Screen time was linked to teen anxiety. Then researchers compared twins

By Pooja Toshniwal PahariaReviewed by Susha Cheriyedath, M.Sc.Oct 12 2026

From gaming at age nine to hours spent online as teenagers, a study following nearly 22,000 Swedish twins examines how recreational screen habits relate to mental health later in life.

Study: Recreational screen use and internalizing problems from preadolescence to young adulthood in a longitudinal twin cohort with co-twin comparison. Image Credit: Marie Zedig / Shutterstock

A recent study published in the journal Nature Human Behaviour found that recreational screen use, measured by frequency at age nine and duration during adolescence, was associated with internalizing symptoms and subsequent clinical outcomes involving recorded anxiety or depression diagnoses and/or antidepressant dispensing in young adulthood.

At ages 15 and 18, boys spent more time on interactive screen activities, including chatting, web browsing, and gaming, but cross-sectional associations with internalizing symptoms were stronger in girls and weakened in comparisons between twins. The findings suggest that shared familial factors contribute to these associations and could inform clearer, evidence-based guidance on youth screen use.

Recreational screen use has become common in childhood and adolescence. Potential links between screen-based entertainment on mobile phones, television, and computers and youth mental health have raised concerns; however, the causal nature of these relationships remains poorly understood.

About the study

In the present study, researchers evaluated associations between screen use for recreational activities from preadolescence through adolescence and internalizing symptoms (anxiety and depressive symptoms) through early adulthood.

The study included 21,797 twins from the Child and Adolescent Twin Study in Sweden (CATSS). The team examined associations across the cohort and compared twins within pairs who differed in screen use to assess the contribution of shared environmental and genetic factors.

The researchers collected survey data on recreational screen habits and emotional or psychological difficulties at age nine (parent-report, CATSS-9), age 15 (self-report, CATSS-15), and age 18 (self-report, CATSS-18). Emotional symptoms were also assessed at age 24 (self-report, CATSS-24). Data were also obtained on participants' sex, birth year and twin zygosity, along with parental countries of birth, educational attainment and civil status.

At age nine, parents assessed their children’s anxiety using the Screen for Child Anxiety-Related Emotional Disorders (SCARED) and depressive symptoms using the Short Mood and Feelings Questionnaire (SMFQ). The Strengths and Difficulties Questionnaire (SDQ) measured emotional symptoms at age 15, while the Center for Epidemiologic Studies Depression Scale (CESD) and SCARED measured depression and anxiety symptoms at age 18. At age 24, researchers used the revised CESD and the Hospital Anxiety and Depression Scale (HADS) to assess internalizing symptoms.

The team also linked CATSS data to clinical diagnoses and pharmacy-dispensed medication records from national registers. The team assessed clinical outcomes using national health records documenting first diagnoses of anxiety or depression during hospital or outpatient care, along with records of dispensed antidepressants, which served as an additional indicator of treatment. Antidepressant dispensing did not necessarily indicate a confirmed diagnosis of either disorder.

Using regression models, they examined associations between screen use and depressive or anxiety symptoms. Cox proportional hazards models assessed associations with subsequent clinically recorded depressive or anxiety disorders and/or antidepressant dispensing, providing estimates as adjusted hazard ratios (aHRs). Sensitivity analyses also accounted for physical activity and earlier bullying victimization where data were available.

Results

Of the 21,797 twins initially included, 73% completed the CATSS-15 questionnaire, 60% completed CATSS-18, and 27% completed CATSS-24. A recorded anxiety or depression diagnosis and/or antidepressant dispensing occurred in 11.5% of CATSS-9 participants by age 22 and 17.7% of CATSS-15 participants by age 28.

At age nine, 66% of participants watched television, videos, or DVDs daily. Boys were more likely to game daily than girls (38% vs. 23%). At ages 15 and 18, boys reported spending more time on interactive screen activities, including chatting, gaming, and web browsing, especially on weekends.

Among CATSS-15 participants, 23% of boys and 12% of girls spent over six hours daily on interactive screen activities during weekends. Heavy use was less common among adolescents with university-educated parents.

At ages 15 and 18, longer screen use was associated with higher internalizing symptom scores. Cross-sectional associations between screen use and internalizing symptoms were stronger for girls, particularly for interactive activities, and were attenuated in comparisons involving identical twins.

Screen use at age nine was not significantly associated with internalizing symptoms at age 15. Heavy interactive screen use among 15-year-olds was linked to more pronounced anxiety and depressive symptoms by age 18, while heavy usage at age 18 was associated with more depressive symptoms by age 24. These associations weakened in within-twin comparisons and were not statistically significant after correction for multiple testing.

Researchers also found that adolescents with more internalizing symptoms at age 15 reported slightly greater interactive screen use at age 18. Additional adjustments for physical activity and earlier bullying victimization made little difference to the findings involving later emotional symptoms.

Daily gaming at age nine, compared with gaming twice weekly or less, was associated with an increased risk of a recorded anxiety or depression diagnosis and/or antidepressant dispensing in the overall sample (aHR, 1.38) and among boys (aHR, 1.56). Yet these relationships weakened and were no longer significant in the within-twin analyses.

Watching television, videos, or DVDs showed no credible association with these outcomes. Among 15-year-olds, more than six hours per day of interactive screen activities on weekdays or weekends was linked to a higher risk of these recorded outcomes than two hours or less. 

The corresponding aHRs were 3.37 and 2.30, respectively. Within-twin estimates were smaller and no longer significant after correction for multiple testing. Analyses of later symptoms and clinical outcomes did not consistently show stronger associations in girls.

Conclusion

The findings link adolescent screen use to emotional difficulties and subsequent anxiety or depression outcomes, but the weaker associations in twin comparisons indicate that shared familial factors account for part of the relationship.

Data collection began before smartphones and social media became widespread, while broad screen-use categories could not distinguish specific online experiences. Self-reported screen time, declining participation, and residual confounding also limited causal interpretation. 

Future research should examine how specific screen activities, individual vulnerabilities and developmental differences relate to mental health before establishing more targeted recommendations for young people.