the easy availability of energy-dense, highly palatable food and the increasing barriers to, and decreasing need for, physical activity – that every child who is at the slightest risk of becoming fat has become fat, and the others are naturally resistant to fatness.

A third possibility is that the anti-obesity messages and initiatives – promoted by just those experts who are so reluctant to admit there is a plateau in prevalence – are actually working. Every time a mother packs a healthy lunch, every time a school renovates its play areas, and every time a council builds a new cycleway, they are firing another shot in the war against obesity.

The Obesity Paradox

Although the plateau in childhood overweight and obesity should be seen as good news, something unusual is happening at the other end of life. Fatter people are more likely to get heart disease, diabetes or cancer, and are likely to die younger, right? Well that’s what we always thought, but one of the most surprising findings in epidemiology over the last few years has been the “obesity paradox”. It appears that overweight people may have a lower risk of dying than people of normal weight.

The obesity paradox was first noticed by Mercedes Carnethon and colleagues in relation to people with heart disease and diabetes (Journal of the American Medical Association 2012, 308(6): 581–90), where being overweight appeared to be protective. People who had heart disease, or diabetes, or other types of diseases and were overweight, or even obese, survived longer than those who were normal weight.

Soon, however, evidence emerged that this also applied to elderly Australians who were healthy (, and then, it seemed, even to younger people. One study followed more than 16,000 Norwegians for 9 years, noting when they died and looking at the relationship between their weight status and their death rate. Jan-Magnus Kvamme and colleagues found, to their surprise, that the risk of death from any cause was lower in overweight and moderately obese people than in people of normal body weight (Journal of Epidemiology and Community Health, doi:10.1136/jech.2010.123232).

A volley of similar studies have appeared around the world. A study of 50,994 US adults by Anthony Jerant and Peter Franks found that, after 6 years of follow-up, overweight individuals were 20–25% less likely to die than those of normal weight (Journal of the American Board of Family Medicine 2012, 25(4): 422–31). Even at ages 18–60, overweight Hispanics were less likely to die than their leaner peers, and by age 70-plus even grossly obese (BMI ≥ 35) Hispanics were 25% less likely to die, according to Kevin Fontaine (International Journal of Obesity 2011, 36(8): 1121–26).

In a study of Canadians followed for 12 years, Heather Orpana and colleagues found that overweight men, and both overweight and obese women, had better survival prospects than those of normal weight (Obesity 2011, 18(1): 214–18). A 17-year follow-up of 140,000 Europeans by Xin Song and colleagues found that people with a BMI of 28 (in the middle of the overweight band) were at less risk than those with a BMI of 20 (at the bottom of the normal weight band) (European Journal of Clinical Nutrition 2011, 66(2): 156–65).

These results were met with a great deal of skepticism by researchers. How could being fatter actually protect us against dying when being fatter makes it more likely that we will get diseases that can kill us, such as various types of cancer, heart disease and diabetes? Could it be that leaner people were suffering from wasting diseases that had not been diagnosed, making it seem as if lean people were more likely to die?

It seemed a neat solution, but when people with any sign of disease, or with risk factors for these diseases (such as smoking) were eliminated, the same results were found. Even when deaths in the first few years of follow-up (which might be expected if some unknown disease were lurking) were not considered, the results were the same.

A 2005 study from the US Centers for Disease Control found that it hasn’t always been like this. Following up data from a series of large American surveys, the National Health and Nutrition Examination Surveys, Katherine Flegal and her team found that, among those followed up from 1972, both the overweight and obese had a greater risk of dying than those of normal weight (Journal of the American Medical Association, 293(15):1861–7). Following up from the 1978 survey, the overweight survived longer but the obese had shorter lives. In the follow-up of the 1993 survey, both the overweight and obese lived longer.

So what changed between 1972 and 1993? Perhaps fatter people now get better medical care. If a lean person walks into a doctor’s surgery with high blood pressure, the doctor will probably advise them to get a bit of exercise and avoid stress. If an obese person walks in, the doctor is more likely to put them on drugs to lower blood pressure, and more drugs to control cholesterol for good measure. She will also be more likely to check their blood sugar and perhaps prescribe drugs for that. The fat person walks out of the surgery with a whole basket of life-prolonging drugs while the lean person is told to chill out. In other words, fat people may live longer because they get better care. There is some evidence to support this idea: health costs are considerably higher for obese people than for lean people.

Recent data also suggest that how much fat we have may not be as important as where that fat is located. A study just released (Francisco Lopez-Jiminez, presentation at ESC Congress 2012) followed up American adults for 14 years. Who was most at risk? Normal-weight people with high waist-to-hip ratios – that is, people who accumulated fat around the middle. Fat elsewhere in the body appears to actually protect us against heart disease.

Fighting the Obesity Wars

So where do these new and surprising developments leave us in the war against obesity? The first important message is that they do not mean that the war is won, or that it is a war not worth fighting. The commonsense approaches we have been using – social and environmental engineering encouraging us to eat well and get a bit of exercise – appear to be working. Medical awareness appears to be better at treating the consequences of overweight and obesity in adults. And while overweight people may not die younger, they certainly lead less healthy lives.

The second message is that we’re not going to hell in a handbasket. Fatness is not increasing exponentially, we are not sitting on a “fat bomb”, and children will not die younger than their parents. There is a way to go, but we seem to be on the right track.

The fundamental messages have not changed in a century: eat good food and get some exercise. Both food and exercise have benefits that run way beyond controlling fatness. The work of US public health expert Steven Blair, for example, has shown that people who are fat but fit are healthier than those who are lean and unfit. In a study of 116,564 women by Frank Hu and colleagues, those who were overweight but active had a 20% increase in risk of death compared with those who were lean and active. But those who were lean and inactive had a 60% greater risk (New England Journal of Medicine 2004, 3512: 694–703). In a very large sample of US men in a study by Ming Wei and colleagues, those who were fit but obese had a 20% increase in risk compared with those who were fit and lean, while those who were unfit and lean had a 120% increase in risk (Journal of the American Medical Association 1999, 282(16): 1547–53).

Studies relating physical activity to mortality suggest that every hour we spend exercising extends our lives by about one -and-a-half hours. That sounds like a pretty good deal to me.

Professor Tim Olds is group leader of the University of South Australia’s Health and Use of Time program.