Childhood Sugar-Sweetened Beverage Consumption: an Agent-Based Model of Context-Specific Reduction Efforts

Abstract

Introduction: Despite widespread recognition among public health experts that childhood sugar-sweetened beverage consumption should be reduced, doing so has proven to be a challenge. An agent-based model of early childhood sugar-sweetened beverage consumption was applied to data from three high-quality, longitudinal cohort studies to gain insight into potentially effective intervention strategies across contexts.

Methods: From 2021 to 2023, a single agent-based model design was applied to data sets derived from three separate cohorts of children followed from infancy to childhood, with very different populations and environments (participants recruited in 1999-2002; 2003-2010; and 2009-2014). After assessing its ability to reproduce observed consumption patterns across cohorts, it was used to simulate potential impacts of multiple intervention strategies across contexts.

Results: Interventions reducing home availability of sugar-sweetened beverages consistently had the largest potential effects. Impact differed between cohort settings: a complete decrease in availability resulted in an estimated 87% decrease in overall early childhood consumption for one of the cohorts, compared with 61% and 54% in the others. Reducing availability in center-based child care resulted in substantially greater reduction in one cohort relative to the other two.

Conclusions: There is untapped potential for strategies targeting children's sugar-sweetened beverage consumption in the home, but in some instances, other approaches might also yield meaningful effects. Tailoring approach to setting may be important, and agent-based models can be informative for doing so. This agent-based model has broad generalizability and potential to serve as a tool for designing effective, context-specific strategies to reduce childhood sugar-sweetened beverage consumption.

Figure 1:. Conceptual figure of model design.

Ovals represent static model elements, and rectangles those that are potentially dynamic; a black border indicates that the model element explicitly changes as a function of child age. Arrows denote relationships between model elements, with a dashed line indicating that there is interaction between individuals (i.e., families adjust home SSB availability based on their observation of others). Our primary model output, SSB consumption, shown in a grey rectangle, is a dynamic function of availability of SSBs in settings and time children spend in those settings. Figure abbreviations: Body mass index (BMI); Sugar-sweetened beverages (SSB)

Figure 2:. Impact of selected interventions.

Interventions are described on the Y-axis. The uppermost intervention reflects an increase in access to and effectiveness of pediatrician advice for high-BMI children’s families to reduce SSB availability, while others represent reductions in SSB availability in specified settings. Mean impact on weekly SSB servings between ages 2 and 7 is shown with bars and 95% confidence intervals with whiskers.