The Obesity Paradox

Being fat is bad except when it’s good. It’s called “the obesity paradox.” (No, that isn’t a mis-spelling for “two physicians who treat fat people.”) The adverse health effects of obesity are well established, but there are exceptions. Obesity appears to confer an advantage in certain subgroups with conditions like heart disease and diabetes.

In the News

Casual consumers of some recent media reports might interpret them as an excuse to stop trying to lose excess weight, especially if they are diabetic. Others might think we have been lied to about the dangers of the obesity epidemic. The reality is more complicated.

An article in the NY Times asks:

Obesity is the primary risk factor for Type 2 diabetes, yet sizable numbers of normal-weight people also develop the disease. Why?

The question is a bit misguided and the answer is simple. Obesity is the primary risk factor but that doesn’t mean it is the only risk factor or the cause of the disease, and non-obese patients who develop diabetes obviously have other risk factors. We know Type 2 diabetes is a multifactorial disease involving interactions between genetic, environmental, and lifestyle factors.

The article highlights recent research showing that:

Diabetes patients of normal weight are twice as likely to die as those who are overweight or obese.

And not just in diabetes:

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.

That’s true. Overweight and obese patients undergoing dialysis have better outcomes. Overweight people have better outcomes from heart failure and peripheral artery disease. After cardiac revascularization surgery, obese patients have similar or lower mortality rates compared to non-obese patients.

A simple interpretation of these findings might be that obesity must not really cause diabetes, that being obese is healthier than being non-obese, and that the concern about an “obesity epidemic” is misplaced. As usual, simple is wrong; reality is more complex.

Obesity and Diabetes

If you are overweight, the most effective thing you can do to prevent diabetes is to lose weight. The Diabetes Prevention Program study of patients with prediabetes showed that weight loss was the primary predictor of risk for developing the disease. Weight loss and exercise lowered the risk by 58%. Losing 5-10% of body weight significantly reduces blood sugars and allows many type 2 diabetic patients to get off their insulin and medications. The ADA recommends weight loss both to improve diabetic control and to reduce the risk of complications.

If you already have diabetes, there is evidence that losing weight is beneficial. In the Cancer Prevention study, patients with diabetes who lost weight had a 25% reduction in total mortality and a 28% reduction in cardiovascular plus diabetes mortality. And the benefit persisted even if they regained weight. But patients who lost more than 70 pounds had small increases in mortality.

In the National Health Interview Survey, patients who were trying to lose weight had a 23% lower mortality rate than those who were not trying to lose weight, even if they didn’t lose weight! Actual weight loss was associated with increased mortality only if the weight loss was unintentional.

A JAMA study suggested that people who are normal weight at the time of diabetes diagnosis may not be comparable to overweight diabetics. They may be at a disadvantage due to other factors like their ratio of muscle to fat (“thin outside, fat inside,” or TOFI).  It may be a matter of fitness, not fatness.

Obesity and Cardiovascular Risk

A systematic review of 40 studies in The Lancet looked at the relative risk of total and cardiovascular mortality compared to a baseline of 1.0 for people with a “normal” BMI of 20-25. Their data show that the death rate is lower for people who are mildly to moderately overweight, but it is greater for those who are underweight or very overweight.

BMI RR for Total Mortality RR for CV Mortality
<20 1.37 1.45
25-29.9 0.87 0.88
30-35 0.93 0.97
>35 1.1 1.8

Several hypotheses have been proposed to explain these findings. For instance, one study suggested that obese patients have an advantage because they are treated more aggressively.  The increased risk in the underweight category is not surprising: some of those underweight people are malnourished, cachectic from cancer, or are in poor health for other reasons.

Morbidity: Death Isn’t the Only Consideration

If obesity doesn’t kill you it could still hurt you in lots of other ways and make your life less pleasant. In addition to the obvious practical, social, and aesthetic disadvantages, there are a number of serious health consequences short of death. The Wikipedia article on obesity-associated morbidity  provides a handy list of all the health problems associated with obesity. It’s a long list, divided into 10 categories from cardiology to urology. It includes gallstones, infertility, congenital defects, gastroesophageal reflux disease, stroke, carpal tunnel syndrome, multiple sclerosis, obstructive sleep apnea, erectile dysfunction, and many others.


There really is no obesity “paradox” — there is just a complicated situation. Obesity predisposes to developing a number of health problems. Once those problems have developed, in some cases patients who are overweight have a survival advantage over patients who are not. That advantage diminishes as weight increases and eventually turns into a disadvantage when the weight gets high enough. Being underweight is also a health risk. Fitness may be as important as fatness. We don’t yet understand what all this means: we have hypotheses, but so far they are not supported by credible studies.

The Bottom Line

Obesity is a health hazard but it should not be over-simplified or categorically demonized. It is a complex issue with exceptions to the general rule. Science supports trying to control weight in general but it doesn’t support forcing every individual into the same mold of an ideal BMI range of 20-25.



Posted in: Nutrition

Leave a Comment (85) ↓