The relationship between screen-based behaviors and adolescent mental health: a phenotypic and genetic analysis

Evgeniia Frei, Piotr P. Jaholkowski, Pravesh Parekh, Oleksandr Frei, Alexey A. Shadrin, Nora Refsum Bakken, Viktoria Birkenæs, Helga Ask, Ole A. Andreassen, Olav B. Smeland
doi: https://doi.org/10.1101/2023.09.14.23295537

Abstract

Objective Technological advances made screen devices readily available, with adolescents spending an increasing amount of time on screens. Despite rising concerns, evidence on the relationship between screen-based activities and adolescent mental health remains inconsistent.

Method Using self-reports from the Norwegian Mother, Father, and Child Cohort Study and ICD-10 codes for mental illness from health registries, we estimated the odds of psychiatric diagnoses among adolescents across different levels of screen usage (n = 22,096). Additionally, we computed polygenic risk scores (PRSs) for schizophrenia (SCZ), bipolar disorder (BP), major depression, attention-deficit hyperactivity disorder (ADHD), autism spectrum disorder (ASD), and alcohol use disorder (AUD) for adolescents without a history of mental illness and evaluated their association with time spent on screen-based activities.

Results We observed higher odds of psychiatric diagnoses among individuals with most time spent across all screen-based behaviors. However, the lowest social media usage was also associated with mental illness (OR 2.18, 95% CI: 1.80, 2.65), particularly with increased ASD occurrence. Minimum time spent on gaming was associated with lower odds of psychiatric diagnoses (OR 0.74, 95% CI: 0.65, 0.84). PRSADHD was positively associated with time spent on social media and gaming, while PRSASD was positively associated with gaming, but negatively – with social media usage. Additionally, social media usage was significantly associated with lower PRSSCZ and PRSBP, but higher PRSAUD.

Conclusion We show that screen time behaviors are associated with mental illness among adolescents, which may partly reflect a shared genetic basis.

Introduction

The availability of screen-based devices continues to increase, and children and adolescents spend more time on screen-based activities than ever before.1 Digital media devices provide various easily accessible stimuli, and the time youth spend on screen-related activities has been increasing over the past decades.2 In Europe, 80% of youth either own a mobile or a smartphone.3 Similarly, around 95% of US adolescents own a mobile device.4 Globally, it is estimated that one in three children is an internet user, and that one in three users of the internet are under the age of 18.5

Sound mental health is of particular importance during adolescence, as this is a critical and formative period when individuals begin their transition to adulthood. Adolescents with mental health problems have lower quality of life, and are vulnerable to social exclusion, educational difficulties, and stigma.6 The prevalence of mental health problems among young people is increasing globally,7,8 and mental disorders are becoming major contributors to health-related disability in children and youth.9 Studies from the United States, the United Kingdom, and Norway have reported that around 1/10 to 1/5 of children and adolescents are diagnosed with a mental disorder.10–12 The causes of the increase in mental illness are not fully known.13

Several studies have suggested that the use of digital devices is associated with mental health problems. Excessive screen time may negatively affect young people’s brains both from a neurobiological perspective (for instance, some studies have reported structural reorganization of the brain in young people with internet gaming disorder),14,15 and from a lifestyle and social perspective such as physical activity, getting adequate sleep, in-person social interactions, and spending time in nature.16–18 However, study results are conflicting, and the link between mental illness and screen time use is yet to be clearly defined.19,20 For example, more screen time was moderately associated with worse mental health in children aged 9 to 10 years in the Adolescent Brain Cognitive DevelopmentSM Study (ABCD Study®), but the effect sizes were modest.21 Studies that investigated the link between digital media use and attention-deficit hyperactivity disorder (ADHD) support the connection between excessive screen time and ADHD symptoms.22 However, the associations are more consistent for problematic use of digital media than for screen time in general, and the complex relationships between digital media use and ADHD symptom levels are still poorly understood.23,24

Genetic factors contribute to psychiatric disorders in childhood and adolescence,25,26 with multiple common genetic variants linked to psychiatric disorders.13 It has been demonstrated that screen-based behaviours are heritable: for example, twin heritability of problematic internet use was estimated at 0.48.27 Another study reported single nucleotide polymorphism (SNP) heritability of television watching and leisure computer use as 0.16 and 0.09, respectively.28 However, the potential of shared genetic factors between screen time and mental health problems remains unexplored.

In the present study, we leveraged the Norwegian Mother, Father, and Child Cohort Study (MoBa),29 a prospective population-based pregnancy cohort, to investigate the phenotypic and genetic link between screen time and mental illness among adolescents (n = 22,096, 14-16 years of age). Specifically, we focused on activities in leisure time, including four screen-based behaviors: watching movies/series/TV; playing games (on PC, TV, tablet, mobile etc.); sitting/ lying down with PC, mobile or tablet (irrespective of activity); and communicating with friends on social media. First, we investigated whether the odds of having a psychiatric disorder among adolescents were associated with time spent on different screen-based behaviors. Next, we investigated if the screen-based behaviors were associated with the genetic liability to major psychiatric disorders, as indicated by polygenic risk scores (PRSs) derived from recent large-scale genome-wide association studies (GWASs) on schizophrenia (SCZ), bipolar disorder (BD), major depression (MD), autism spectrum disorder (ASD), ADHD, and alcohol use disorder (AUD).

