Using Quantile Regression to Measure the Differential Impact of Economic and Demographic Variables on Obesity

Abstract

The fight against obesity in the U.S. has become a pressing priority for policy makers due to many undesirable outcomes including escalating health care costs, reduced quality of life and increased mortality. This analysis uses data from the 2007 Behavioral Risk Factor Surveillance System (BRFSS) to evaluate the relationship between behavioral, economic, and demographic factors with BMI while explicitly accounting for systematic heterogeneity using a quantile regression. Results suggest that the effect of exercise, smoking, occupation, and race vary by sizeable amounts from high to low BMI-quantiles. This strongly indicates that future research efforts and policy responses to obesity need to account for these differences in order to develop more effective policies.