Understanding the Obesity Paradox: Why being overweight may be less concerning than being unfit


It’s one of the paradoxes of the health community: slightly overweight people can be healthier than those with trim bodies. And being skinny can be just as dangerous, or worse, than being fat.

How could that be?

Harvard Medical School estimates that more than 300 million people worldwide are obese (defined as having a body mass index [BMI] >/= 30). And this matters because being obese presents a risk to your health and ultimately, an increased likelihood of dying early.

Except it’s not as simple as that. Take, for example, a study by the Cooper Institute, a nonprofit organization in Dallas, that looked at body composition and fitness levels in 22,000 men, ages 30 to 83. Over the course of the eight study years, 428 participants died. What’s interesting is that those who were overweight and fit were two times less likely to have died than those who were lean and fit. There was also no significant difference in the overall death rates (the all-cause mortality) between the overweight fit men and the lean fit men.

So perhaps the pursuit of health should be directed away from a singular focus on body composition toward fitness over fatness.

Mortality is a U-shaped curve

What came out of reports a few years ago was something called the obesity paradox or sometimes reverse epidemiology. What the data show are that those who are either at the low side of the weight curve or the very high side are at increased risk of death. Those who are slightly overweight have a 6 percent lower risk of death than their ‘normal’ weight peers. These factors were shown to be particularly important in certain groups of people, such as very elderly individuals or those with certain chronic diseases. It further postulates that normal to low body mass index or normal values of cholesterol may be detrimental and associated with higher mortality in asymptomatic people.


A 2004 study published in the Journal of the American Medical Association found very similar results to the Cooper Institute data, determining that fitness, rather than body weight alone, was the stronger predictor of health. An earlier study of 2014 men followed over a 16-year period found that physical fitness appears to be a predictor of mortality from cardiovascular disease in healthy men and “a high level of fitness was also associated with lower mortality from any cause”. These results are further corroborated by Another study that tested the prognostic capacity of exercise capacity by following 6,213 men who were referred for exercise testing, concluded that exercise capacity was the most powerful predictor of cardiovascular disease (CVD), even among other risk factors for CVD.

A study published in the New England Journal of Medicine conducted by Harvard researchers looking at a longitudinal study of nurses that included data on both weight and activity level also found that activity level attenuated some of the effects of being overweight. But those data also suggested that the nurses who were overweight or obese tended to have a higher mortality rate; however, this particular study wasn’t designed to capture robust fitness measures or body composition, simply self-reports from the study participants of activity level and their weight.

Research published in Progress in Cardiovascular Disease reported:

Overweight and obese-fit individuals had similar mortality risks as normal weight-fit individuals. Furthermore, the obesity paradox may not influence fit individuals. Researchers, clinicians, and public health officials should focus on physical activity and fitness-based interventions rather than weight-loss driven approaches to reduce mortality risk.

BMI has tended to be a rather poor measure of health or relative fitness, based on its agnosticism of muscle content and other physiological indices. While it can serve as a quick heuristic as a short-hand measure for the average of a population, it has almost no measurement precision for individual cases where it matters most. It’s more like weighing your vehicle to estimate that it has enough fuel to cover a set distance  – there are so many other factors that could impact the result (e.g., other additional passengers, insufficient coolant level, low tire pressure). The issue is that incorporating BMI, and not more accurate body composition analyses, into research studies on weight, fitness, and longevity adds to the error rate of the results due to the imprecision of the BMI formula at measuring physiological indices.


Weight loss itself appears to not offer the same effects as improving fitness

Another study found that those who are overweight or obese in middle age have a higher risk of hospitalization and mortality from coronary heart disease, cardiovascular disease, and diabetes compared to those who are normal weight. Keep in mind that this study did not consider activity level, so weight is considered in isolation without the moderating variable of fitness.

Additionally, the NEJM reported on a long-term analysis of the effects of weight loss on cardiovascular disease in those with type 2 diabetes. They concluded, “An intensive lifestyle intervention focusing on weight loss did not reduce the rate of cardiovascular events in overweight or obese adults with type 2 diabetes.

Instead of a pure focus on weight, an approach that includes–and favors–fitness provides the greatest opportunity to increase survival.

Ben Locwin, Ph.D., M.B.A., M.S. is a contributor to the Genetic Literacy Project and is an author of a wide variety of scientific articles for books and magazines. He is also an expert contact for the American Association for the Advancement of Science (AAAS) and a consultant for a many industries including biological sciences, pharmaceutical, psychological, and academic. Follow him at @BenLocwin.

