Nonparametric Comparison of Longitudinal Profiles of Healthy Eating Index Scores

• Activity Number: 177
• Type: Contributed
• Date/Time: Monday, August 10, 2015 : 10:30 AM to 12:20 PM
• Sponsor: ENAR
• Author(s): Aiyi Liu*
• Companies: NICHD/NIH
• Keywords: ratio of variables ; normality ; permutation ; longitudinal ; nonparametric

Abstract

The healthy eating index (HEI) 2005 is a popular measure of diet quality that assesses conformance to Federal dietary guidance, and is the total score of 12 food components, each is derived by truncating the ratio of two random variables (food amount to total energy). The complex nature of the HEI scores, including measurement errors, zero-inflated data, and heterogeneity in distribution across time, poses great challenges in analyzing longitudinally collected HEI data, especially when using parametric modeling. For group comparison of longitudinal profiles of HEI2005 scores, we propose a nonparametric procedure and demonstrate its performance via simulation studies, and exemplify its application to a recently completed dietary intervention study among children with type 1 diabetes.