Abstract
OBJECTIVE: Digital parenting restrictions are widely used to manage adolescent screen use, yet it is unclear how effective these strategies remain as children age and gain autonomy. This study examined longitudinal patterns of digital parenting restrictions and their relationship to adolescent screen time using self-reported and passive measures. METHOD: Data came from Years 3-5 of the Adolescent Brain Cognitive Development (ABCD) Study (n = 8,096). Latent Transition Analysis identified Low, Moderate, and High Restriction profiles from 14 Parent Screen Time Questionnaire items. Linear mixed-effects models tested associations between profiles and screen use outcomes (self-reported screen time, self-reported smartphone use, passive smartphone use from the ABCD-EARS app, and smartphone keystroke activity). Interaction terms examined age effects. Gradient boosting (XGBoost) identified the most influential parenting practices. RESULTS: High and Moderate Restriction profiles were linked to lower screen use across all measures. Compared to the Low Restriction group, High Restriction participants reported 56 fewer daily minutes of screen time (p < 0.0008) and logged 286 fewer daily keystrokes (p < 0.0008). A significant interaction (β = 13.0; 95% CI: 7.9, 18.2) indicated that the effect of high restriction diminished with age. Machine learning highlighted bedroom device access (e.g., screens playing at bedtime, mobile devices in bed) as the strongest predictor of higher screen time. CONCLUSION: Parenting restrictions reduce adolescent screen time, but their impact declines as adolescents gain independence. Limiting device access in the bedroom may be one of the most effective and lasting strategies for moderating screen use and fostering healthier digital habits.