How large is the energy gap that accounts for the obesity epidemic?

Humans are gaining excess weight, and alarmed scientists generally agree on the cause: positive energy balance beyond that needed to maintain a healthy weight. After all, body mass largely reflects a mix of water and energy-yielding organic chemicals, and when considered as a thermodynamic system, stores can only increase when substrate intake exceeds losses. Agreement, however, stops beyond this simple conjecture, and a consensus is lacking on fundamental unanswered questions: What is the magnitude of ‘‘energy imbalance’’? Can the energy imbalance be explained by excessive intake, reduced levels of physical activity, or a combination of both? Whereas these questions are debated among members of the academic community, their resolution has immense practical implications. By how much should people in developed nations reduce what they eat and increase the amount they exercise to ‘‘return’’ to weights before the ‘‘obesity epidemic’’ began? Public health efforts can be aimed at arresting or preventing weight gain by fostering actions aimed at achieving neutral energy balance and thus closing the ‘‘energy gap’’ (1). Our questions could easily be answered if we had quantitative estimates of energy intake and expenditure, including that related to activity before the obesity epidemic and in the current population, particularly among thosewith excess adiposity. Food intake is highly variable from day to day, and available information on what people ate a half-century ago is based largely on inaccurate or biased self-report or food disappearance data (2). Activity estimates similarly are founded on self-report, observational studies, or indirect measures such as heart rate analyses, step counting, and fitness evaluations in selected populations. Because the energy imbalance causing the obesity epidemic is small for the population as a whole, in the range of several hundred calories per day, all agree that these approaches are far too imprecise to answer our questions with an acceptable level of accuracy. Our ability to accurately quantify human energy intake and expenditure under nonlaboratory conditions only became available on a limited basis in the 1980s with the introduction of the doubly labeled water method (3). Subjects’ ingestion of 2 stable isotopes of water provided a means of quantifying total energy expended over a period of several weeks. When subjects are in near-energy equilibrium with stable weight over the study period, the measured expended energy is equivalent to ingested energy, and doubly labeled water evaluation provides an objective measure of calories ingested from food. The doubly labeled water method can be used to estimate energy expended in activity in which the investigator also measures subject resting metabolic rate (3). These types of objective data are lacking for populations before the obesity epidemic; in addition, doubly labeled water is costly and application so far includes only several thousand subjects participating in clinical trials in many laboratories throughout the world. New methods of quantifying energy balance components are now being introduced, but, at present, these important advances do not help us answer the energy imbalance questions. The current controversy surrounds the definition of ‘‘energy gap’’ (1) and the means by which this energy imbalance is calculated. What happens to weight and energy expenditure if a person decides to increase his or her intake while maintaining a stable activity level? Positive energy balance and net protein and fat accretion with weight gain will follow. The creation of new molecules in addition to those created with daily turnover comes with a cost that is typically described as the ‘‘efficiency’’ of new substrate synthesis or tissue growth. The gain in weight and metabolically active cell mass will also impose an increase in the energy expended in activity and during rest. A new steady state weight will eventually be reached when all of these factors combine to achieve neutral energy balance. What happens if this person reduces his or her activity level while maintaining a stable food intake? Positive energy balance with protein, fat, and weight gain will again follow. Food intake is ‘‘clamped’’ and will not change, although the subject will experience a weight-related increase in energy expended with activities and during rest. Weight gain will cease once the person reaches energy equilibrium. We need to add yet another level of complexity when we consider children and adolescents. Variable levels of positive energy balance are present throughout the growth period, and normal growth must be accounted for when estimating the imbalance leading to overweight and obesity. Subjects who gain excess weight are thus dynamically moving over time between various growth and nutritional ‘‘planes’’ with corresponding body composition and metabolic effects. These mass and functional changes can be exploited as energy imbalance biomarkers, and, through their measurement, investigators have used variable approaches in estimating the energy gap. In 2003 Hill et al (1) were the first investigators to propose a definition of ‘‘energy gap’’ as the ‘‘required change in energy expenditure relative to energy intake necessary to restore energy