Keeping the baby and throwing out the bathwater.

Abstract

The Perspective in this issue of Obesity, “Redefining Obesity: Beyond the Numbers” by Arya Sharma and Denise Campbell-Scherer (1), suggests that it is time to redefine how we diagnose the disease of obesity. Even if it is not a totally novel idea, we fully agree that BMI should not be the only criteria for a diagnosis of obesity in the clinic. But BMI categories have utility on a population basis and for screening. Let's not throw out the baby with the bathwater. By this we mean, don't discard use of BMI altogether just because it's not completely diagnostic of the disease. The discovery of the US obesity epidemic dates to the national surveys measuring BMI conducted on a representative sample of the US population, with the NHANES III survey (1988-1994) creating shock waves through the popular media when released in 1998 (2). The percent of adults with obesity, defined as BMI 30 kg/m2 and above, had varied little from 1960 to 1980 but increased considerably between 1980 and 1994, from 13% to 21% among men and from 17% to 26% among women. This trend continued in the survey conducted from 1999 to 2000, with an increase in the prevalence of obesity to 28% in men and 34% in women (3). In the latest survey (2013-2014), women had higher rates (40.4%) than men (35.0%) (4). Using BMI as a population measure of body size and as a reasonable, even if not perfect, surrogate for body fat in populations has had merit. It has focused much-needed attention on the remarkable changes that have taken place in the US population since the early 1980s; in the 36 years since, the rates of BMI >30 kg/m2 have increased more than 2-fold in women and 2 1/2-fold in men. These population changes go a long way in explaining the dramatic increases seen in obesity comorbidities, especially type 2 diabetes. However, the utility of BMI as sole diagnostic criteria ends when one considers obesity and its consequences in given individuals. BMI only measures body size, not body fat amount or location or how that body fat may be affecting health. Sharma and Campbell-Scherer propose a simple change in the World Health Organization's definition of obesity (5) from “abnormal or excessive fat accumulation that may impair health” to “abnormal or excessive fat accumulation that impairs health.” The trend in guidelines is to not use BMI as the sole diagnostic criteria for obesity. This is certainly true for the 2013 Obesity Guidelines (6), in which BMI is positioned as a screening tool, not a diagnostic tool. The recent American Association of Clinical Endocrinologists Guidelines (7) are also “complications-centric, rather than BMI-centric.” The solution is to use the definition “abnormal or excessive fat accumulation that impairs health” in the medical record and BMI criteria only for screening or following population trends. Healthcare providers are used to assessing multiple factors (symptoms, physical findings, laboratory findings, or imaging findings) and making a diagnosis based on multiple factors. BMI does not have to be the sole criterion for the diagnosis of obesity. And we agree with Sharma and Campbell-Scherer that public health stakeholders in association with groups involved in obesity research and advocacy need to join forces to generate guidelines on how to better characterize and prioritize for treatment those patients who have clear health impairments related to the presence of abnormal body fat.