Table 1 Descriptive characteristics of the included studies

Chen et al. [22]

China

Cross-Sectional

N: 1,331

M Age: 7–9 th grades

48,69% girls

To investigate the associations between different types of (screen-based sedentary behavior - SSB) and anxiety symptoms.

Anxiety symptoms: Zung Self-Rating Anxiety Scale

Self-reported. SSB was categorized into television/movie time, video game time, and internet-surfing time

Video game time of 6 or more hours was positively associated with boys’ anxiety symptoms (p < .05), with a large effect size (OR = 5.12, 95% CI: 1.56–17.44). There were no associations between different types of SSBs and anxiety symptoms in girls.

Forte et al. [23]

Ireland

Cross-Sectional

N:1,756

M Age: 15,2

56,7% girls

To investigate the associations between

screen time (ST) and physical activity (PA) level with depressive symptoms.

Depressive Symptoms

Quick Inventory of

Depressive

Symptomatology

Self-reported weekly ST (TV, computer (PC), and phone use)

The most significant correlations with depressive symptoms were shown by higher levels of computer (β = 0.106, p ≤ .000) and phone use (β = 0.138, p ≤ .000), with a small effect size. Only correlations between phone use and depressive symptoms were moderated by PA level.

Kandola et al. [24]

UK

Longitudinal

N: 4,599

M Age: 14,0

55%girls

To investigate how theoretically replacing different screen-uses with exercise might influence future adolescent emotional distress.

The outcome was emotional distress at age 17: Strengths and Difficulties Questionnaire, emotional symptoms subscale

Self-reported. Daily time-use variables were recorded by 24-hour time-use diaries completed over two randomly selected days, one in the week and the other at the weekend.

Substituting 60 min of television or social media use with team sports was associated with a decrement of 0.17 (95%CI, -0.31, -0.04) and 0.15 (95%CI, -0.29, -0.01) in emotional symptom scores, at 17, respectively. Small effect size.

Kjellenberg et al.

[25]

Sweden

Cross-Sectional

N: 1,139

M Age: 13,4

51% girls

To investigate the associations between physical activity pattern, sports participation, screen time and mental health.

Anxiety and health-related quality of life (HRQoL): short version of the Spence Children’s Anxiety Scale and Kidscreen-10.

Self-reported: time spent with screens on a weekday and weekend, not including schoolwork.

Only the group of girls who reported ≥ 5 h ST on weekdays or weekends has shown significant associations. These groups had considerably higher anxiety rates compared with those who reported up to 2 h. When controlling for MVPA, this comparison held significant (B = 3.39, 95% CI 1.33 to 5.46). Large effect size.

Kidokoro et al.

[26]

Japan

Cross-Sectional

N: 7,847

M Age: 14,0

53% girls

To investigate

associations between different types of screen behavior and depression, considering sleep and exercise.

Depression symptoms were measured using a modified version of the depression questionnaire developed by the American Psychiatric Association

Self-reported: time spent on recreational (i.e., screen behavior outside of school) screen, in a week.

Longer time spent on newer sorts of screen behavior, including social media, and online games (among junior high school girls), was associated with a higher prevalence of depression, with a small effect size. Longer time was spent on TV correlation with a lower prevalence of depression

Marciano et al. [27]

Switzerland

Longitudinal

N: 674

M Age: 14,4

56,7% girls

To investigate how changes in screen-media use from Spring 2019 (T1) influenced adolescents’ mental health in (T2) Autumn 2020

Mental health: adapted version of the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5). Loneliness: 3-item version of the UCLA Loneliness Scale. Depression: seven items from the Center for Epidemiologic Studies Depression Scale.

Self-reported: Screen-media activities were measured at T1 and T2 as time spent in different online activities on “a typical school day” and

on “a typical weekend day”

Worse mental health at T2 has only been significantly associated with increased time spent on social media (β = 0.112, p = .016), with medium effect size. Longer time dedicated to structured media activities as television viewing decreased rates of inattention (β = −0.091, p = .021) and anxiety (β = −0.093, p = .014).

Khan et al. [28]

42 European and North American countries.

Cross-sectional

N: 577,475

M Age: 13.6

51.35% girls

To examine the associations between the combination of ST and Physical Activity (PA) with mental well-being.

