For years, we have been told that overweight and obesity are associated with serious medical morbidity, including certain cancers like breast, prostate, and colon, metabolic disturbances such as diabetes and abnormal blood lipid levels, as well as heart disease, orthopedic disability, sleep apnea, and hypertension. Most, but not all, researchers have led us to believe that overweight and obesity, as defined by body mass index (BMI), can substantially shorten our life expectancy, and the worse the level of obesity, the more likely, an earlier demise. This week, though, a new study received considerable media coverage and raises questions about some of these assumptions. Can being overweight or even somewhat obese actually be protective against mortality? For most in the field of obesity, this notion certainly seems counter-intuitive, surprising, and even nonsensical, and hence it has been called the “obesity paradox.”
Katherine M. Flegal, Ph.D., from the Centers for Disease Control and Prevention (CDC), and her colleagues, writing in the Journal of the American Medical Association, (and extending their original 2005 study) conducted a rigorous, systematic review of almost 100 other studies involving a combined sample of 2.88 million people and 270,000 deaths to assess whether overweight and all levels of obesity are in fact related to increased mortality risk as most believe. Dr. Flegal et al believe that knowing "relative mortality risks" associated with different levels of weight "may help to inform decision making in the clinical setting.” They note that in the U.S., for example, almost 40% of adult men and almost 30% of adult women are considered overweight by today’s BMI standards, while 36% of adults are considered obese. Significantly, though, more than half of those in the obese category are in the grade 1 category (BMI of 30 kg/m2 to 34.9 kg/m2.) In general, mortality-weight statistics follow a U-shaped curve, with extremes of being severely below normal weight or severely above it are at highest risk. What these researchers found is that Grade 1 obesity was not associated with higher mortality and what’s more, being overweight (BMI of 25 kg/m2 to 29.9 kg/m2) was actually protective and associated with significantly lower all-cause mortality! Flegal and her colleagues, though, acknowledge that many of the studies included in their review relied on notoriously inaccurate self-reporting of height and weight that could result in misclassifications of BMI and bias their results. They also note that their study is limited in not addressing morbidity,“cause-specific mortality,” and issues of body composition (e.g. where fat is distributed on the body.)
In an Editorial accompanying the study, Drs. Steven B. Heymsfield and William T. Cefalu raise the provocative question, “...are the concerns about overweight as currently defined unfounded?” Professor of Law and author of The Obesity Myth: Why America’s Obsession with Weight is Hazardous to your Health, Paul Campos, writing an editorial in the New York Times, in response to the Flegal study, certainly thinks they are and believes “our current definition of ‘normal weight’ makes absolutely no sense.” He asserts that we are in an “absurd situation” in which we are serving “the economic interests of...the weight-loss industry and large pharmaceutical companies” and “baselessly categorizing at least 130 million Americans--and hundreds of millions in the rest of the world--as people in need of ‘treatment.’”
Heymsfield and Cefalu, though, note that the Flegal study does confirm that higher levels of obesity are associated with increased mortality, but they also explain that the situation is far more complicated when we consider lower levels of obesity and even overweight. First, BMI (the ratio of weight in kilograms divided by height in meters squared) is not a perfect measure of fat: it does not account for differences in sex, race, or age or even a person’s cardiopulmonary level of fitness, as well as other risk factors such as blood pressure, glucose and lipid levels, and waist circumference. For example, fat around our internal organs is more dangerous than subcutaneous fat so that the “apple” distribution of fat around the middle is worse than the so-called “pear" distribution. Furthermore, these authors note that physicians may be “increasingly aggressive” in treating the risk factors associated with obesity and hence may be affecting mortality statistics, particularly in these lower weight categories, such that “overweight and grade 1 obesity might lead to greater morbidity that is not captured when evaluating associations between all-cause mortality and BMI.” They also point out that when there is a chronic wasting disease, being overweight and even slightly obese can be somewhat protective in providing “needed energy reserves” and even protection against trauma. There have been, for example, earlier reports that those overweight and mildly obese patients with chronic heart failure and coronary artery disease had decreased mortality compared to those of normal weight. It is always important to consider whether weight loss is intentional or unintentional.
Bottom line: One size does not fit all. The relationship between weight and mortality is a complex one. For most people, being of normal weight is still healthiest but for some, particularly with advancing age and chronic wasting diseases, those extra pounds may not be so detrimental.