Method

Study sample

MoBa is a population-based pregnancy cohort study conducted by the Norwegian Institute of Public Health.29 Participants were recruited from all over Norway from 1999-2008. The women consented to participation in 41% of the pregnancies. The cohort includes approximately 114,500 children, 95,200 mothers, and 75,200 fathers. The current study is based on version 12 of the quality-assured data files released for research in January 2019. The study sample includes all adolescents (14-16 years of age) who answered the MoBa questionnaire Q-14year (n = 22,096). The response rate for the Q-14year questionnaire was approximately 19.5%. Data of participants who withdrew their consent before March 2023 were not included in the analyses. The establishment of MoBa and initial data collection was based on a license from the Norwegian Data Protection Agency and approval from The Regional Committees for Medical and Health Research Ethics. The MoBa cohort is currently regulated by the Norwegian Health Registry Act. The current study was approved by The Regional Committees for Medical and Health Research Ethics (2016/1226/REK Sør-Øst C) in Norway.

Screen time use measures

We used the questions about activities in leisure time from the MoBa Q-14year questionnaire (Table 1). For each variable, participants gave themselves a score from 1 to 6 depending on the time spent on the specific screen-based activity.

View this table:
Table 1. Basic sociodemographic characteristics of the MoBa cohort at baseline, demographic characteristics of the adolescent sample, and descriptive information on key study variables

NA indicates missing values.

Psychiatric diagnoses

We retrieved information about psychiatric diagnoses among the MoBa participants who responded to the Q-14year questionnaire from the Norwegian Patient Registry (NPR). The retrieved information contained ICD-1030 codes received by the participants between 2008-2018. Diagnoses included organic, including symptomatic, mental disorders (F00-F09), mental and behavioral disorders due to psychoactive substance use (F10-F19), schizophrenia, schizotypal and delusional disorders (F20-F29), mood [affective] disorders (F30-F39), neurotic, stress-related and somatoform disorders (F40-F49), behavioral syndromes associated with physiological disturbances and physical factors (F50-F59), intellectual disability/mental retardation (F70-F79), disorders of psychological development (F80-F89), and behavioral and emotional disorders with onset usually occurring in childhood and adolescence (F90-F98).

Phenotypic Analysis

We applied logistic regression models to explore the association between the occurrence of psychiatric disorders in our sample and the time spent on each screen-based activity. We used a single outcome variable defined as the presence of any psychiatric disorder (see below) and tested it for association with a given screen time use score for a given screen-based activity (a predictor variable), leading to 24 models in total (one model for each of the four screen-based activity, and each of the 1-6 scores characterizing the time spent on each screen-based activity). The results are presented using odds ratios (OR), with OR above 1.0 indicating an increased risk of having a psychiatric diagnosis in a subgroup with a certain screen time use score. Age and sex assigned at birth were used as covariates in all models. For any pair of full siblings and maternal half-siblings (i. e. participants who had identical mother IDs), only one member was randomly retained in the analyses, which resulted in a sample of 20,581 individuals.

The analysis was performed by combining all psychiatric diagnoses, as well as for F90 (hyperkinetic disorders) and F84 (autism spectrum disorders) diagnostic categories, so that the outcome variable in the model was an indicator of a subject being diagnosed with any of these disorders. Original p-values for statistical significance are reported and compared to the family-wise error rate (α = 0.05/N, after Bonferroni correction for the number of tests). Analyses were done using Python ver. 3.7.4, and R ver. 4.0.3.

Genotyping of the MoBa cohort

Blood samples for the children were taken from the umbilical cord after their birth. Biological samples were sent to the Norwegian Institute of Public Health where DNA was extracted by standard methods and stored. Genotyping of the MoBa cohort was conducted through multiple research projects spanning several years, involving various selection criteria, and genotyping centres. Full details about the genotyping and QC procedure are provided elsewhere.31

Polygenic risk score (PRS) analyses

The sample for PRS analyses was restricted to participants without a history of a psychiatric disorder, based on the data from the NPR (i.e., we removed any subject that had any of the ICD-10 diagnoses codes specified above during the period 2008−2018). Subjects with ambiguous sex (based on genetics) and subjects with chromosomal abnormalities (as indicated in the Norwegian Medical Birth Registry – a national health registry containing information about all births in Norway) were removed from analyses. The analysis was restricted to the individuals of European ancestry, defined through comparison of the first seven genetic principal components (PCs) of the samples in the MoBa cohort with the equivalent PCs of the 1000 Genomes phase 1 samples, as described previously31. Further, for each related pair of study participants with a kinship coefficient greater than 0.05, one member was randomly excluded from analyses, which resulted in a sample of 12,980 unrelated individuals.

We applied PRSice (v2.3.3)32 to calculate PRSs of six psychiatric disorders using recent large-scale GWASs of SCZ, BP, MD, ASD, ADHD, and AUD.33–38 For each disorder, PRSs were calculated across several p-value thresholds (5e-08, 1e-06, 1e-05, 0.0001, 0.001, 0.01, 0.05, 0.1, 1). Then, following a widely applied method,39 the first principal component across all p-value thresholds was extracted and used to test for association with screen time phenotypes.

Linear regression models were used to test the association between time spent on each screen-based activity (Table 1) versus each PRS. We fit separate linear regression models for every questionnaire item and every PRS, resulting in 24 linear regression models in total (4 questionnaire items times 6 PRSs). Within each linear regression model, we included the first 10 genetic PCs and sex as covariates. In the association analysis, PRSs were standardized to zero mean and unit variance. As before, original p-values for statistical significance are reported and compared to the family-wise error rate using Bonferroni correction. Finally, we applied Kruskal-Wallis test to compare PRSs between the groups with different levels of screen time use.

Results

Sample characteristics

Descriptive information on key study variables and basic baseline sociodemographic characteristics of the sample are presented in Table 1. Gender identity is reported according to the participant’s replies. Detailed baseline sociodemographic characteristics of the MoBa cohort are provided in Table S1 (available online). Information about specific MoBa questions and variable names is provided in Table S2 (available online).