Satisfaction with life: Cantril Ladder Psychosomatic complaints: feeling low; irritability or bad mood; nervousness; difficulty falling asleep; dizziness; headaches; stomachaches; and back pain.

Self-reported use of all screen types and all purposes, excluding motion and fitness games.

In a dose-dependent manner, there was negative association between ST levels and life satisfaction and positive association between ST and psychosomatic complaints. Detrimental correlations between ST and mental well-being started when ST exceeded 60 min/d for girls and 75 min for boys (p < .001). Small to medium effect sizes.

McAllister et al. [29]

UK

Cross-sectional

N: 4,243

M Age: 13.7

55% girls

To examine associations between depressive symptoms and self-harm by gender and screen types.

Depression: Short version of the Mood and Feelings Questions (SMFQ);

Self-harm: reported whether they had engaged in self-harm in the past year

Self-reported: Time Use Diary (TUD)

activities that involved screen media: texting/e-mailing, social media, internet use, gaming

and “watching TV,

DVDs, downloaded videos” (TV/videos).

More than 3 h/d on social media (n = 92) more likely to self-harm (29%) and less than 1 h/d (n = 852) less likely (19%) girls. For depressive symptoms the results were similar 31% vs. 19%. Small effect sizes. Self-harm or depression among boys was not associated with media use. The correlation was not significant between MH and games or TV/video for boys and girls.

Sampasa-Kanyinga et al. [30]

Canada

Cross-sectional

N: 6,364

M Age: 15.1

56.7% girls

To investigate associations between compliance with the Canadian 24 H Movement Guidelines recommendations and psychological distress.

Psychic distress: 6-item Kessler Psychological Distress Scale (K6)

Self-reported: Time spent in the last 7 days on any screen device in free time.

ST was only associated with sleep duration when combined with anxiety and depression (p < .001), with a small effect size. No direct correlation between ST and anxiety or depression was observed.

Ren et al. [31]

China

Cross-sectional

N: 1,771

M Age: 12 to 19

51.8% girls

To examine the psychological impacts of the pandemic COVID-19

on the development of mood disorders.

Anxiety: General Anxiety Disorder-7 (GAD-7, Chinese version). Depression: Patient Health Questionnaire-9 (PHQ-9, Chinese version); Perceived Social Support: Multidimensional Scale of Perceived Social Support (MSPSS, Chinese version); Psychological resilience: Connor-Davidson Resilience Scale-10 (CD-RISC-10, Chinese version).

Self-reported: average time spent per day, in leisure and study, on electronic devices (TV, smartphones, tablets and computers).

Symptoms of anxiety and depression were 28.3% and 30.8%, respectively. The risk of anxiety/depression decreases with less ST (P < .05). Between ages, the association between TS and depression and anxiety did not differ.

Brown et al. [32]

Canada

Longitudinal

wave 1 (1994–1995) and wave 2 (1996)

N: 6,436

M Age: 16.03

52% girls

To determine whether adherence to a movement behavior profile was associated with differences in depressive symptoms.

Depression: Center for Epidemiological Studies Depression Scale (CES-D)

Self-reported: 7-day recall Daily hours spent on TV/videos; computer games.

High profile AF and low ST reported minor depressive symptoms, and these differences were evident one year later

Brown & Kwan

[33]

Canada

Cross-sectional

N: 1,118

M Age: 15.92

54.5% girls

To examine the effects of reallocation/substitution of PA, ST, and sleep on mental well-being in adolescents.

Flourishing: Diener Flourishing Scale.

Self-esteem: modified Rosenberg Self-Esteem Scale.

Resilience: two articles from the Canadian Campus Well-Being Study.

Self-reported: standard 7-day recall. How many hours per day on average are spent watching TV, using the PC, tablet or smartphone during free time.

ST (combined with < 8 h of sleep) is negatively associated with self-esteem, resilience, and flourishing. Replacing 60 min/d of ST with moderate/intense PA or sleep was associated with better self-esteem and resilience scores. For flourishing, benefits were observed when replacing ST with moderate/intense PA or sleep (in the group that slept less than 8 h). Small effect size.

Gilchrist et al. [34]

Canada

Cross-sectional

N: 46,413

M Age: 9th to 12th grade

51.5% girls

To examine the impact on MH when reallocating 15 min spent on one health behavior to 15 min spent on another behavior.

