Dose-dependent and joint associations between screen time, physical activity, and mental wellbeing in adolescents: an international observational study

Mental wellbeing in adolescents has declined considerably during past decades, making the identification of modifiable risk factors important. Prolonged screen time and insufficient physical activity appear to operate independently and synergistically to increase the risk of poor mental wellbeing in school-aged children. We aimed to examine the gender-stratified dose-dependent and joint associations of screen time and physical activity with mental wellbeing in adolescents.

We used data from three rounds of Health Behaviour in School-aged Children cross-sectional surveys (2006, 2010, and 2014) from 42 European and North American countries. Survey participants, aged 11 years, 13 years, and 15 years, provided self-reported information by completing an anonymous questionnaire that included items on health indicators and related behaviours. We used the self-reported variables of life satisfaction and psychosomatic complaints as indicators of adolescents’ mental wellbeing, combining these with the self-reported discretionary use of screens and engagement in physical activity. We used generalised additive models and multilevel regression modelling to examine the gender-stratified relationships between mental wellbeing and screen time and physical activity.

Our sample included 577 475 adolescents (mean age 13·60 years, SD 1·64), with 296 542 (51·35%) girls and 280 933 (48·64%) boys. The mean reported life satisfaction score (on a scale of 0–10) was 7·70 (95% CI 7·69–7·71) in boys and 7·48 (7·46–7·50) in girls. Psychosomatic complaints were more common among girls (mean 9·26, 95% CI 9·23–9·28) than boys (6·89, 6·87–6·91). Generalised additive model analyses showed slightly non-linear associations of screen time and physical activity with life satisfaction and psychosomatic complaints for girls and boys. Detrimental associations between screen time and mental wellbeing started when screen time exceeded 1 h per day, whereas increases in physical activity levels were beneficially and monotonically associated with wellbeing. Multilevel modelling showed that screen time levels were negatively associated with life satisfaction and positively associated with psychosomatic complaints in a dose-dependent manner. Physical activity levels were positively associated with life satisfaction and negatively associated with psychosomatic complaints in a dose-dependent manner. Joint associations of screen time–physical activity with mental wellbeing showed that, compared with the least active participants with more than 8 h per day of screen time and no physical activity, most of the other screen time–physical activity groups had considerably higher life satisfaction and lower psychosomatic complaints.