Of the 22,096 individuals (who responded to the Q-14years questionnaire), 2298 (10.40%) had at least one psychiatric diagnosis registered in the NPR between 2008 and 2018. The most frequent diagnostic subgroups were behavioral and emotional disorders with onset usually occurring in childhood and adolescence (F90-F98) (n = 1598, 7.23%), neurotic, stress-related and somatoform disorders (F40-F49) (n = 483, 2.19%), and disorders of psychological development (F80-F89) (n = 478, 2.16%). From these subgroups, the most frequent codes were F98 (n = 693, 3.14%) and F90 (n = 538, 2.43%), F43 (n = 182, 0.82%), and F84 (n = 191, 0.86%). Detailed information on the counts in each diagnostic subgroup is presented in Table S3 (available online).

Association between mental illness and time spent on screen-based activities

We evaluated the odds of having a psychiatric disorder within specific subgroups (defined by their screen time use score) after stratifying the sample by time spent on each screen-based activity (Figure 1 and Table 2). Given significant associations with ASD and ADHD in PRS analyses, we also performed separate phenotypic analyses for hyperkinetic disorders and autism spectrum disorders (Figures S1 and S2, Tables S3 and S4, available online).

Figure 1.
Figure 1. Odds ratios of having any psychiatric diagnosis among participants with different levels of screen-based activities

Odds ratios of having any psychiatric diagnosis among participants with different levels of screen-based activities were estimated using logistic regression models with age and sex as covariates (n = 20,581). Error bars indicate the 95% CIs for the estimated values. Asterisks indicate statistically significant results after comparison to the family-wise error rate (α = 0.002, after Bonferroni correction for the number of tests). The horizontal line indicates no association (OR = 1).

View this table:
Table 2. Odds ratios of having any psychiatric diagnosis among participants with different levels of screen-based activities

Odds ratios of having any psychiatric diagnosis among participants with different levels of screen-based activities were estimated using logistic regression models with age and sex as covariates (n = 20,581). Original p-values are reported. Asterisks indicate statistically significant results after comparison to family-wise error rate (α = 0.002, after Bonferroni correction for the number of tests).

For movie/TV watching, we observed lower odds of having any mental illness among participants who spent less than 1 h/day on this activity (OR 0.81, 95% CI: 0.73, 0.90). On the other hand, the subgroups of adolescents who spent more time watching TV/movies, had a significantly higher odds of having a mental illness. Specifically, the subgroup that spent 3-4 h/day, had OR = 1.29 (95% CI: 1.15, 1.45), while the subgroup who spent ≥ 7 h/day, had OR = 1.96 (95% CI: 1.41, 2.72).

Time spent on gaming had a similar association with mental illness occurrence. The two groups of adolescents who spent the lowest amount of time gaming (i.e., never/rarely or < 1 h/day) had significantly lower odds of having a psychiatric diagnosis: 0.74 (95% CI: 0.65, 0.84) and 0.82 (95% CI: 0.73, 0.92), respectively. Conversely, individuals within the groups who gamed 3-4 h/day, 5-6 h/day, and ≥ 7 h/day, had significantly higher odds of having a mental illness: 1.31 (95% CI: 1.17, 1.47), 1.58 (95% CI: 1.34, 1.87), and 2.19 (95% CI: 1.74, 2.75), respectively. We then evaluated differences across specific diagnostic codes. Participants who gamed never/rarely had significantly lower odds of having ADHD (OR 0.43, 95% CI: 0.31, 0.60) and ASD (OR 0.66, 95% CI: 0.37-1.18). In comparison, the odds of having ADHD and ASD were significantly higher for individuals who spent most time gaming (≥ 7 h/day): 2.91 (95% CI: 2.22, 3.82) and 3.72 (95% CI: 2.30, 6.02), respectively.

For unspecific screen-based activity (sitting/lying down with PC, mobile, or tablet), we observed a decrease in the occurrence of mental illness in the group that used screen devices for 1-2 h/day (OR 0.84, 95% CI: 0.76, 0.92). Individuals within the two subgroups who spent the most time sitting/lying with a screen device (i. e. 5-6 h/day and ≥ 7 h/day), had significantly higher odds of having a mental illness: 1.29 (95% CI: 1.14, 1.47), and 2.05 (95% CI: 1.74, 2.41), respectively. The higher occurrence of mental illness among participants who spent ≥ 7 h/day with a screen device was primarily due to larger proportions of ADHD and ASD cases. Odds ratios for having these disorders were 2.91 (95% CI: 2.22, 3.82) and 4.42 (95% CI: 3.00, 6.51), respectively.

For social media communication, we observed an L-shaped relationship between time spent on this activity and the odds of having a mental illness. Adolescents with both the least and the most time spent had significantly higher odds of having a psychiatric diagnosis: OR 2.07 (95% CI: 1.45, 2.95), and OR 2.39 (1.81, 3.16), respectively. Participants who never/rarely used social media had significantly higher odds of having ADHD (OR 2.07, 95% CI: 1.45, 2.95) and ASD (OR 7.45, 95% CI: 5.15, 10.77). In comparison, individuals with medium time spent (1-2 h/day) had lower odds of having a mental illness in general (OR 0.82, 95% CI: 0.75, 0.91), as well as ADHD (OR 0.62, 95% CI: 0.50, 0.77) specifically.

