Importance: Excessive screen time has been associated with a higher risk for mental health problems, but whether the associations differ by screen content types is unclear.
Objective: To examine the allocation of and longitudinal changes in screen exposure across different content types and to explore their associations with mental health in children aged 3 to 6 years.
Design, setting, and participants: This cohort study used 3-wave, lagged generalized estimating equation models to analyze data from the Shanghai Children's Health, Education and Lifestyle Evaluation-Preschool (SCHEDULE-P) study in Shanghai, China. The cohort was a representative sample of kindergarten children. Data were collected between November 2016 and May 2019 when children were aged 3 to 4 years (wave 1), 4 to 5 years (wave 2), and 5 to 6 years (wave 3). Data analysis was performed between June 2022 and May 2023.
Exposure: Screen exposure (total daily time and time with each type of content, including educational programs, entertainment programs, non-child-directed programs, electronic games, and social media) was collected when children were aged 3, 5, and 6 years.
Main outcomes and measures: Mental health of children at age 3, 5, and 6 years was reported by parents using the Strengths and Difficulties Questionnaire.
Results: Of the 15 965 children included in the representative sample, 8270 were males (51.7%) and the mean (SD) age at wave 1 was 3.73 (0.30) years. As children developed from ages 3 to 6 years, the proportion of screen exposure to educational programs (≤1 hour per day: 45.0% [95% CI, 43.5%-46.5%] to 26.8% [95% CI, 25.3%-28.3%]) and entertainment programs (≤1 hour per day: 44.4% [95% CI, 42.8%-45.9%] to 32.1% [95% CI, 30.4%-33.9%]) decreased, whereas exposure to social media increased (≤1 hour per day: 1.5% [95% CI, 1.2%-1.9%] to 27.1% [95% CI, 25.5%-28.7%]). The associations between on-screen content and mental health varied. For a given total screen time, a higher proportion of screen exposure to educational programs was associated with a lower risk for mental health problems (adjusted odds ratio [AOR], 0.73; 95% CI, 0.60-0.90), whereas non-child-directed programs were associated with a higher risk for such problems (AOR, 2.82; 95% CI, 1.91-4.18). Regardless of the content, total screen time was consistently associated with mental health problems.
Conclusions and relevance: Results of this study indicated that both total screen time and different types of content were associated with mental health problems in children aged 3 to 6 years. Limiting children's screen time, prioritizing educational programs, and avoiding non-child-directed programs are recommended.