Depression: 10-item Center for Epidemiologic Studies Revised Depression Scale − 10 (CESD-R-10)

Anxiety: (GAD-7) Generalized Anxiety Disorder 7-item scale

Flourishing: Diener Flourishing Scale.

Self-reported. How much daily time is spent watching TV or movies, playing video/PC games, surfing the Internet, and texting/messaging/e-mailing.

Replacing ST with any behavior (homework, PA, or sleep) may be better for MH outcomes. Small effect size.

Khan & Burton

[35]

Australia

Cross-sectional

N: 2,946

M Age: 16.9

49% girls

To investigate associations between two common recreational screen activities and psychological well-being.

Psychological well-being: Strengths and Difficulties Questionnaire (SDQ).

Self-reported: Average daily time spent on electronic games and TV separately.

Playing electronic games was inversely associated with psychological well-being for male and female adolescents (p < .001), with medium and large effects size. Watching TV was inversely associated with psychological well-being for female adolescents (p < .001), with large effect size.

Nigg et al. [36]

Germany

Longitudinal

N: 686

M Age:

wave 1 (T2) 11.85 years

wave 2 (T3) 16.86 years

55.2% girls

To investigate the relationships between PA, ST and MH.

MH status: Strengths and Difficulties Questionnaire (SDQ), subscales: emotional symptoms; conduct problems; hyperactivity/inattention; relationship problems; pro-social behavior

Self-reported: daily time spent on TV/video, PC/internet and PC games

In girls, TV time at T1 and T2 predicted emotional symptoms at T2 and T3, overall SDQ score and conduct problems, hyperactivity and inattention at T3; PC time at T1 predicted hyperactivity/inattention at T2 and conducted problems at T3; For boys, PC time at T1 predicts emotional symptoms at T2. Conduct problems, hyperactivity/inattention, peer relationship problems, and SDQ score at T2 predicted TV time and PC time at T3.

Twenge & Farley

[37]

UK

Cross-sectional

N: 11,427

M Age: 13.77

50% girls

To examine associations between different types of screen activities and MH indicators. And whether there are differences in these associations between genders.

Self-harm behaviors: “In the past year have you purposely hurt yourself in any way?“ with response options of “yes” or “no”; Depression: Short form mood and feelings questionnaire; Self-Esteem: 5 questions based on Rosenberg’s Self-Esteem Scale; Life Satisfaction: 6 questions based on scales that measure life satisfaction.

Self-reported: How much daily time, outside of school, is spent with all types of screen devices (smartphone, tablet, TV, internet, Digital Versatile Disk (DVD), video games, social media)

Among girls, the association of MH problems with social media and Internet use was more evident than for gaming and TV (p < .001). Boys: social media and Internet use are associated with self-harm behavior. Girls: MH impairment when ST ≥ 2 h; for boys, only with > 5 h of ST. In individuals who used social media for more than 5 h, MH problems were more relevant for girls than for boys (p < .05). Small effect size.

Xiao et al. [38]

China

Cross-sectional

N: 1,680

M Age: 7th to 12th grade

48.7% girls

To evaluate PA and ST during the COVID-19 pandemic, and also their associations with mood disorders and conflict with parents.

Mood State: Shortened Chinese version of the Mood States Profile (POMS).

Conflict with parents: “Approximately how many times have you had a conflict with your parents in the last 7 days?“

Self-reported:

Students reported the number of hours per day they spend studying online and other uses of ST.

More time studying online (p < .05) and other ST (p < .001) verified more conflict with parents. Girls had significantly higher mood disorder scores (p < .01). Other ST use was associated with mood disorders (p < .01). A 1 h addition of ST was associated with a 1.6-to-1.8-point increase in mood.

Bang, et al. [39]

Canada

Cross-sectional

N: 1,477

M Age: 12 to 17

50.9% girls

Correlations to adherence to the 24-Hour Movement and MH Guidelines.

Social difficulties: Strengths and Difficulties Questionnaire (SDQ);

Stress: self-reported;

MH: self-report

Self-reported: how much time on screen devices.

Adolescents who adhered to ST recommendations were more likely to have good psychosocial health (OR = 2.64; 95% CI: 1.21–5.73) when compared to those who did not (p < .05). Mean effect size.

Barthorpe et al. [40]

UK

Cross-sectional

N: 4,032

M Age: 13 to 15

55.2% girls

To investigate associations between time spent on social media and self-harm, depressive symptoms and self-esteem.

