ABSTRACT
Obesity that results from chronic positive energy balance (1) has serious health consequences (2), and despite recent advances in the understanding of the molecular biology controlling energy homeostasis (3) and large efforts to uncover the genetic underpinnings of obesity in human populations (4), the exact cause of weight gain in the majority of people remains unknown. This is not entirely surprising because it is inherently difficult to study the etiology of obesity in humans. The methodologies available to measure various components of daily energy metabolism are either barely precise (energy expenditure) or profoundly inaccurate (energy intake) and thus unable to detect the small differences in energy balance that, when chronically sustained, are likely to be responsible for the development of obesity in the majority of people. For the past 30 years, regulation of energy expenditure has been a dominant theme in human obesity research. The majority of studies have been conducted under the controlled, artificial conditions of a metabolic study unit and often in the resting state. Some studies have indicated that low resting energy expenditure (5) and low wholebody lipid oxidation rates (6,7) are inherited phenotypes, which can predispose some individuals to development of obesity. However, these findings are not universal, and an unequivocal molecular explanation for their regulation is still missing (8). It is only recently, with the advent of the doubly labeled water technique (9), that a limited number of studies have started to provide information on energy expenditure in individuals who are unencumbered by the confines of the laboratory setting. However, to date these areas of research have not produced convincing evidence that abnormal regulation of energy expenditure is either a common or a major risk factor for weight gain in humans. The study of molecular mechanisms and resulting behaviors that underlie energy intake in humans has been even less conclusive. In the vast majority of studies, eating behavior, food intake, and macronutrient preferences are measured by instruments (Visual Analog Scales, food frequency questionnaires, diaries, recalls, diet histories, weighed records) sharing the same weakness; they depend on subjects honestly telling researchers their interoceptive sensations and what they consumed (Table 1). The inability of accurately measuring what people eat and, why, represents the most fundamental flaw in obesity research (10). Faced with these difficulties, many researchers in this field continue to shy away from measurements of eating behavior and food intake as the proximal, genetically dissectible phenotypes and rely instead on measurements of body weight and body composition to identify the molecular cause(s) of the disease.