  • marcbrazeau

    Other confounders:
    People who wasted away right before dying due to illness correlates low BMI with mortality but get the causality is going the other way.

    Obese people who have a genetic disregulation and don’t have insulin resistance or metabolic syndrome can be metabolically healthy without any extra exercise – other than the exercise of moving around while obese, which is not to be under-rated.

  • WeGotta

    So, scientists are still learning about the complicated human body; a very small slice of this thing we call the natural world.
    So, experts were wrong in the past and scientific consensus has shifted as new data emerged.

    So what can we really say, scientifically, about anything so complicated then? That “it is thus” or can we only really say that “it seems like it is thus”?

    Seems like science is just a way of understanding the world (one of many) and “consensus” has changed and will always change as new information is learned.

    • Jim Gordon

      You need to get out more…

    • Roy Williams

      In the case of the issue being discussed here, this is not a case of the “experts were wrong in the past…” The article title is: “The obesity paradox…”. What we have here are some observations that do not fit the existing model. That does not mean that either the observations or the model is wrong. The most probable resolution to the paradox is that additional observations will explain the paradox, and show how the epidemiological observations reported in this essay are not in conflict with the well-understood link between obesity and poor health outcomes.

      You asked: ” So what can we really say, scientifically, about anything so complicated then?” In fact we know a staggering amount of information about “the complicated human body”. Just one example of hundreds: throughout history, until very recently, cystic fibrosis was a disease that was typically a death sentence in the first few years of life. We now understand the genetic defects that cause the disease, we know what is happening in the CF lung that is wrong, and we have developed treatments that allow most individuals with CF to live a somewhat normal life into their 30’s and 40’s.
      The fact that there are now millions of people in the US today who are still alive many years after being treated for cancer is testimony to the extensive knowledge about basic cell biology that has been acquired over the past 50 years.
      Science is not a “right or wrong” proposition. Scientific knowledge about “the human body” expands at the rate of over a thousand pages of scientific research reports daily, from all over the world. New data permits a refinement of existing models and an extension of knowledge to that which was not previously included in existing models. The data reported in this article is just one small step in the continuing effort to understand the many components of health and longevity.
      Have you ever tried to assemble a puzzle without a picture of the completed puzzle to guide you? That is very much like the scientific enterprise.
      You state that science is only one of many ways to understand the world. Scientific research has given us our modern world of instant communication, rapid global travel, long and healthy lives through an understanding of nutrition, sanitation, and disease, and thousands of products that make our lives easier, safer, more productive than those of our ancestors 200 years ago. There may be other ways to “understand” the world, but you and I are both fortunate that so many tens of thousands of dedicated research scientists and industrial engineers chose to understand the world through science – otherwise very, very few of us would be here today, and if were here, we would be constantly worried about staying warm at night, not being eaten by predators, finding our next meal, and dying by age 20 from accident, disease, or predators.
      * So, you suggested that “experts were wrong”, that “science” changes, and you conclude that “science is just a way of understanding the world…and will always change..”. As I have tried to explain, that is not a correct characterization of how our knowledge of the world expands over time. The fact that we continually learn more does not mean that what we now understand is wrong or not useful. Yes, we can and do make specific statements “it is thus” about complex biological processes, and we are able to build on those statements to develop new drugs, identify toxic materials, make replacement body parts, understand and in some cases cure previously fatal diseases, and develop new research methods that let us continually make a better, healthier life for everyone.

      • WeGotta

        All that you say is true. I don’t deny the power of science.
        I do deny claims that it’s “the only important thing” or that “all technology is good” or that “we know all about [insert anything here]”.

        Things that are true today may be false tomorrow. Our detailed and expanding knowledge about the human body today will no doubt look rather primitive in 1000 years wouldn’t you say?
        So it would be wise to be humble and always keep an open mind. When you think you know it all you stop asking questions.

        “Scientific research has given us our modern world of instant communication, rapid global travel, long and healthy lives through an understanding of nutrition, sanitation, and disease, and thousands of products that make our lives easier, safer, more productive than those of our ancestors 200 years ago.”

        Definitely true.
        But, where would we be without clean air, clean water, and fertile soil, natural diversity and human cooperation?
        Where would you be if you were not loved?

        If I made a list of the most important things in life I would but those sorts of things at the tippity top, wouldn’t you say?
        Science would be utterly useless if we didn’t have social cohesion, civility, peace, understanding, cooperation, imagination, and an open mind.