Association between time spent on screen-based activities and PRSs for major psychiatric disorders

We identified several statistically significant associations between screen time use behaviors and PRSs for major psychiatric disorders, after correcting for multiple comparisons (Figure 2). The time spent watching movies/TV was positively associated with PRS for ADHD (β = 0.035; 95% CI: 0.019, 0.052; p = 3.70e-05). For time spent on gaming, we found a positive association with PRSs for ADHD (β = 0.026; 95% CI: 0.0078, 0.0044; p = 5.29e-03) and for ASD (β = 0.052; 95% CI: 0.034, 0.070; p = 2.73e-08). Moreover, time spent on unspecific screen-based activity was positively associated with PRS for ASD (β = 0.029; 95% CI: 0.011, 0.047; p = 1.75e-03). Finally, time spent on social media communication was negatively associated with PRSs for SCZ (β = −0.036; 95% CI: −0.056, −0.016; p = 4.37e-04), BP (β = −0.030; 95% CI: − 0.050, −0.0095; p = 3.95e-03), and ASD (β = −0.031; 95% CI: −0.051, −0.011; p = 2.71e-03).

Figure 2.
Figure 2. Association between time spent on screen-based activities and polygenic risk scores for six major psychiatric disorders

Association between time spent on screen-based activities and polygenic risk scores for schizophrenia (SCZ), bipolar disorder (BP), major depression (MD), autism spectrum disorder (ASD), attention-deficit hyperactivity disorder (ADHD), and alcohol use disorder (AUD), derived from linear regression models. Error bars indicate the 95% CIs of the estimated regression coefficients. The vertical lines indicate an effect estimate equal to zero.

Conversely, PRSs for ADHD and AUD were positively associated with time spent on social media communication (β = 0.059; 95% CI: 0.039, 0.079; p = 8.36e-09, and β = 0.030; 95% CI: 0.010, 0.050; p = 3.73e-03 respectively).

Further analysis revealed several significant dose-dependent associations between hours spent communicating with friends on social media and higher PRSs for ADHD and AUD, as well as lower PRSs for SCZ and ASD. Moreover, we observed significant dose-dependent associations between higher PRS for ASD and time spent gaming, along with time spent on unspecific screen activity. Detailed results are presented in Figure S3 (available online).

Discussion

In the present study, we provide new insights into the relationship between screen-based behaviors and mental illness among adolescents, a widely debated issue in society today. Leveraging one of the largest birth cohorts in the world, as well as large-scale psychiatric genetics summary data from international consortia, we investigated this relationship from both a phenotypic and genetic perspective. Overall, we observe that the occurrence of mental illness, and ADHD and ASD in particular, was significantly higher among adolescents on both extremes of screen time use. We further performed genetic analyses, demonstrating that the polygenic risk of major psychiatric disorders, most strongly ADHD and ASD, was significantly associated with several screen-based behaviors. Altogether, the study demonstrates that screen-based behaviors are associated with mental illness among adolescents, which may partly reflect a shared genetic basis.

We found clear patterns of association between psychiatric diagnoses and time spent on screen-based activities (Figure 1, Table 2). Odds of having a psychiatric diagnosis were significantly higher within the groups who spent the most time on each of the four screen-based behaviors assessed. Taken together, the results imply that excessive screen time is associated with poorer mental health among adolescents. However, we also observed a higher occurrence of mental illness among those with minimum time spent on social media, who had around twice the odds of having a mental illness. This may not be unexpected given that social media represent a powerful arena for socialization and network building for young people,40 where too much or too little interaction may reflect difficulties in social skills or communication preferences, which may be linked to mental illness. Thes results can also be thought of in the context that adolescents with mental illness are left out of these social arenas, which are not mutually exclusive hypotheses. Overall, the findings indicate that better mental health among adolescents are linked to neither of the extremes for time spent on these screen-based activities.

For gaming, the odds of having a psychiatric diagnosis were almost twice as high among adolescents who spent more than 7 h/day on this activity. Importantly, we demonstrate that the increased occurrence of psychiatric disorders appears to be driven by higher proportion of ADHD and ASD cases (Tables S3 and S4, Figures S1 and S2, available online). Although some studies have linked mental health problems and problematic game use,41 the causal relationships are not yet established. Moreover, the concept “problematic game use” has its own controversies (for example, lack of unified criteria).42 The relationship between gaming, the consequences of its abuse, and associated mental health problems is complex, and might be influenced by many factors, warranting more studies. We also observed twofold odds of having a psychiatric diagnosis among participants who spent ≥ 7 h/day on unspecific screen-based activities (sitting/lying down with PC, mobile, or tablet). High levels of sedentary behavior are linked to poorer lifestyle choices and worse health outcomes, and it has also been shown that physical activity is associated with better mental health among children and adolescents.43 Thus, excessive screen time in general may affect mental health indirectly, by limiting potential benefits of physical activity and reducing sleep duration and quality. However, the possibility of reverse causation is also likely, as excessive screen time may be a coping strategy for individuals who already suffer from mental illness.

In general, the positive association between excessive screen time and poorer mental health is in line with previous research,21 though in many preceding studies associations were weak, and effect sizes were small.20 The slight discrepancies could be partially explained by participant’s age: gaming and online social activity is steadily increasing over ages 9-12.44 Accordingly, data for various age groups could be difficult to compare. Moreover, methods for screen-time assessment are limited, with no standardized instruments available to estimate how and when different modern devices are used along with traditional screen-based devices, as well as the variety of ways that different types of screens could be used, all of which might be associated with different types of screen-based behaviors and health outcomes. Nevertheless, the questions about screen-time in MoBa are formulated broadly enough to include most types of screen-based devices adolescents use, and to cover the majority of screen-based behaviors. The six different levels of screen time usage analyzed here, with the maximum label ≥ 7 h/day, allows for examining both the high and low extremes in screen time, which are evidently the most strongly associated with mental health problems. It is also important to mention that the ICD-10 diagnostic codes, obtained from the NPR, represent more valid and reliable outcome measures than self-reports and parent evaluations, which can be considered as another strength of our study.

