Diabetes and the “Obesity Paradox”

We hear constant warnings about how fatness leads to diabetes. But a wide range of studies show heavy people live longer and do better with diabetes and heart disease. This is called the “obesity paradox.” What does it mean?


In an article in The New York Times, science writer Harriet Brown reports on the work of Dr. Mercedes Carnethon at Northwestern University. Dr. Carnethon has found that thinner people with diabetes have twice the death rate as those who are overweight or obese.

Carnethon’s findings are typical. As Brown writes,

In study after study, overweight and moderately obese patients with certain chronic diseases often live longer and fare better than normal-weight patients with the same ailments.

Protective effects of weight have been found in heart failure, people undergoing dialysis, and those with coronary artery disease, stroke, kidney disease, and high blood pressure. Not only that — researcher Katherine Flegal PhD, has shown that overweight Americans have less risk of dying, on average, than normal-weight people. Mildly obese people have about the same risk of death as normal-weight people and a much lower risk of death than underweight people.

Studies of people over 70 years of age in Australia, Canada, and the US have each found that overweight (but not obese) people had the lowest risk of dying from any cause.

I have to ask. How can they keep calling these findings “paradox?” In how many cases can a theory (like “obesity kills”) fail before its failure is no longer a “paradox?” When do we start to think that maybe the theory is wrong?

There are other explanations of what is happening here. Maybe it’s that being sick puts a strain on the body, and some extra weight helps the body meet those demands. Without it, bodies may possibly become “malnourished” even at “normal” weight. That’s how fatness might be a risk factor for illness but still help you once you get the illness.

Another study by Dr. Carnethon seems to back this up. She found that at diabetes diagnosis, only about 20% of people with Type 2 were normal weight. But people among those 20% much more likely to die during 15 years of follow-up than the 80% who were overweight and obese.

Another theory is that thin people who get heart disease or diabetes may have worse genes. Otherwise they wouldn’t get sick in the first place. Some, like cardiologist Carl Lavie, MD, even say that heart disease in thin people may be a different illness than heart disease in heavier people.

A different illness? It sounds like these guys are desperate to protect their theory from the data. But other scientists seem more open-minded.

Dr. Neil Ruderman, an endocrinologist at Boston University School of Medicine, believes that thin people can still have a lot of insulin resistance and bad cholesterol. They may not have much fat, but what they have is around their middle. This abdominal fat is likely to creep into the liver, pancreas, and heart where it can shorten life.

Ruderman calls these people “metabolically obese normal weight people.” Eek! So you be thin and still be “obese!” What chance do we have?

Here’s a possibility. Weight is acting as a marker for fitness. In other words, fat people are at risk because they tend to be out of shape. A series of studies at the Cooper Institute in Dallas showed that being fat and fit is better than being thin and out of condition. Exercise may not lead to weight loss, but it does strengthen the heart and reduce fat in the liver, where fat seems to do damage.

The explanation of the “obesity paradox” may be simple — it’s not weight that causes health problems. It’s the stress, lack of fitness, and the insulin resistance that are often seen in heavy people that make them sick.

Fatness itself may be somewhat protective. That may be one reason bodies become fat, as protection. So thinner people who share those risk factors are at the most risk of all and are most likely to die from illnesses like heart disease and diabetes.

Exceptions to this might be obstructive sleep apnea, knee, and hip problems. Those do seem related to fatness in some studies.

In fact, it seems probable that insulin resistance causes fatness more than the other way around. They’re a vicious cycle.

My point is that losing weight by getting fit might be great, especially the “getting fit” part. Losing weight by restricting calories is much less likely to help and may even hurt if you already have diabetes, heart disease, or other major illness.

Hey, it’s my birthday this week (November 9). Give me a present by going to www.reasons2live.net, reading a blog entry or two, and maybe leaving a comment.

Learn more about the health and medical experts who who provide you with the cutting-edge resources, tools, news, and more on Diabetes Self-Management.
About Our Experts >>

  • jim snell

    Happy Birthday; David.

