Selection of measures in epidemiologic studies of the consequences of obesity

Abstract

The most popular measure for conducting analyses in studies of adiposity is body mass index (BMI); however, BMI does not discriminate between muscle and adipose tissue and does not directly assess regional adiposity. In this article, we address the question of whether alternatives to BMI should be used in epidemiologic analyses of the consequences of obesity. In general, measures of fat distribution such as waist circumference and sagittal abdominal diameter are more highly correlated with cardiovascular disease risk factors and diabetes than BMI; however, differences are usually small. Precise measures of adiposity from methods such as dual-energy x-ray absorptiometry may provide more specific and larger associations with disease, but published studies show that this is not always true. Further, practical considerations such as cost and feasibility must influence the choice of measure in many studies of large populations. Measures of adiposity are highly correlated with each other, and the additional cost of a more precise measure may not be justified in many circumstances. Validated prediction equations that include multiple anthropometric measures, along with demographic variables, may offer a practical means of obtaining assessments of total adiposity in large populations, whereas waist circumference can provide a feasible assessment of abdominal adiposity. Finally, public health messages to the public must be simple to be effective. Therefore, investigators may need to consider the ease of translation of results to the public when choosing a measure.