Dietary Energy Density and Weight Loss

Energy density (ED) is defined as the energy content of a food relative to its weight (kcal/g). Water and fiber are primarily responsible for lowering ED while fat greatly increases ED. The 2010 Dietary Guidelines for Americans assert, “Eating patterns that are low in calorie density improve weight loss and weight maintenance.” Theoretically, by reducing energy density, people can eat a larger volume of food and control their appetites while ingesting the same or less energy. This is proposed to improve adherence to an energy-restricted diet and result in sustained weight loss. Although diets incorporating low ED foods may be healthful, careful scrutiny raises questions about the effectiveness of this approach for weight management.

The first concern with this concept stems from the methods used to calculate a food’s or diet’s ED score. Criteria vary across studies, particularly regarding inclusion or exclusion of beverages, and this markedly changes the final ED value (Cox & Mela, 2000). Inclusion of beverages frequently results in weak or no association between ED and energy intake, body weight, body mass index (BMI) or weight loss (Ledikwe et al., 2007; Johnson et al., 2008, 2009; Perez-Escamilla et al., 2012). Yet exclusion of beverages from calculations lacks a physiological basis. Some argue that because beverages elicit different behavioral and physiological responses compared to solid foods, they should not be included in the ED score (see Ledikwe et al., 2005, for summary). However on this basis, fat and fiber should also be excluded because they also evoke different behavioral and physiological effects (altered gastric emptying, increased viscosity in the stomach or intestines, and so on). Further, exclusion of beverages distorts the overall computation since beverages contribute approximately 20% of daily dietary energy (Popkin, 2010).

Second, if valid, an association between ED and an index of body weight or composition should be apparent with foods at the extremes of the ED spectrum. However, this is not supported by literature. The least energy dense diet component is beverages (0.0-0.6 kcal/g), so high consumption should aid weight management. Yet strong evidence implicates consumption of energy-yielding beverages in weight gain. Alternatively, the ED concept would predict that very energy dense foods, such as nuts (ED = 5.7-6.6 kcal/g), should lead to weight gain. However, studies with nuts consistently indicate very little or no weight gain. They may actually aid weight loss as the body fails to completely utilize the energy provided by nuts (Mattes et al., 2008). Hence, the foods most likely to support the ED concept fail to do so.

Third, the literature does not consistently support a direct relationship between ED and body weight or BMI. While some studies reveal a positive association (Ello-Martin et al., 2007; Savage et al., 2008; Vergnaud et al., 2009), most trials or interventions, particularly those that are the most scientifically rigorous, show no association (Saquib et al., 2008; Du et al., 2009). Analysis of data from the Nurses’ Health Study II shows women who maintained high ED diets over eight years gained less weight than those who maintained low ED (Bes-Rastrollo et al., 2008). Additionally, interventions increasing fruit and/or vegetable consumption (resulting in lower average ED) may lead to either no change in body weight (Pierce et al., 2007) or increased body weight (Houchins et al., 2011) as people do not always adequately compensate for the additional energy.

Fourth, attributing a role for ED in the increasing BMI trends over the past three decades requires, at a minimum, a shift toward diets with higher ED. However, contrary to this prediction, water consumption has remained stable or increased (Barnard, 2010; Popkin, 2010), fat consumption has been largely stable (Wright & Wang, 2010), and fiber intake has remained stable (Briefel & Johnson, 2004; King et al., 2012). Fat availability in the United States’ food supply did rise after 2000 (USDA, 2011), but this was concurrent with a reduction in the incidence of BMI (Flegal et al., 2010). Thus, none of the primary determinants of ED have changed in a direction consistent with a hypothesis that ED shifts have driven BMI increases.

--- PAGE BREAK ---

Fifth, despite claims that low ED foods promote satiety, trials attempting to use the concept to manage body weight have had mixed success. As with many dietary approaches, short-term losses are not sustained (Lowe et al., 2008; Saquib et al., 2008). There may be multiple reasons for this, but prominent among them is the inconvenience of using many of the foods, such as fresh produce, that comprise a low ED diet.

Overall, there are considerable inconsistencies in the evidence and methods for correlating ED and weight status. To reiterate, diets with low ED may be healthful, but they will not necessarily lead to spontaneous reductions of energy intake and body weight. To classify foods on a single dimension, such as ED, for a therapeutic purpose while ignoring other attributes such as nutrient density poses risks for nutrient imbalances.

Cordelia A. Running , M.Sc.
Graduate Research Assistant,
Department of Food Science,
Purdue University, West Lafayette, IN 
[email protected]

Richard D. Mattes, Ph.D., R.D. 
Distinguished Professor of Nutrition Science,
Purdue University, West Lafayette, IN 
[email protected] 

References