Associations between early-life screen viewing and 24 hour movement behaviours: findings from a longitudinal birth cohort study

Abstract

Screen viewing is a sedentary behaviour reported to interfere with sleep and physical activity. However, few longitudinal studies have assessed such associations in children of preschool age (0-6 years) and none have accounted for the compositional nature of these behaviours. We aimed to investigate the associations between total and device-specific screen viewing time at age 2-3 years and accelerometer-measured 24 h movement behaviours, including sleep, sedentary behaviour, light physical activity, and moderate-to-vigorous physical activity (MVPA) at age 5·5 years. The Growing Up in Singapore Towards healthy Outcomes (GUSTO) study is an ongoing longitudinal birth cohort study in Singapore, which began in June 2009. We recruited pregnant women during their first ultrasound scan visit at two major public maternity units in Singapore. At clinic visits done at age 2-3 years, we collected parent-reported information about children's daily total and device-specific screen viewing time (television, handheld devices, and computers). At 5·5 years, children's movement behaviours for 7 consecutive days were measured using wrist-worn accelerometers. We assessed the associations between screen viewing time and movement behaviours (sedentary behaviour, light physical activity, MVPA, and sleep) using Dirichlet regression, which accounts for the compositional nature of such behaviours. This study is active but not recruiting and is registered with ClinicalTrials.gov, NCT01174875. Between June 1, 2009, and Oct 12, 2010, 1247 pregnant women enrolled and 1171 singleton births were enrolled. 987 children had parent-reported screen data at either 2 or 3 years, of whom 840 attended the clinic visit at age 5·5 years, and 577 wore an accelerometer. 552 children had at least 3 days of accelerometer data and were included in the analysis. Total screen viewing time at age 2-3 years had a significant negative association with sleep (p=0·008), light physical activity (p<0·0001), and MVPA (p<0·0001) in relation to sedentary behaviour at age 5·5 years. Compared with children who spent 1 h or less per day screen viewing at age 2-3 years, children who screen viewed for 3 h or more per day at 2-3 years engaged in more sedentary behaviour (439·8 mins per day [≤1 h screen viewing time] vs 480·0 mins per day [≥3 h screen viewing time]), and less light physical activity (384·6 vs 356·2 mins per day), and MVPA (76·2 vs 63·4 mins per day) at age 5·5 years. No significant differences in time spent sleeping were observed between the groups (539·5 vs 540·4 mins per day). Similar trends were observed for television viewing and handheld device viewing. Longer screen viewing time in children aged 2-3 years was associated with more time spent engaged in sedentary behaviour and shorter time engaged in light physical activity and MVPA in later childhood. Our findings indicate that screen viewing might displace physical activity during early childhood, and suggest that reducing screen viewing time in early childhood might promote healthier behaviours and associated outcomes later in life. Singapore National Research Foundation, and Singapore Institute for Clinical Sciences, Agency for Science Technology and Research (A*STAR).