Self-mutilation: yes/no response to the question “In the past year, have you hurt yourself on purpose?“

Depressive symptoms: Short Mood and Feelings Questionnaire.

Self-esteem: short Rosenberg scale

ST on social media was recorded in time use diaries (TUD). Social media use was assessed through the activity ‘browsing and updating on social networking sites, e.g., Twitter, Facebook, BBM, Snapchat’.

Significant associations in girls only. Increased time on social media increased the risk of: Automutilation on weekdays (OR = 1.13, 95%CI 1.06 to 1.21). Depression on weekdays (OR = 1.13, 95%CI 1.06 to 1.21). weekend use = 0.19, 95%CI 0.06 to 0.32). Lower levels of self-esteem on weekdays (adjusted OR by 30-min increase in social media use = 1.13, 95%CI 1.06 to 1.21). (adjusted β by 30-min increase in: weekday use = -0.12, 95%CI -0.20 to -0.04; weekend use = -0.12, 95%CI -0.18 to -0.07). Medium effect sizes.

Cao et al. [41]

China

Cross-sectional

N: 4,178

53.4% girls

M Age: 14.2

To identify clustering patterns of health-related behavior and their association with depressive symptoms.

Depressive symptoms: Depression Scale of the Center for Epidemiologic Studies.

Self-reported: daily time spent using all kinds of screen devices.

ST > 2 h/d higher risk of developing depressive symptoms 1,24 (1,06 − 1,45) (p < .01). Cluster of high ST pattern 4.83 ± 1.66 h/d, adolescents were 1.37 times (AOR = 1.37, 95% CI: 1.08–1.73) more likely to develop depressive symptoms compared to Cluster < 2 h/d of ST (p < .01). Small effect size.

Coyne et al. [42]

USA

Longitudinal

N: 487

M Age: 13 to 20

51.6% girls

To assess the correlation between time spent on social media, depression, and anxiety.

Depression: 20-item Depression Scale, Center for Epidemiological Studies (CES-DC).

Anxiety: 6-item Spence Child Anxiety Inventory.

Self-reported: Time spent on social networking sites on a typical day.

Girls spend more time on social networking sites. Increased time spent on social networking sites was not associated with increased MH problems when examined for both genders.

Faria et al. [43]

Brazil

Cross-sectional

N: 217

M Age: 16.08

49.3% girls

To identify PA-related lifestyles and sedentary behavior (SB) and their association with health outcomes.

Common Mental Disorders (CMD): General Health Questionnaire, 12-item version, validated in Brazil.

Tobacco and alcohol use: short version of the Global School-Based Student Health Survey, validated for Brazilian adolescents.

Self-reported: daily time spent in front of any screen device.

Adolescents included in the “Inactive and Sedentary” class had higher CMD scores than those assigned to the “Active and Non-Sedentary” class (p < .05). Girls with signs of CMD were 9.20 (95% CI 1.17–71.52) more likely to be in the “Inactive and Sedentary” class than in the “Active and Non-Sedentary” class. Medium effect size.

Faulkner et al. [44]

Canada

Longitudinal

N: 2,292

M Age: 16.3

54% girls

To examine whether changes in adherence to the Canadian 24 h Movement Guidelines (PA, ST, and sleep) are associated with changes in flowering.

Flourishing: Diener Flourishing Scale.

Depression: Center for Epidemiological Studies Depression Scale-(revised)-10 (CESD-R-10)

Self-reported: Daily time spent watching TV/movies, playing video/PC games, talking on the phone, surfing the Internet, and texting, messaging, or e-mailing.

The majority did not meet the guidelines for total recreation ST (90.5% of the girls and 94.3% of the boys). Meeting the ST guideline was not associated with flowering in both sexes. For girls and boys, meeting more guidelines was not associated with higher flourishing scores when controlling for depressive symptoms and other covariates (p > .05).

Kim et al. [45]

Canada

Cross-sectional

N: 2,320

M Age: 14.58

50.3% girls

Quantify the strength of the association between passive and active forms of ST and adolescent major depressive episodes and anxiety disorders.

Major depressive episode, social phobia, generalized anxiety disorder and specific phobia: Mini International Neuropsychiatric Interview for Children and Adolescents, according to DSM-IV criteria.

