It then becomes possible to look for correlations between those trends and the varying trends in life expectancy.

The standard way of assessing a person’s weight relative to height is with body mass index (BMI), defined as a person’s weight in kilograms divided by the square of the person’s height measured in meters. Thus, a person who weighs 75 kilograms (165 pounds) and is 1.75 meters high (5 feet, 9 inches) has a BMI of 24.5.

According to the World Health Organization (2006), a person is of normal weight if he or she has a BMI between 18.50 and 24.99. A person with a BMI of 25 or more is considered overweight, while anyone with a BMI of 30 or more is considered obese. Obesity is further subdivided into Class 1 (30–34.99), Class 2 (35–39.99), and Class 3 (40 and above). For instance, a person who is 1.6 meters (5 feet, 3 inches) tall and weighs 77 kilograms (169 pounds) has a BMI of slightly more than 30, or is borderline obese.

BMI is the most widely used measure for determining obesity mainly because it is convenient. It is easy and straightforward to obtain a person’s height and weight and then apply the BMI formula. Other measures of obesity, such as percentage of body fat, are generally much more difficult to determine accurately. However, errors are not uncommon in self-reports of height and weight (Ezzati et al., 2006). Balancing out its convenience, moreover, is the fact that BMI is no more than a rough measure of how fat a person is. People who are particularly muscular will have a high BMI, for instance, even if they have very low body fat. Furthermore, there is evidence that not all fat is created equal and that certain fat distributions—such as intra-abdominal fat around the waist area—are unhealthier than others (Bergman et al., 2006; Snijder et al., 2006). Thus, many studies use other measures, such as waist-to-hip ratio or body fat percentage, to gauge health risks. Still, BMI has the advantage that, because it is so convenient, far more data—particularly historical and international data—exist for it than for any of the other measures. For that reason, this chapter focuses on BMI data while noting the limitations of relying strictly on these data and describing other relevant data where they exist.

One of the most striking health-related trends in the United States over the past 50 years has been the rise of obesity. As can be seen in Figure 3-1, in 1960−1962 only 10.7 percent of adult males and 15.8 percent of adult females in the country were obese. By 2007−2008, those numbers had risen to 32.2 and 35.5 percent, respectively. For both sexes, the growth in obesity