Furthermore, we show that genetic risk of major psychiatric disorders, as indexed by PRSs, was differently associated with screen-based behaviors (Figure 2). Importantly, the associations were independent of a history of mental illness, as the analysis was restricted to individuals without any psychiatric diagnoses, although we cannot exclude the possibility that any of the participants would develop mental illness later on. In particular, genetic risk for ADHD was associated with more time spent on watching TV/movies, gaming and social media. These findings are in line with phenotypic analysis results: the occurrence of ADHD was significantly higher among participants who spent ≥ 3 h/day on gaming, and ≥ 7 h/day on social media (Table S4 and Figure S1, available online). Overall, our results indicate that adolescents with attention deficits are more prone to spend more time on their screens, and that children and adolescents with ADHD may be particularly susceptible to problematic media use. Many studies have reported associations between exposure to screen-based devices and attention in children and adolescents,45 although the underlying mechanisms linking attention skills and screen time are not well established. It has been suggested that ADHD and problematic gaming may share a common psychobiological mechanism of reward and sensitization, mainly mediated by dopamine activity.46 Other neurotransmitters like serotonin, glutamate, or GABA, might play an important role too – further studies are needed to illuminate their potentially shared biological mechanisms.

Conversely to ADHD PRS, ASD PRS was positively associated with the gaming score, but negatively with social media use. These findings are also in line with phenotypic analysis results, where we observed almost four times higher odds of having an ASD diagnosis among participants who spent most time on gaming (OR 3.72), and about seven times higher odds of having an ASD diagnosis among adolescents who used social media never/rarely (OR 7.45) (Table S5 and Figure S2, available online). The present findings are in compliance with the social difficulties associated with ASD. Moreover, individuals with ASD may be more vulnerable to addictive behaviors, including internet or gaming addiction.47 The positive association between AUD PRS and social media use is in line with the hypothesis on the addictive potential of social media,48 which is especially important for the target population as adolescence is known as a vulnerable stage for development of addictive behaviors. It has been suggested that social media usage is involved in the control of brain inhibitory mechanisms,49 though an effect of prolonged social media usage on neural structures and activity have yet to be investigated.

Generally, in our study, the results of phenotypic and genetic analyses were consistent with each other, indicating that PRSs can be used as a proxy of an actual diagnosis in studying associations with psychiatric disorders. However, due to the young study population, we did not have enough cases of SCZ and BP to perform statistically valid phenotype analysis. Nevertheless, the PRS analyses results, computed for adolescents without any psychiatric disorders, indicate significant negative associations with social media use for both the SCZ and BP PRSs. Therefore, as the study population becomes older, it is important to perform a follow-up analysis for these diagnoses. The pattern of results between PRS and phenotype analyses was also different for ADHD and unspecific screen-based activity. This might be attributed to the limited accuracy of the PRS for ADHD, although the ADHD PRS results were consistent with the phenotype analyses for other screen-based behaviors. Future studies based on more well-powered ADHD GWAS are needed to address this further.

Our study is not without limitations. Despite well-formulated questions and fine-grained response categories, we are unable to objectively assess screen usage with self-report data. Additionally, these questions do not capture different aspects of the screen-based activities such as active vs. passive engagement or the social aspects of gaming, etc. Further, selection bias is a major challenge in cohort studies, and MoBa participants were found to not be representative of the entire Norwegian population.50 Although the contributions of MoBa are still highly valid, these issues should be taken into account when interpreting the results of the present study.

In conclusion, our study demonstrates that screen-based behaviors and mental illness are both phenotypically and genetically linked among adolescents in a large population cohort. For three of the screen-based activities, better mental health was linked to medium time usage, while for gaming, minimum time spent was associated with the lowest occurrence of mental illness. The results were diagnosis-specific, with ADHD driving the associations between higher occurrence of psychiatric diagnoses and maximum time spent on gaming, unspecific screen-based activities, and social media communication. Conversely, the occurrence of ASD was significantly higher among individuals who spent the least time on social media. Although we cannot conclude on the causal relationship underlying these associations, the PRS analyses indicate a shared genetic basis between screen behaviors and mental illness. Accordingly, adolescents with increased genetic risk of a given mental disorder may be more susceptible to display more extreme screen-based behaviors. Overall, our results provide new insights into the relationship between screen time use and mental health problems during adolescence.

Data Availability

All use of data and biological material from MoBa is subject to Norwegian law. Information about the requirements needed to gain access to data from MoBa is provided here: https://www.fhi.no/en/ch/studies/moba/for-forskere-artikler/research-and-data-access/

Disclosure statement

Ole A. Andreassen is a consultant to HealthLytics and has received speaker’s honorarium from Lundbeck and Sunovion. No potential conflict of interest was reported by other authors.

Acknowledgements and Funding

The Norwegian Mother, Father and Child Cohort Study is supported by the Norwegian Ministry of Health and Care Services and the Ministry of Education and Research. We are grateful to all the participating families in Norway who take part in this on-going cohort study. We thank the Norwegian Institute of Public Health (NIPH) for generating high-quality genomic data. This research is part of the HARVEST collaboration, supported by the Research Council of Norway (#229624). We also thank the NORMENT Centre for providing genotype data, funded by the Research Council of Norway (#223273), South East Norway Health Authorities and Stiftelsen Kristian Gerhard Jebsen. We further thank the Center for Diabetes Research, the University of Bergen for providing genotype data and performing quality control and imputation of the data funded by the ERC AdG project SELECTionPREDISPOSED, Stiftelsen Kristian Gerhard Jebsen, Trond Mohn Foundation, the Research Council of Norway, the Novo Nordisk Foundation, the University of Bergen, and the Western Norway Health Authorities. We also thank deCODE genetics for genotyping of the main part of the MoBa sample.

