chapter  9
Recursive Partitioning: Predicting Body Fat and Glaucoma Diagnosis
Pages 16

Worldwide, overweight and obesity are considered to be major health problems because of their strong association with a higher risk of diseases of the metabolic syndrome, including diabetes mellitus and cardiovascular disease, as well as with certain forms of cancer. Obesity is frequently evaluated by using simple indicators such as body mass index, waist circumference, or waist-tohip ratio. Specificity and adequacy of these indicators are still controversial, mainly because they do not allow a precise assessment of body composition. Body fat, especially visceral fat, is suggested to be a better predictor of diseases of the metabolic syndrome. Garcia et al. (2005) report on the development of a multiple linear regression model for body fat content by means of p = 9 common anthropometric measurements which were obtained for n = 71 healthy German women. In addition, the women’s body composition was measured by Dual Energy X-Ray Absorptiometry (DXA). This reference method is very accurate in measuring body fat but finds little applicability in practical environments, mainly because of high costs and the methodological efforts needed. Therefore, a simple regression model for predicting DXA measurements of body fat is of special interest for the practitioner. The following variables are available (the measurements are given in Table 9.1):

DEXfat: body fat measured by DXA, the response variable,

age: age of the subject in years,

waistcirc: waist circumference,

hipcirc: hip circumference,

elbowbreadth: breadth of the elbow, and

kneebreadth: breadth of the knee.