Identifying developmental trajectories of body mass index in childhood using latent class growth (mixture) modelling: associations with dietary, sedentary and physical activity behaviors: a longitudinal study

Abstract

Background

To date, many epidemiologic studies examining associations between obesity and dietary and sedentary/physical activity behaviors have focused on assessing Body Mass Index (BMI) at one point in time. Recent developments in statistical techniques make it possible to study the potential heterogeneity in the development of BMI during childhood by identifying distinct subpopulations characterized by distinct developmental trajectories. Using Latent Class Growth (Mixture) Modelling (LCGMM) techniques we aimed to identify BMI trajectories in childhood and to examine associations between these distinct trajectories and dietary, sedentary and physical activity behaviors.

Methods

This longitudinal study explored BMI standard deviation score (SDS) trajectories in a sample of 613 children from 4 to 12 years of age. In 2006, 2009 and 2012 information on children’s health related behaviors was obtained by parental questionnaires, and children’s height and weight were measured. Associations with behaviors were investigated with logistic regression models.

Results

We identified two BMI SDS trajectories; a decreasing BMI SDS trajectory (n = 416; 68 %) and an increasing BMI SDS trajectory (n = 197; 32 %). The increasing BMI SDS trajectory consisted of more participants of lower socio-economic status (SES) and of non-western ethnicity. Maternal overweight status was associated with being in the increasing BMI SDS trajectory at both baseline and follow-up six years later (2006: Odds Ratio (OR), 2.9; 95 % confidence interval (CI) 1.9 to 4.3; 2012 OR, 1.8; 95 % CI 1.2 to 2.6). The increasing BMI SDS trajectory was associated with the following behaviors; drinking sugared drinks > 3 glasses per day, participation in organized sports < 1 h per week, and TV viewing > 2 h per day, though participation in organized sports at follow-up was the only significant result.

Conclusions

Our results indicate the importance of healthy lifestyle behaviors at a young age, and indicate that maternal BMI is a very important risk factor for the development of childhood overweight. Comprehension of heterogeneity in the development of BMI and associations with modifiable health related behaviors is interesting for prevention by targeting high risk behaviors in early childhood, especially in low SES children, children of non-western ethnicity and children whose mother is overweight.