Australasian Science: Australia's authority on science since 1938

The “Obesity Paradox” Paradox

By Tim Olds

Three recent studies have cast new darkness on the paradox that overweight adults are more likely to get diseases such as diabetes yet seem to live longer.

I have written before about the Obesity Paradox – the finding that overweight adults are more likely to get diseases such as diabetes yet are less likely to die from them. In fact, overweight people appear to live longer. Now three recent studies have cast new darkness on the Obesity Paradox.

But before we deal with these studies, let’s review what we already know. The relationship between fatness (usually measured by Body Mass Index, or BMI, which is your weight in kilograms divided by your height in metres squared) and your risk of dying from anything at a given age is U-shaped. People who are very thin or very fat are more likely to shuffle off the mortal coil.

The issue is where the bottom of the U occurs. Traditionally it was thought to fall in the “normal weight” range (a BMI of 20–25). Recent analyses suggest it falls in the middle of the “overweight” range (BMI 25–30). This is a difference of about 10 kg for an average male, so we’re talking about quite a bit of lead in the saddlebags.

Now there have been three theories to explain the Obesity Paradox. The first is the Healthy Survivor Theory. To understand this theory you need to know how these epidemiological studies are done. Researchers get a group of people, measure their fatness and follow them until they die. Typically, researchers start following people in their 50s or 60s.

But what if the unhealthy fat people had already died before the study started? Then the study would start with only healthy fat people and a mix of healthy and unhealthy lean people, so it wouldn’t be surprising to see the fat people last longer.

A second theory is the Differential Medical Care Theory. This theory argues that fat people get better medical care. A fat person who walks into a GP’s office with high blood pressure is likely to walk out with a bundle of life-saving drugs. A lean person may be told to get a bit more exercise and cut down on the salt.

The third theory is the Real Benefits Theory. This theory argues that fatness confers real, life-saving benefits. Perhaps it protects against falls or nutritional deficits and frailty in old age.

Now let’s look at the three studies that have muddied the muddy waters.

A study of 2.3 million Jewish conscripts over a 43-year period found that there was a strong relationship between their BMI at age 16–19 and their risk of death from all causes. The fattest 5% were more than five times as likely to die as the leanest 5%. It’s important to note that this study refers to premature death – the oldest participants who died were just over 60 years old, and only a very small percentage of the cohort had died (barely 1% of those in the fattest 5% had departed this vale of tears). This study supports the Healthy Survivor Theory: the unhealthy fat people may indeed die young.

A second study looked at the relationship between BMI and risk of death in three large Danish cohorts from 1976–78, 1991–94 and 2003–13. The researchers found that the BMI associated with the lowest all-cause mortality (i.e. the bottom of the U) has been increasing over time. It was 23.7 in the 1970s, 24.6 in the 1990s and 27.0 in the 2000s. One previous American study found similar trends. These studies seem to be telling us that the world has changed so that fatter people now survive longer, so it may support the Differential Medical Care Theory.

The third study was the biggest: a megastudy of 10.6 million people from Asia, Australasia, Europe and North America. The headline finding was that mortality was lowest at normal weight (BMI 18.5–25), and was 11% higher in overweight adults and 171% higher in the obese group. This study seems to go directly counter to the Obesity Paradox.

But there’s a catch. The authors excluded all participants who had known disease or had ever smoked. This reduced the initial sample from over 10 million to about 4 million, so 60% of all participants were excluded.

These adjustments are likely to bias findings. In particular, people who have never smoked are likely to lead healthier lifestyles, be leaner, and may live longer not because they are lean but because of their healthy lifestyles.

When the analysis is redone adjusting only for known disease, overweight people have a 5% lower risk of dying at any given age. This is much more in line with all the other studies.

So these three studies have brought a bit of darkness to the light. One thing that these studies agree on is that there is not all that much difference – in terms of risk of dying – between a BMI in the normal range and a BMI in the low overweight range. They also agree that the BMI at the bottom of the U increases with age.

So if you’re my age, and carrying a few extra kilos, you shouldn’t lose too much sleep over it. If you’re young it might be a good idea to get to the gym.


Professor Tim Olds leads the Health and Use of Time Group at the Sansom Institute for Health Research, University of South Australia.