This work was supported by the Research Council of Norway (grants 324499, 326813, 324252, 271555/F21, 274611 MoBaGen, 324620 C-Me), The European Economic Area and Norway Grants (grants EEA-RO-NO-2018-0535, EEA-RO-NO-2018-0573), South-Eastern Norway Regional Health Authority (grant 2022072), European Union’s Horizon 2020 Research and Innovation Programme Under the Marie Skłodowska-Curie Grant (grant 801133), European Union’s Horizon 2020 Research and Innovation Programme (grant 964874 RealMent, grant 847776 CoMorMent).

This work was performed on Services for sensitive data (TSD), University of Oslo, Norway, with resources provided by UNINETT Sigma2 − the National Infrastructure for High Performance Computing and Data Storage in Norway.

References

  1. 1.↵
    Bucksch J, Sigmundova D, Hamrik Z, et al. International Trends in Adolescent Screen-Time Behaviors from 2002 to 2010. J Adolesc Heal. 2016;58(4):417–425. doi:10.1016/j.jadohealth.2015.11.014
    OpenUrlCrossRef
  2. 2.↵
    2014 Global Summit on the Physical Activity of Children. Journal of Physical Activity and Health. Published 2014. Accessed August 21, 2023. https://journals.humankinetics.com/view/journals/jpah/11/s1/jpah.11.issue-s1.xml
  3. 3.↵
    Mascheroni G, Ólafsson K. Net Children Go Mobile: Risks and Opportunities. Second Edition. Milano: Educatt.; 2014. Accessed April 24, 2023. www.netchildrengomobile.eu
  4. 4.↵
    Rideout V, Robb MB. Social Media, Social Life: Teens Reveal Their Experiences.; 2018.
  5. 5.↵
    Keeley B, Little C. State of the Worlds Children 2017: Children in a Digital World. UNICEF. 3 United Nations Plaza, New York, NY 10017. Tel: 212-326-7000; Fax: 212-887-7465; Web site: http://www.unicef.org/education; 2017. Accessed April 24, 2023. www.soapbox.co.uk
  6. 6.↵
    Celebre A, Stewart SL, Theall L, Lapshina N. An Examination of Correlates of Quality of Life in Children and Youth With Mental Health Issues. Front Psychiatry. 2021;12:1518. doi:10.3389/fpsyt.2021.709516
    OpenUrlCrossRef
  7. 7.↵
    Bor W, Dean AJ, Najman J, Hayatbakhsh R. Are child and adolescent mental health problems increasing in the 21st century? A systematic review. Aust N Z J Psychiatry. 2014;48(7):606–616. doi:10.1177/0004867414533834
    OpenUrlCrossRefPubMed
  8. 8.↵
    Collishaw S. Annual research review: Secular trends in child and adolescent mental health. J Child Psychol Psychiatry. 2015;56(3):370–393. doi:10.1111/jcpp.12372
    OpenUrlCrossRefPubMed
  9. 9.↵
    Gore FM, Bloem PJN, Patton GC, et al. Global burden of disease in young people aged 10-24 years: A systematic analysis. Lancet. 2011;377(9783):2093-2102. doi:10.1016/S0140-6736(11)60512-6
    OpenUrlCrossRefPubMedWeb of Science
  10. 10.↵
    Heiervang E, Stormark KM, Lundervold AJ, et al. Psychiatric disorders in Norwegian 8- to 10-year-olds: An epidemiological survey of prevalence, risk factors, and service use. J Am Acad Child Adolesc Psychiatry. 2007;46(4):438–447. doi:10.1097/chi.0b013e31803062bf
    OpenUrlCrossRefPubMedWeb of Science
  11. 11.
    Ford T, Goodman R, Meltzer H. The British child and adolescent mental health survey 1999: The prevalence of DSM-IV disorders. J Am Acad Child Adolesc Psychiatry. 2003;42(10):1203–1211. doi:10.1097/00004583-200310000-00011
    OpenUrlCrossRefPubMedWeb of Science
  12. 12.↵
    Kessler RC, Avenevoli S, Costello EJ, et al. Prevalence, persistence, and sociodemographic correlates of DSM-IV disorders in the National Comorbidity Survey Replication Adolescent Supplement. Arch Gen Psychiatry. 2012;69(4):372–380. doi:10.1001/archgenpsychiatry.2011.160
    OpenUrlCrossRefPubMedWeb of Science
  13. 13.↵
    Andreassen OA, Hindley GFL, Frei O, Smeland OB. New insights from the last decade of research in psychiatric genetics: discoveries, challenges and clinical implications. World Psychiatry. 2023;22(1):4–24. doi:10.1002/wps.21034
    OpenUrlCrossRef
  14. 14.↵
    Pan N, Yang Y, Du X, et al. Brain structures associated with internet addiction tendency in adolescent online game players. Front Psychiatry. 2018;9(MAR):6. doi:10.3389/fpsyt.2018.00067
    OpenUrlCrossRef
  15. 15.↵
    Weng CB, Qian RB, Fu XM, et al. Gray matter and white matter abnormalities in online game addiction. Eur J Radiol. 2013;82(8):1308–1312. doi:10.1016/j.ejrad.2013.01.031
    OpenUrlCrossRefPubMed
  16. 16.↵
    Oswald TK, Rumbold AR, Kedzior SGE, Moore VM. Psychological impacts of “screen time” and “green time” for children and adolescents: A systematic scoping review. PLoS One. 2020;15(9 september). doi:10.1371/journal.pone.0237725
