Longitudinal Associations Among Diet Quality, Physical Activity and Sleep Onset Consistency With Body Mass Index z-Score Among Toddlers in Low-income Families

Abstract

Background: Habits surrounding health behaviors (i.e., sleep, physical activity, diet) are developed in toddlerhood. Lack of consistent health habits may increase obesity risk among toddlers in low-income families.

Purpose: To compare the role of sleep onset consistency, physical activity and diet quality as mediators between household poverty and toddler weight.

Methods: Two hundred and seven toddlers (mean age = 20.2 months, 46% female, 68.1% Black) participating in an obesity prevention trial were assessed at three time points over 12 months. Using Actical accelerometers, we assessed sleep and physical activity at each time point for up to 1 week. We defined sleep onset consistency as the standard deviation of sleep onset across all days. We calculated the Healthy Eating Index-2015 from a 24-hr dietary recall. We used WHO standards to calculate BMI-for-age z-scores from toddlers' weight/length, and calculated poverty ratio from parent-reported income and family size. Multilevel mediation models tested toddler sleep onset consistency, physical activity, and diet quality as mediators between household poverty and toddler BMI z-score.

Results: Toddlers from households with higher poverty ratios had more inconsistent sleep onset times. Toddlers with more inconsistent sleep onset times had higher BMI z-scores across all timepoints, even when accounting for physical activity and diet quality. Sleep onset consistency indirectly explained the association between household poverty and BMI z-score.

Conclusions: Inconsistent sleep schedules could help explain the association between poverty and BMI. Future research should examine strategies to support low-income families to develop and maintain routines as a mechanism to prevent obesity and reduce disparities.

Trial registration number: NCT02615158.

Keywords: Bedtime; Diet quality; Obesity; Physical activity; Poverty; Toddlers.

© Society of Behavioral Medicine 2020. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.

Figures

Fig. 1.

Multilevel mediation model illustrating within-person effects at level 1 and between-person effects at level 2. All observed variables measured at three time points (Time 1 [enrollment], Time 2 [6-month follow-up], Time 3 [12-month follow-up]).

Fig. 2.

Data processing from baseline sample to analytic sample.