Screen Time is Associated with Cardiometabolic and Cardiovascular Disease Risk in Childhood and Adolescence

Abstract

BackgroundScreen time in children and adolescents may be linked to cardiometabolic and cardiovascular risk. This study examines the relationship between screen time and cardiometabolic risk (CMR) factors.MethodsWe analysed data from over 1,000 participants in the Copenhagen Prospective Studies on Asthma in Childhood cohorts (COPSAC2010 and COPSAC2000). This longitudinal study utilised objective measures of physical activity, sleep, pubertal development, and dietary intake as covariates, and assessed mediating and moderating effects of lifestyle factors on parental- and self- and reported discretionary screen time. Our primary outcome of interest was a CMR score which was made from standardised z-scores of metabolic syndrome components (waist circumference, systolic blood pressure, HDL cholesterol, triglycerides, and glucose), adjusted for sex and age. Secondary outcomes were insulin resistance, inflammation, atherogenic lipoproteins, and anthropometric measures. We utilised supervised machine learning modelling of blood NMR metabolomics to identify a unique metabolic signature of screen time. Finally, we assess screen time associations with a predicted Cardiovascular Risk Score derived from Cox proportional hazards models of 10-year CVD events trained in the UK Biobank.ResultsIncreased screen time was significantly associated with CMR in children and adolescents, with each additional hour of screen time linked to a higher CMR z-score (children at 10-years: 0.08 [0.01 - 0.14], p=0.021; adolescents at 18-years: 0.13 [0.07 - 0.2], p=0.001). In childhood, sleep duration (p=0.029) and time of onset (p=0.009) significantly moderated the relationship between screen time and CMR; less sleep combined with high screen time significantly increased cardiometabolic risk. In adolescence, sleep duration likewise significantly moderated the association between screen time and CMR (p=0.012), replicating the findings from childhood. A supervised machine learning model trained in the childhood cohort identified a unique metabolic signature in the blood NMR metabolome associated with screen time, which was validated in the adolescent cohort (0.14 [0.03-0.26], p=0.014). CVD-risk scores modelled from CVD-events were directionally associated with screen time in childhood (0.06 [−0.02 - 0.13], p=0.15) and significantly associated with screen time in adolescence (0.07 [0.01 - 0.13], p=0.017) in fully adjusted models.ConclusionIncreased screen time is significantly associated with higher cardiometabolic risk in children and adolescents, with sleep duration moderating this relationship. A unique metabolic signature of screen time was validated across cohorts, and screen time was associated with higher cardiovascular risk scores in adolescence. These findings underscore the importance of considering screen time and sleep duration in addressing cardiometabolic and cardiovascular risks.