    OpenUrlCrossRef
  17. 17.
    Twenge JM, Spitzberg BH, Campbell WK. Less in-person social interaction with peers among U.S. adolescents in the 21st century and links to loneliness. J Soc Pers Relat. 2019;36(6):1892–1913. doi:10.1177/0265407519836170
    OpenUrlCrossRef
  18. 18.↵
    LeBourgeois MK, Hale L, Chang AM, Akacem LD, Montgomery-Downs HE, Buxton OM. Digital media and sleep in childhood and adolescence. Pediatrics. 2017;140(Suppl 2):S92–S96. doi:10.1542/peds.2016-1758J
    OpenUrlCrossRefPubMed
  19. 19.↵
    Stiglic N, Viner RM. Effects of screentime on the health and well-being of children and adolescents: A systematic review of reviews. BMJ Open. 2019;9(1):e023191. doi:10.1136/bmjopen-2018-023191
    OpenUrlAbstract/FREE Full Text
  20. 20.↵
    Tang S, Werner-Seidler A, Torok M, Mackinnon AJ, Christensen H. The relationship between screen time and mental health in young people: A systematic review of longitudinal studies. Clin Psychol Rev. 2021;86. doi:10.1016/j.cpr.2021.102021
    OpenUrlCrossRef
  21. 21.↵
    Paulich KN, Ross JM, Lessem JM, Hewitt JK. Screen time and early adolescent mental health, academic, and social outcomes in 9- And 10-year old children: Utilizing the Adolescent Brain Cognitive DevelopmentSM (ABCD) Study. PLoS One. 2021;16(9 September):e0256591. doi:10.1371/journal.pone.0256591
    OpenUrlCrossRef
  22. 22.↵
    Thorell LB, Burén J, Ström Wiman J, Sandberg D, Nutley SB. Longitudinal associations between digital media use and ADHD symptoms in children and adolescents: a systematic literature review. Eur Child Adolesc Psychiatry. 2022;1:1–24. doi:10.1007/s00787-022-02130-3
    OpenUrlCrossRef
  23. 23.↵
    Engelhard MM, Kollins SH. The Many Channels of Screen Media Technology in ADHD: a Paradigm for Quantifying Distinct Risks and Potential Benefits. Curr Psychiatry Rep. 2019;21(9). doi:10.1007/s11920-019-1077-1
    OpenUrlCrossRef
  24. 24.↵
    Beyens I, Valkenburg PM, Piotrowski JT. Screen media use and ADHD-related behaviors: Four decades of research. Proc Natl Acad Sci U S A. 2018;115(40):9875–9881. doi:10.1073/pnas.1611611114
    OpenUrlAbstract/FREE Full Text
  25. 25.↵
    Manoli DS, State MW. Autism spectrum disorder genetics and the search for pathological mechanisms. Am J Psychiatry. 2021;178(1):30–38. doi:10.1176/appi.ajp.2020.20111608
    OpenUrlCrossRef
  26. 26.↵
    Faraone S V., Larsson H. Genetics of attention deficit hyperactivity disorder. Mol Psychiatry. 2019;24(4):562–575. doi:10.1038/s41380-018-0070-0
    OpenUrlCrossRefPubMed
  27. 27.↵
    Vink JM, Van Beijsterveldt TCEM, Huppertz C, Bartels M, Boomsma DI. Heritability of compulsive Internet use in adolescents. Addict Biol. 2016;21(2):460–468. doi:10.1111/adb.12218
    OpenUrlCrossRef
  28. 28.↵
    van de Vegte YJ, Said MA, Rienstra M, van der Harst P, Verweij N. Genome-wide association studies and Mendelian randomization analyses for leisure sedentary behaviours. Nat Commun. 2020;11(1):1–10. doi:10.1038/s41467-020-15553-w
    OpenUrlCrossRefPubMed
  29. 29.↵
    Magnus P, Birke C, Vejrup K, et al. Cohort Profile Update: The Norwegian Mother and Child Cohort Study (MoBa). Int J Epidemiol. 2016;45(2):382–388. doi:10.1093/IJE/DYW029
    OpenUrlCrossRefPubMed
  30. 30.↵
    ICD-10. International Statistical Classification of Diseases and Related Health Problems 10th Revision. Published 2019. Accessed August 10, 2023. www.who.int
  31. 31.↵
    Corfield EC, Frei O, Shadrin AA, et al. The Norwegian Mother, Father, and Child cohort study (MoBa) genotyping data resource: MoBaPsychGen pipeline v.1. bioRxiv. Published online September 6, 2022:2022.06.23.496289. doi:10.1101/2022.06.23.496289
    OpenUrlAbstract/FREE Full Text
  32. 32.↵
    Choi SW, O’Reilly PF. PRSice-2: Polygenic Risk Score software for biobank-scale data. Gigascience. 2019;8(7):1–6. doi:10.1093/gigascience/giz082
    OpenUrlCrossRefPubMed
  33. 33.↵
    Grove J, Ripke S, Als TD, et al. Identification of common genetic risk variants for autism spectrum disorder. Nat Genet. 2019;51(3):431–444. doi:10.1038/s41588-019-0344-8
    OpenUrlCrossRefPubMed
  34. 34.
    Trubetskoy V, Pardiñas AF, Qi T, et al. Mapping genomic loci implicates genes and synaptic biology in schizophrenia. Nature. 2022;604(7906):502-508. doi:10.1038/s41586-022-04434-5
    OpenUrlCrossRefPubMed
  35. 35.
    Demontis D, Walters RK, Martin J, et al. Discovery of the first genome-wide significant risk loci for attention deficit/hyperactivity disorder. Nat Genet. 2019;51(1):63–75. doi:10.1038/s41588-018-0269-7
