tively, while there was a relatively small increase in average height. The associated shift in average body mass index (BMI)—the typical weight-for-height measure used to assess obesity—was 3 kg/m2 in adults, in both men and women, from approximately 25 to 28 (Ogden et al., 2004). Figure 1-1 shows the shift in the distribution of BMI among U.S. adults between 1976-1980 and 2005-2006, with the gain being most pronounced among the heaviest people (Ogden et al., 2007).
Data from the 2007-2008 National Health and Nutrition Examination Surveys (NHANES)2 show that among adults aged 20 or older, nearly 34 percent have weight levels in the obese range (BMI ≥ 30) and another 34 percent are classified as overweight (BMI between 25 and 30), thus the combined prevalence of overweight and obesity is nearly 68 percent (Flegal et al., 2010). Among children and adolescents aged 2 through 19, nearly 17 percent are classified as obese (BMI for age and sex ≥ 95th percentile), and 15 percent are classified as overweight (BMI for age and sex ≥ 85th but < 95th percentile), thus close to 32 percent are either overweight or obese (Ogden et al., 2010). There is no evidence that underlying biological susceptibility to weight gain has changed, although there is ample evidence of increases in such factors as the amount of food available and where it is available, creating eating environments that are highly conducive to the consumption, often unintentional, of excess calories (Kral and Rolls, 2004; Ledikwe et al., 2005; Story et al., 2008; Wansink, 2004). There is also no evidence that the prevalence of individual weight control attempts has decreased as average population weights have increased. For example, the proportion of U.S. adult men and women who reported attempting to lose weight was, respectively, 29 and 44 percent in 1996 and 33 and 46 percent in 2000 (Bish et al., 2005; Serdula et al., 1999). Accordingly, the increasing prevalence of obesity in diverse populations reveals the limitations of biological and behavioral controls on caloric balance in the face of high caloric availability and the relatively low levels of physical activity that are common in urban and westernized societies (Booth et al., 2001; French et al., 2001; IOM, 2005; Kumanyika et al., 2002; Popkin and Gordon-Larsen, 2004). The need for multifaceted actions to prevent obesity should no longer be a subject of serious debate.
Many multifaceted interventions for obesity prevention influence the calorie balance by focusing on either energy intake or energy expenditure. According to Swinburn and colleagues (2009), it is important to know whether one of these two factors is the primary driver of the obesity epidemic in order to prioritize potential interventions. The relative contributions of increased energy intake and decreased energy expenditure to the obesity epidemic have been debated, with some suggesting that increased energy intake is sufficient to explain the weight gain of the American population since the 1970s (Bleich et al., 2008; Swinburn et al., 2009). Both factors are clearly important for energy balance and other health outcomes, and this report does not aim to tell decision makers whether to focus on energy intake, energy expen-