Confounding Effects of Food Intake and Physical Activities on Student Body Mass Index Using General Loglinear Models

Abstract

Background: There is need to prioritize the promotion of physical activity and eating habit to promote healthy practices and prevent the occurrence of disease and obesity. Objective: This study aims to examine the association strength between types of food intake and physical activities using General Loglinear Models (GENLOG). Methods: A cross-sectional survey was conducted on selected secondary schools in the eastern states of Peninsular Malaysia. It involved a sample of 326 students (aged between 13 to 17 years). A questionnaire was administered to the students within the school compound and a face-to-face interview was conducted. Ratio Chi-square statistic was used to determine the association between the types of food intake and physical activities; while, Mantel Haenszel’s test of homogeneity was used to identify the presence of confounding variables. Results: The results showed that the consumption of breakfast, fast food, and snacks was significantly associated with various physical activities with some confounding effects. Conclusion: Comparison of several loglinear models confirmed the homogenous loglinear model as most parsimonious in explaining the association between types of food intake and physical activity among adolescents.