Researchers debunk 'obesity paradox' for people with diabetes

Researchers debunk 'obesity paradox' for people with diabetes
A detailed analysis of thousands of people with diabetes found no support for the "obesity paradox," Harvard researchers reported. (Susana Gonzalez / Bloomberg)

For people with Type 2 diabetes who had hoped that their love handles might serve some purpose by reducing their risk of premature death, Harvard researchers have some bad news: The "obesity paradox" does not exist.

"We found no evidence of lower mortality among patients with diabetes who were overweight or obese at diagnosis, as compared with their normal-weight counterparts, or of an obesity paradox," the research team reported in a study that appears in Thursday's edition of the New England Journal of Medicine.


What they did find will be pretty disheartening to those with a body mass index outside the 22.5-to-24.9 range, which is considered the higher end of "normal" weight:

Compared with people in this index group, people with a BMI between 30 and 34.99 (considered "obese") were 24% more likely to die during the course of the study, and people with a BMI above 35 (considered "severely obese") were 33% more likely to die. People whose BMIs in the 25-to-29.99 range who were merely "overweight" also had higher rates of premature death, though the difference wasn't large enough to be statistically significant.

The data also offer sobering news for diabetes patients on the lower end of the normal range. Compared with people in the index group, those with a BMI between 18.5 and 22.4 were 29% more likely to die of any cause during the study period.

These findings are in line with some previous studies, but they contradict several other reports that identified a surprising relationship that came to be known as the obesity paradox. In these studies, people who were obese were less likely to die than people with a "healthy" weight. Examples of the obesity paradox have been seen among people with diabetes, people with various kinds of heart disease and people on dialysis.

Considering that excess weight is widely recognized by medical experts as a risk factor for early death due to ailments like cardiovascular disease and cancer, the team from the Harvard School of Public Health, Harvard Medical School, and Brigham and Women's Hospital decided to take a detailed look at the issue.

They suspected that previous studies failed to take smoking into account, which is a problem because smokers are more likely to have lower BMIs and to die prematurely. They also figured the results could be skewed because people who are frail or suffer from chronic diseases can also have lower weight and a higher risk of premature death.

So they turned to two large long-term studies that tracked thousands of people in copious detail: The Nurses Health Study and the Health Professionals Follow-up Study. Together, these studies provided the researchers with data on 11,427 people who were diagnosed with diabetes before 2010. The study subjects were tracked for an average of 15.8 years. During that time, 3,083 of the people died.

The researchers found that the lowest risk of death from any cause was among people with a BMI between 22.5 and 24.9. After adjusting for factors like smoking history, alcohol consumption, exercise habits, diet quality and demographic factors like age, race and marital status, they discovered that the severely obese had the highest risk of premature death, followed by those on the low end of normal weight and those who were obese. They described the results as a "J-shaped" association.

Among the subset of diabetes patients who had never smoked, the increased risk of death was large enough to be statistically significant only in people who were obese (36% higher odds) or severely obese (56% higher odds).

But among smokers, the greatest risk of premature death was seen among people with BMIs in the low-health range (32% higher odds).

More than 90% of the people included in the analysis were white, and the researchers cautioned that their results might not apply to people of "other racial and ethnic groups."

Why did these results contradict studies that found an obesity paradox? The researchers raised several possibilities.

One problem may be that the earlier studies followed volunteers for a shorter period of time and recorded fewer deaths, making the true signal harder to find. Also, some of the studies did not take account of smoking behavior or undiagnosed chronic diseases, which could also skew the results. In one case, a study that reported an obesity paradox separated study subjects into only two BMI groups – those above and below 25.


"The maintenance of a healthy body weight should remain the cornerstone of diabetes management," they concluded.

If you're interested in the latest medical research, you like the things I write about. Follow me on Twitter and "like" Los Angeles Times Science & Health on Facebook.