    OpenUrlCrossRefPubMed
  36. 36.
    Mullins N, Forstner AJ, O’Connell KS, et al. Genome-wide association study of over 40,000 bipolar disorder cases provides new insights into the underlying biology. Nat Genet. 2021;53(6):817–829. doi:10.1038/S41588-021-00857-4
    OpenUrlCrossRefPubMed
  37. 37.
    Howard DM, Adams MJ, Clarke TK, et al. Genome-wide meta-analysis of depression identifies 102 independent variants and highlights the importance of the prefrontal brain regions. Nat Neurosci. 2019;22(3):343–352. doi:10.1038/s41593-018-0326-7
    OpenUrlCrossRefPubMed
  38. 38.↵
    Zhou H, Sealock JM, Sanchez-Roige S, et al. Genome-wide meta-analysis of problematic alcohol use in 435,563 individuals yields insights into biology and relationships with other traits. Nat Neurosci. 2020;23(7):809–818. doi:10.1038/s41593-020-0643-5
    OpenUrlCrossRefPubMed
  39. 39.↵
    Coombes BJ, Ploner A, Bergen SE, Biernacka JM. A principal component approach to improve association testing with polygenic risk scores. Genet Epidemiol. 2020;44(7):676–686. doi:10.1002/gepi.22339
    OpenUrlCrossRef
  40. 40.↵
    Armstrong-Carter E, Telzer EH. Advancing Measurement and Research on Youths’ Prosocial Behavior in the Digital Age. Child Dev Perspect. 2021;15(1):31–36. doi:10.1111/cdep.12396
    OpenUrlCrossRef
  41. 41.↵
    González-Bueso V, Santamaría JJ, Fernández D, Merino L, Montero E, Ribas J. Association between internet gaming disorder or pathological video-game use and comorbid psychopathology: A comprehensive review. Int J Environ Res Public Health. 2018;15(4). doi:10.3390/ijerph15040668
    OpenUrlCrossRefPubMed
  42. 42.↵
    Przybylski AK, Weinstein N, Murayama K. Internet gaming disorder: Investigating the clinical relevance of a new phenomenon. Am J Psychiatry. 2017;174(3):230–235. doi:10.1176/appi.ajp.2016.16020224
    OpenUrlCrossRef
  43. 43.↵
    Sampasa-Kanyinga H, Sampasa-Kanyinga H, Colman I, et al. Combinations of physical activity, sedentary time, and sleep duration and their associations with depressive symptoms and other mental health problems in children and adolescents: A systematic review. Int J Behav Nutr Phys Act. 2020;17(1):1–16. doi:10.1186/s12966-020-00976-x
    OpenUrlCrossRefPubMed
  44. 44.↵
    Bagot KS, Tomko RL, Marshall AT, et al. Youth screen use in the ABCD® study. Dev Cogn Neurosci. 2022;57:101150. doi:10.1016/j.dcn.2022.101150
    OpenUrlCrossRef
  45. 45.↵
    Santos RMS, Mendes CG, Marques Miranda D, Romano-Silva MA. The Association between Screen Time and Attention in Children: A Systematic Review. Dev Neuropsychol. 2022;47(4):175–192. doi:10.1080/87565641.2022.2064863
    OpenUrlCrossRef
  46. 46.↵
    Weinstein A, Weizman A. Emerging association between addictive gaming and attention-deficit/ hyperactivity disorder. Curr Psychiatry Rep. 2012;14(5):590–597. doi:10.1007/s11920-012-0311-x
    OpenUrlCrossRefPubMedWeb of Science
  47. 47.↵
    Kervin R, Berger C, Moon SJ, Hill H, Park D, Kim JW. Behavioral addiction and autism spectrum disorder: A systematic review. Res Dev Disabil. 2021;117:104033. doi:10.1016/j.ridd.2021.104033
    OpenUrlCrossRef
  48. 48.↵
    Cheng C, Lau YC, Chan L, Luk JW. Prevalence of social media addiction across 32 nations: Meta-analysis with subgroup analysis of classification schemes and cultural values. Addict Behav. 2021;117:106845. doi:10.1016/j.addbeh.2021.106845
    OpenUrlCrossRef
  49. 49.↵
    Cataldo I, Lepri B, Neoh MJY, Esposito G. Social Media Usage and Development of Psychiatric Disorders in Childhood and Adolescence: A Review. Front Psychiatry. 2021;11:508595. doi:10.3389/fpsyt.2020.508595
    OpenUrlCrossRef
  50. 50.↵
    Nilsen RM, Vollset SE, Gjessing HK, et al. Self-selection and bias in a large prospective pregnancy cohort in Norway. Paediatr Perinat Epidemiol. 2009;23(6):597–608. doi:10.1111/J.1365-3016.2009.01062.X
    OpenUrlCrossRefPubMedWeb of Science
Back to top
Posted September 14, 2023.
Download PDF

Supplementary Material

Data/Code
Email
The relationship between screen-based behaviors and adolescent mental health: a phenotypic and genetic analysis
Evgeniia Frei, Piotr P. Jaholkowski, Pravesh Parekh, Oleksandr Frei, Alexey A. Shadrin, Nora Refsum Bakken, Viktoria Birkenæs, Helga Ask, Ole A. Andreassen, Olav B. Smeland
medRxiv 2023.09.14.23295537; doi: https://doi.org/10.1101/2023.09.14.23295537
Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area