    Excellent blog and very interesting feedback.

    The comments you point out suggest that the current
    thinking on Type 2 diabetes has some holes in it
    and could use a boost.

  • Kathy

    My boyfriend is 40, not obese, and has type 2 diabetes. He currently takes both metformin and long-acting insulin (once a day).

    My “theory”, and I”m not a doctor at all, although more informed than the average person (and I work in human subjects research), is that some of the people categorized with type 2 diabetes, specifically the less obese people, are more like type 1’s than the obese type 2’s with cell resistance. I really wonder how well my boyfriend’s pancreas is working. Rather than his cells having insulin resistance, I wonder if he his pancreas is subpar, but operating enough that he was originally diagnosed with type 2. Call people like him Type 1.5’s. lol

    Anyway, if you could pull type 2 diabetics w/working but poorly working pancreases out of the group of type 2 diabetics with normal pancreases but with cells resistant to insulin, I wonder if that would pull enough of the non-obese out of the type 2’s to make that statistical difference of fatter people living longer go away.

    Of course, I am assuming that Type 1’s have a shorter lifespan than Type 2’s. It’s an idea to consider, anyway.

    I think that thin people with diseases like heart disease are more likely to have them for genetic reasons than being obese – and perhaps having the disease for a genetic reason makes it more deadly.

    And of course, thinner people are the ones outside mowing their lawn and having heart attacks…the obese people stay inside on their sofas and watch tv.

    Anyway, it’s an interesting article and a lot to ponder!

  • Mary

    Happy Birthday! By now, it is a little belated, however, since it only comes once a year, one could have a birthday week! I have learned much from your articles that appear in Diabetes Self Management, and been muchly encouraged. Thank you for keeping on keeping on & sharing with others in a meaningful way.

  • Joe

    Pretty much forever we have associated “fit” with “lean”. But if you look at professional athletes, many have excellent cardio fitness and stamina, along with great strength, but fall into the medical definition of obesity. Other people who are very trim and look perfectly fit get winded walking across the street.

    Maybe we should start measuring fitness with things like endurance tests and oxygen levels instead of pounds and inches?

  • Pam

    I was diagnosed with type 2 about 6 years ago. At that time i was somewhat overweight, but not obese. As a kid, I was super skinny (underweight) & was almost force-fed by my family in hopes of getting me to gain weight. I had a totoal hysterectomy at 45 & went into instant menopause. I began to gain weight & also started the vicious circle of yo-yo dieting. At about 63 I got my diagnosis of diabetes. I followed the diabetes guidelines faithfully, took Metformin and managed my diabetes very well, but could NEVER rid myself of the belly fat. Skinny arms, legs, not butt….but forever stuck with the belly fat! Recently, I began to eat differently & took up high intensity interval training 4 days a week. Instead of the recommended 45g of carbs per meal, I now strive for 59g PER DAY of carbs in the form of fruits & veggies. I’ve dropped 25 lbs. and many inches of fat from my belly. I’m working on decreasing my body fat percentage rather than just losing weight & it seems to be working, at least for me. I’d say getting fit rather than losing weight is the better way to go. My a.m. glucose readings are not in the 90s instead of the 190s.

  • Mike

    Has it occurred to scientists that the BMI (Body-Mass Index) (which defines “Obese” vs. “Normal”) may be inadequate or out-of-date? I am very healthy and moderately athletic and fit, although nothing special in background nor appearance – a typical white-collar middle-aged worker. In height and weight, I almost match NFL receiver Wes Welker. (He is certainly a superior athlete to me and almost anybody else.) Yet, these same height and weight metrics score me as borderline obese. Is anyone going to call Wes Welker fat?
    If I were to be at the low end of the (BMI-) “normal” weight for my height, I would almost surely be skinny and sickly – certainly not athletic.
    I think something about the BMI fails to take into account the fact that muscle is a) denser than fat, and b) helpful to health in ways that are not being adequately represented in many studies.