Portal de Periódicos da CAPES

Reporting Weight Loss: Is Simple Better?

2010; Wiley; Volume: 18; Issue: 2 Linguagem: Inglês

10.1038/oby.2009.289

1930-739X

Arya M. Sharma, Shahzeer Karmali, Daniel W. Birch,

Pharmacology and Obesity Treatment

TO THE EDITOR: In their recent article titled "Is it Time to Change the Way We Report and Discuss Weight Loss?" in Obesity, Bray and colleagues recommend the use of percentage of excess body weight loss (%EBWL) to report and discuss weight loss (1). Interestingly, in an article recently published in Surgery of Obesity and Related Diseases (2), we argue against the use of both excess weight loss (EWL) and %EBWL—rather, we recommend the use of percentage of initial weight loss together with baseline weights in all reports and discussions of weight loss. Although we agree with the need for a uniform measure of reporting weight loss in trials and in discussions with patients across disciplines, we disagree with both of the two prime arguments presented by the authors in favor of using %EBWL. The first objection relates to the proposed definition of "excess" weight, the second objection relates to the putative value of communicating this idea to patients. Any definition of "excess" requires a definition of "normal". Bray and colleagues recommend using 25 kg/m2 as the upper limit of normal in line with current anthropometric definitions of obesity. It is, however, important to note that this concept of "normal" is not based on results from intervention studies. Thus, there is currently no evidence to suggest that the maximum benefit of intentional weight loss can only be achieved when patients reduce their body weight to a BMI ≤25 kg/m2. In fact, in both surgical and nonsurgical studies, significant improvements in morbidity and mortality (in surgical studies) can be achieved with far more modest weight loss (in the 5–30% range) (3). Not only is achieving a BMI of 25 kg/m2 virtually impossible and unsustainable for the vast majority of patients (4), there is also no evidence that this is indeed necessary for optimal outcomes. Till such data becomes available, defining a BMI of 25 kg/m2 as a target for weight loss remains arbitrary. Indeed, we feel that presenting 25 kg/m2 simply serves to further perpetuate the feeling of "hopelessness" and low self-esteem already present in this vulnerable population (5). The other argument challenges the notion that %EWL presents patients with a more optimistic view of their achievements. Although we concede that for a patient with BMI 30 kg/m2, who loses 10% of initial weight, 60%EBWL sounds more impressive, it does not take much to realize that a BMI of 25 kg/m2 is still out of range. In fact, to explain to patients what exactly the term "excess" weight means, one cannot but enter into a discussion that reminds patients exactly how far away from "normal" they actually are and how hopeless any chance of ever becoming "normal" really is. In contrast, a discussion that focuses on the substantial health benefits that can be achieved with even modest (5–10%) but realistically sustainable weight loss without actually coming anywhere close to a BMI of 25 kg/m2 is far more empowering than harping on an unrealistic and, as far as we know unnecessary, "ideal" target weight. The concept of describing weight loss as a simple percentage of initial weight is a calculation that almost anyone can immediately follow. In contrast, as outlined in their article, %EBWL requires more complex computation, especially, if as suggested by Bray et al., one chooses to use a different "normal" BMI for different populations (1). Imagine the confusion in the literature, when trying to compare %EBWL from studies in Asians using a BMI of 23 kg/m2 as normal to studies using a BMI of 25 kg/m2 in whites —an issue that would be hopelessly confounded in multiethnic populations. Finally, the use of %EBWL needlessly deviates from reporting in other fields of medicine. Thus, we do not generally report percentage changes of "excess" blood pressure, "excess" blood glucose, or "excess" blood cholesterol levels. As a rule, changes in these, like in virtually all other clinical parameters, are presented as absolute or percentage changes from initial values. We see no reason why weight should be treated any differently. Indeed, as the goal of weight loss interventions is to improve health and not just to lower weight, we believe that the uses of a clinical staging system for obesity, as recently recommended (6), will prove far more valuable to determining outcomes for patients, health-care providers, researchers, payers, and policy makers than any reporting system that focuses on weight alone. The authors declared no conflict of interest.