Passive ST: Hours /day watch TV, movies, or videos, including YouTube?“

Active screen: hours /day outside of school on average on a PC, laptop, tablet, or smartphone?“

Adolescents with 4 h or more of passive ST p/day, compared to < 2 h, were 3 times more likely to meet DSM-IV-TR criteria for major depressive episode [OR = 3.28 (95% CI = 1.71) -6.28)], social phobia [OR = 3.15 (95% CI = 1.57–6.30)], and generalized anxiety disorder [OR = 2.92 (95% CI = 1.64–5.20)]. Passive ST use was associated with mood and anxiety disorders. Medium effect size.

Weatherson et al. [46]

Canada

Cross-sectional

N: 29,133

50.26% girls

M Age: 15.3

To examine whether complete MH status (CMHS) is associated with adherence to the guidelines for moderate to vigorous PA and recreational ST from the Canadian 24 h Movement Guidelines.

Flourishing: Diener’s Flourishing Scale

Depression: Center for Epidemiologic Studies Revised 10-item Depression Scale

(CESD-R-10). L: wasting

F: flourishing

-DS: low depressive symptoms

+DS: high depressive symptomatology

Self-reported: daily time watching TV or movies, playing video/PC games, talking on the phone, surfing internet and text messaging, email messaging).

To meet the guidelines the sum must be < 2 h p/day of recreational ST.

Presence of DS in 53.85% and F in 50.46%.

L/-DS individuals were 50% more likely to meet ST guidelines compared to L/+DS.

F/+DS individuals were 87% more likely, and F/-DS were 112% more likely to meet ST guidelines (p < .0001).

Lower SD scores and higher F scores were associated with meeting ST guidelines (p < .0001).

Zhang, et al. [47]

China

Cross-sectional

N: 7,200

M Age: 15.50

50% girls

Estimate the combination of exercise time and ST to promote MH. Formulate a benchmark for these variables to prevent the development of psychological problems.

Psychological symptoms: Multidimensional Sub-health Questionnaire of Adolescents (MSQA)

Self-reported: daily time playing video games and watching TV/video programs?“ Every type of ST was counted, from leisure to educational purposes. There were 4 options for teens to choose from: “less than 1 h,“ “1-2 h,“ “2-3 h,“ and “more than 3 h.“

Detection rates for emotional and behavioral symptoms and social adjustment difficulties were higher among boys than among girls.

ST > 2 h/d is a risk factor for emotional, behavioral, psychological symptoms, and social adjustment difficulties, with medium effect size.

Orben & Przybylski [48]

UK

Cross-sectional

N: 355,358

M Age: 16 ± 1.24

50.5% girls

Show that specification curve analysis (SCA) can be a more reliable way to show the relationship between well-being and ST.

Revaluation of data obtained from 3 samples of previous studies:

Monitoring the Future (MTF);

Youth Risk and Behavior Survey (YRBS);

Millennium Cohort Study (MCS).

Psychological Well-being:

Strengths and Difficulties Questionnaire (SDQ);

Rosenberg Self-Esteem;

Moods and Feelings Scale;

4 items from the Center for Epidemiological Studies Depression Scale

Self-reported:

Own computer;

Weekday electronic games;

Hours of social media use;

Weekday TV;

Mean technology;

Use internet at home

Association between digital technology use and well-being is negative, with a 0.4% of variation in well-being. YRBS: When employing the electronic device, the effects were more negative; when including TV use, however, they were less negative. MTF: Watching TV only on the weekend, median positive association with well-being (β = 0.008); Using social media, median negative association with well-being (β = 0.031).

Khouja et al.

[48]

UK

Longitudinal

N: 14,665

M Age: 16 to 18

51% girls

Evaluate the association between ST for device types (watching TV, PC use, and texting), time (weekday or weekend), and anxiety and depression.

Anxiety and depression: computerized version of the Clinical Interview Schedule (CIS-R); Childhood covariates for further adjustment: IQ: Wechsler Intelligence Scale for Children (WISC-III UK); Parental conflict; Father’s presence in the home; Number of people living in the home; Bullying; Early use of family TV.

Self-reported

ST: watching TV, PC use and texting.

Use during the week and at weekends.

More time spent using PC during the week was associated with a small increase in the risk of anxiety (p = .003). More time spent using the PC only on weekends was associated with a small increase in risk of anxiety and depression (p = .003).

