|
Activity Number:
|
438
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
| Abstract - #303688 |
|
Title:
|
Identifying Early Determinants of Overweight/Obesity Among Preschoolers: A Comparison of Classification Trees, Multivariate Adaptive Regression Splines, and Logistic Regression
|
|
Author(s):
|
Panagiota Kitsantas*+
|
|
Companies:
|
George Mason University
|
|
Address:
|
4400 University Blvd, Fairfax, VA, ,
|
|
Keywords:
|
classification trees ; MARS ; overweight/obesity ; preschoolers
|
|
Abstract:
|
Data from the Early Childhood Longitudinal Study-Birth Cohort was utilized to study early determinants of overweight/obesity in preschoolers using three different methodologies, classification and regression trees (CART), multivariate adaptive regression splines (MARS) and logistic regression. CART analysis revealed that children were more likely to be overweight/obese if they were overweight/obese at age 2, their mothers were also overweight/obese (pre-pregnancy BMI), they were Black or Hispanic and they were breastfed less than 2.5 months. MARS analysis indicated that child's weight status at age 2, maternal overweight/obesity, race/ethnicity, smoking and socioeconomic status were strongly associated with overweight/obesity at age 4. The predictive performance of these techniques was similar. Implications of the findings are discussed.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2009 program |