        Science is a tool that can be used to help us or destroy us. Its those other things that determine which it will be.

        • Roy Williams

          You state your perspective well.
          I don’t have an issue with your priorities. However, I see science as an integral part of each of those, not as a separate discussion. We can’t have a peaceful world, free of hunger and civilization-killing diseases, if we don’t have science to tell us that the levels of certain chemicals (like carbon dioxide) in the atmosphere are increasing, we would not know that we need to change what we are doing, or we will have severe disruption of our food supply. Without science, we don’t know that water is or is not contaminated with toxic chemicals or dangerous pathogens. With that knowledge, we are encouraged to seek global participation in solutions. While there is plenty of abuse of internet-enabled communication, our world is very much more “connected” today than it was just 40 years ago. Our growing personal ties with individuals in many other countries help our societies to see each other as friend and not as unknown, suspicious foe – if you are old enough to remember the cold war era from about 1948 through the 1980’s, you know what the world was like when it was divided by the iron curtain, with essentially no communication or social exchange. [I have photos that I took in the early 1970’s when a Soviet freighter docked at the Port of Portland (Ore) – it was a really big deal.]
          For better or worse, we all live in a world that has been defined by technology. We cannot consider any part of our modern world without including a consideration of the science that can lead us forward.
          I’m not sure what you met by “always keep an open mind”, as that is what drives scientific research and inquiry forward, as this article illustrates: while there is plenty of evidence, including really strong biochemical evidence, that obesity leads to bad health, here is some data that appears to be in conflict. Scientists in many related fields will not simply say “that data is wrong”. Instead, over time, various hypothesis will be proposed and tested that try to show that either the existing model of obesity-induced poor health is sometimes wrong, or try to show that the new data is wrong, or that that data and the model are both right because of some factor not previously included in the model. That will be potentially many lines of inquiry, and it will lead to a refinement in our understanding. This is quite different from simply saying “the experts are wrong”, which is not at all open-minded – that amounts to a rejection of the true spirit of inquiry and learning, which involves saying “let’s figure out what is really going on here, and not jump to any conclusions” (such as: “the experts are wrong”).

          • WeGotta

            All completely logical and without question. Eloquent even.
            Believe me, if it were reversed and people were ignoring important science and technology I would be speaking in support of it.

            But right now what we are ignoring are the devastating consequences of moral, spiritual, and ethical poverty. .

            I tend to agree with those who feel a little cheated actually. I feel we should have progressed much further than we have actually.
            But we got stuck in hate, greed and fear causing untold waste of energy, lives, time and resources.

            How much have we wasted on war alone? Is that an “advanced” society?

            Is it “advanced” to wreck the planet? Are we wrecking it because we lack a technology? Or because we lack the humility and wisdom to ask the right questions?
            I say lets give equal consideration and support to those other things we generally consider important such as peace, justice, understanding and fairness.

  • Roy Williams

    All this data is correlative. An alternative hypothesis is that people who are “more fit” are “more fit” precisely because they are healthier. I think that marcbrazeau’s suggestion needs to be investigated further – are people who are on the low end of the BMI scale because of undetected health issues, and therefore more likely to have a shortened life span?
    One hypothesis that has been proposed is that people with a low BMI do not have as much stored nutritional reserves to deal with a serious health challenge as do people with higher BMI – up to a point. This hypothesis can be tested only by looking in detail at the health records of thousands of people over perhaps 20 -40 years.

    The essay cites a study that concludes “that those who are overweight or obese in middle age have a higher risk of hospitalization and mortality from coronary heart disease, cardiovascular disease, and diabetes compared to those who are normal weight”. – nothing new there, and this is a correlation that rises to the level of causative relationship due to substantial molecular biological evidence of a mechanism linking obesity to the observed outcomes.
    Same problem with the study of 6,000 men – do they have cardiovascular disease because they do not exercise, or can they not exercise because they have cardiovascular disease?
    These broad epidemiological studies do not provide useful information for public health policy – they should be used only to inform additional research that is focused on discovering specific causality factors all the way down to the molecular biology level, as has been done with obesity and the high correlation to numerous poor health outcomes.

  • Rob

    As America, Britain, Canada, Ireland become fatter & fatter we read more crap about how being overweight is just fine. Any M.D. anywhere, even the fat ones, will tell you slim is healthier. So, get off your arse, eat low-carbno sugar and sweat some.