Abstract
BACKGROUND: This study investigated the patterns and determinants of smart device usage among children over a period of six years, utilizing latent trajectory class analysis with a growth mixture model (GMM) to identify risk groups based on their screen time trajectories. METHODS: The Kids Cohort for Understanding of internet addiction Risk factors in Early childhood provided data from six waves of surveys conducted from 2018 to 2023, with a final sample size of 313 participants (mostly mothers) who completed all assessments. We used a GMM to analyze six-year patterns of children's smart device usage. Subgroups were identified based on screen time trajectories using GMM, and the impact of sex, age, parental education, family income, and child behavior checklist (CBCL) scores on group membership was assessed with multinomial logistic regression. RESULTS: Three distinct risk groups were identified: a low-risk group (69.86%) with minimal and stable screen time, a mid-risk group (21.92%) with moderate and occasionally increased screen time, and a high-risk group (8.22%) with consistently high and increasing screen time. Factors such as older age of the child, higher early CBCL scores, lower maternal educational levels, and lower economic status were significantly associated with a higher risk of excessive screen time. CONCLUSION: This study highlights the significant variations in children's smart device usage patterns influenced by behavioral and socioeconomic factors. The findings suggest the need for targeted interventions that address the specific needs and risks associated with different screen-time trajectories. Effective strategies should consider both individual and familial factors to promote healthier digital habits among children.