National Health and Nutrition Examination Survey.
Note: This table provides selected information on data available to monitor trends in obesity. For a full list of surveys currently available by country, please refer to NCD‐RisC [1] and GBD [11].
While most national monitoring surveys include measurements of weight and height (Table 2.1), some are conducted via telephone interview and thus rely on self‐report. However, self‐reported weight and height have been shown to underestimate the prevalence of obesity. For example, comparisons of measured and self‐reported height and weight in the United States indicate that women under report their weight, but men do not, and young and middle‐aged men (<65 years) over report their height, but older men do not [17]. These practices result in overall underestimation of BMI and thus obesity prevalence [17]. Under reporting is not negligible. For example, in 2000, the self‐reported prevalence of obesity was above 24% in just three states (Alabama, Mississippi, and District of Columbia). After correcting for under reporting, women in all states except Colorado had an obesity prevalence above 24% [17]. Thus, the most accurate monitoring relies on measured BMI rather than self‐report.
BMI cut‐offs to define overweight, obesity, and severe obesity in adults (Table 2.2) are based on the associations between BMI and mortality risk [18]. However, high BMI levels are even more strongly related to morbidity than mortality [19]. While commonly used, BMI does not differentiate between lean mass and fat mass, which varies by age, sex, and ethnicity. Cut‐offs in children and adolescents are age‐ and sex‐specific [20]. Asian populations have a higher risk of metabolic abnormalities at lower BMI levels. For example, an analysis of data from two US studies, MESA (Multi‐Ethnic Study of Atherosclerosis) and MASALA (Mediators of Atherosclerosis in South Asians Living in America), found that for the equivalent number of metabolic abnormalities at a BMI of 30 kg/m2 in Whites, the corresponding BMI values were 29.9 kg/m2 in Blacks, 27 kg/m2 in Latinx adults, 24.5 kg/m2 in Chinese Americans, and 23.3 kg/m2 in South Asian Americans [21]. The WHO has identified “public health action points” for Asian populations, particularly 23 kg/m2 or higher, representing increased risk, and 27.5 kg/m2 or higher, representing high risk [22]. However, the current WHO BMI cut‐offs are recommended to be retained for international classification and monitoring efforts [22].
Table 2.2 Cut‐offs for defining excess body fat using anthropometric measures and risk of type 2 diabetes and cardiovascular disease [23]
| Adults | Children and adolescents | ||||
|---|---|---|---|---|---|
| BMI (kg/m2) | Asians (kg/m2) | Normal waist circumference | Large waist circumference* | BMI‐for‐age (percentile)** | |
| Healthy weight | 18.5–24.9 | 18.5–22.9 | – | – | 5th–84th |
| Overweight | 25.0–29.9 | 23.0–27.4 | Increased | High | 85th–94th |
| Obesity class I*** | 30.0–34.9 | 27.5–32.4 | High | Very high | ≥95th (obesity) |
| Obesity class II*** | 35.0–39.9 | 32.5–37.4 | Very high | Very high | |
| Obesity class III (severe obesity) | ≥40.0 | ≥37.5 | Extremely high | Extremely high |
Abbreviations: BMI, body mass index; IOTF, International Obesity Task Force; NIH, National Institutes of Health; WHO, World Health Organization.
* Large waist circumference is defined by the NIH and WHO as >102 cm for men and >88 cm for women.
** Percentiles can be calculated using the CDC (US‐based studies) or IOTF (global studies) growth reference [24].
*** Obesity class I, class II, and class III are often combined into a single category of obesity (BMI ≥30 kg/m2 for adults, BMI ≥27 kg/m2 for Asians, and BMI‐for‐age ≥95th percentile for children and adolescents). In global monitoring that includes developing countries, a cut‐off of 25 kg/m2 is typically used (overweight including obesity).
Efforts have been made to identify alternative anthropometric measures that better characterize individuals at increased risk of morbidity and mortality due to excess body fat. One example is waist circumference, which is the next most common anthropometric measurement after BMI, particularly among adults. It is especially useful in those with a normal BMI who are nonetheless at high risk of obesity‐related morbidities due to abdominal obesity. One advantage of waist circumference compared to BMI is that BMI may not decrease after a physical activity intervention due to increased muscle mass, but waist circumference is likely to decrease.
Waist circumference is measured midway between the lower rib margin and the iliac crest. Individuals should be in the standing position with arms relaxed at their sides, without heavy clothing, and the measurement taken at the end of a normal exhalation. It is essential to check that the tape measure lies parallel to the floor and snugly without compressing the skin before reading the measurement. It is also important to train personnel to read the measurement directly in front of the value on the tape measure rather than at an angle or slightly off to the side.
Waist circumference cut‐offs to classify adults at high risk of related morbidity and mortality have been established by the WHO and US National Institutes of Health (NIH) as >102 cm for men and >88 cm for women [7,23]. Similar to BMI, the association of central obesity with morbidity and mortality varies across ethnicities. Asians tend to be at increased metabolic risk at lower waist circumferences than Whites. The International Diabetes Federation has proposed a definition of metabolic syndrome diagnosis that includes central obesity measured by waist circumference and different cut‐offs for South Asian populations [25]. Those proposed cut‐offs are ≥94 cm for Caucasian men and ≥90 cm for South Asian men, and ≥80 cm for women irrespective of ethnicity. For men from sub‐Saharan Africa and the Middle East, it is recommended to use the Caucasian cut‐off, while for men from Latin America, the South Asian cut‐off is recommended.