There was a small positive association between computer use at age 16 and anxiety and depression two years later.

Liu et al. [49]

China

Cross-sectional

N: 11,831

M Age: 15,0

50.9% girls

To examine the association between the time of cell phone use and depressive symptoms.

Depression: The Center for Epidemiological Studies Depression Scale (CES-D) and the Chinese Youth Self-Report (YSR) of the Achenbach Child Behavior Checklist.

Self-reported: “On an average school day, how many hours have you used a smartphone in the past month?”, “On an average weekend day, how many hours have you used a smartphone in the past month?”

Weekdays: depressive symptoms increased with ST ≥ 2 h/day compared with ST < 1 h/day (19.1% and 10.0% respectively). This association was small.

Weekends: depressive symptoms increased with ST ≥ 5 h/day compared with ST < 2 h/day (18.3% and 8.6%, respectively). This association was medium.

Liu et al.

[50]

China

Cross-sectional

N: 13,119

M Age: 15.18

49,5% girls

Analyze moderate to vigorous PA (MVPA) and screen-based sedentary behavior (SSB) and their correlations with depression, anxiety, and self-injurious behavior.

Depression: Centers for Epidemiological Studies 20-item Depression Scale

Anxiety: 39-item Multidimensional Anxiety Scale

Self-injurious behavior: 5-item subscale, Health-Risk Behavior Inventory.

Self-reported.

Average daily hours spent watching TV, playing video games, or using the PC on a typical school day. Clusters: 1 (high MVPA/low SSB), 2 (low MVPA/low SSB), 3 (low MVPA/high SSB), 4 (low MVPA/low SSB).

Boys: Depressive symptoms in class 3 > classes 1 and 2 (p < .001). Total anxiety in class 3 > classes 1 (p < .014) and > class 2 (P < .009). Self-injurious behaviors in class 4 > class 2 and class 3 > than classes 1 and 2 (p < .001).

Girls: depressive symptoms in class 3 > than classes 1 (p < .001) and 2 (p < .005). Anxiety class 2 > class 1 (p < .042). Both sexes: self-injurious behavior is more severe in class 3.

Paulus et al. [51]

USA

Cross-sectional

N: 4,257

M Age: 120 months

47.5% girls

To investigate associations between Screen Media Activity (SMA) and the psychopathology of internalizing and externalizing symptoms.

Psychological symptoms: Child Behavior Checklist CBCL.

Usable structural neuroimaging and SMA data from the Adolescent Brain and Cognitive Development Study (ABCD).

Self-reported: frequency of screen use.

Youth with higher GFA 1 and 4 scores, i.e., had more significant levels of externalizing problems.

Also, they had a thinner occipital cortex and a lower volume in the orbitofrontal areas, such as a thinner hippocampus and a lower inferior temporal cortical volume.

Individuals with higher SMA-related GFA 1 (β = 0.059) and GFA 4 (β = 0.095) scores had significantly higher externalizing scores, with small effect sizes.

Perrino et al.

[52]

USA

Longitudinal

N: 370

M Age: 14.65

47.6% girls

To understand SB based on screen activity and examine the relationship with internalizing psychological symptoms.

Internalizing symptoms: internalizing subscale of the Youth Self-Report.

The variable internalizing problems were created by summing three subscales:

(a) anxious-depressed

b) withdrawn

c) somatic complaints

Physical Activity Questionnaire for Adolescents (PAQ-A) SB subscale. Self-report the amount of time spent in the previous week on TV; video games/PC games; texting; email; surfing the Internet and using the telephone.

Girls had more internalizing symptoms and used more internet, email, messaging and cell phones (p < .001). Boys spent more time with video games (p < .001). Positive association between internalizing symptoms and SB in the early phase (p = .01) and their trajectories (p < .001), in both genders (p = .50).

Hrafnkelsdo-ttir et al.

[53]

Iceland

Cross-sectional

N: 244

M Age: 15.8

59% girls

To examine separate and interactive associations of ST and physical activity level with MH.

MH problems (depression, anxiety and somatic symptoms): “22-item version of the Subscales of the Symptom Checklist 90”.

Self-esteem: Rosenberg Self-Esteem Scale.

Life satisfaction: “Diener’s Satisfaction with Life Scale”.

Self-reported: how many h/d on average played PC games, watched TV/DVD/Internet material, used the Internet for web browsing/Facebook/email, and participated in “other” PC use. Participants were classified into high and low ST groups.

Less ST was associated with lower odds of reporting: Symptoms of depression (RR = 0.33, 95% CI = 0.14–0.76) - p < .001. Anxiety (RR = 0.44, 95% CI = 0.23–0.84) - p < .01.

Low self-esteem (RR = 0.31, 95% CI = 0.15–0.66) - p < .005. Dissatisfaction with life (RR = 0.38, 95% CI = 0.20–0.72) - p < .005. Associations with small effect size.

Gireesh et al. [54]

UK

Cross-sectional

N: 120,115

M Age: 15,0

52.42% girls

To identify modifiable behavioral factors for mental well-being taking into account deprivation, ethnicity, and grouping.

MH: Warwick-Edinburgh Mental Well-Being Scale (WEMWBS).

Self-reported on weekdays, and weekends, time spent on TV, internet, smartphone and PC games.

Categorized into ‘≥7 hours/day’, ‘about 5–6 hours/day’, ‘about 3–4 hours/day’, ‘about 2 hours/day’, and ‘≤1 h/day’.

Being physically inactive, having higher ST, and bullying were associated with decreased well-being in both genders, with the association being more significant in girls (p < .005) than in boys (p < .05). Well-being in both sexes decreased with higher ST in both sexes, starting at about 3 h/d of exposure (p < .001). Associations with small to moderate effect sizes.

Khan et al. [55]

Bangladesh

Cross-sectional

N: 671

M Age: 14.3

49% girls

To explore interactive associations of PA and ST with psychosocial difficulties.

Psychosocial difficulties: Parent-reported Strengths and Difficulties Questionnaire (SDQ)

Self-reported: Adolescent Sedentary Activity Questionnaire (ASAQ) for a typical school day and weekend. Total recreation ST was dichotomized as ≤ 2 h/d (‘low’) or > 2 h/d (‘high’).

Insufficient physical activity + high ST resulted in an 18% increase in total psychosocial difficulties scores, with small effect size. ST was not significantly associated with SDQ difficulties scores (model 2: p = .44).

Twenge et al. [56]

USA

Longitudinal

N: 506,820

M Age: 13 to 18

50.76% girls

To examine cultural changes in 3 generations (GenX, Millennials, and iGen). Determine trends in depressive symptoms, suicide-related outcomes, and suicide rates and examine associations between MH, ST of new and older media, and non-screen activities.

Studies included: Monitoring the Future (MtF) and the Youth Risk Behavior Surveillance System (YRBSS) from 1991 and the Centers for Disease Control data from 1999.

Depressive symptoms: MtF: 6 items from the Bentler Inventory of Medical and Psychological Functioning depression scale

YRBSS: 4 items related to suicide.

Self-reported. Non-school activities only.

2009: TV, video games, PC games, and internet;

2011: Facebook;

2013 and 2015: smartphone and other social media.

Homework. MtF asked, “About how many hours spend on average in a week on all their homework, including

both in school and out of school…”

Since 2010, iGen teens have spent more time in new screen media activities and less time in non-screen activities. Teens who spent more time on screen activities were more likely to have high depressive symptoms. Social media use was correlated significantly with depressive symptoms among girls (p < .001). All suicide-related outcomes were significantly with electronic device use (p < .001). Associations with a small effect size.

Yan et al. [57]

China

Cross-sectional

N: 2,625

M Age: 13 to 18

46.9% girls

To determine time spent on screen activities, associations with adiposity, unhealthy eating behaviors, sleep, PA, academic performance, anxiety, self-esteem, and life satisfaction.

Anxiety: High School Student Mental Health Scale.

Satisfaction with life: Satisfaction With Life Scale.

Self-esteem: Rosenberg Self-esteem Scale.

Self-reported: Time spent watching TV, playing e-games, receiving news or study materials from electronic devices, using social media sites or apps, and watching videos on school and non-school days.

Watching TV during school days for 2 to 4 h was negatively associated with anxiety (p = .047) and self-esteem (p = .032). Receiving news via digital media or studying for 2 to 4 h (p = .036) or > 4 h (p = .002) on school days was positively associated with anxiety. Positive association of social media use with anxiety (p = .009). On school days, watching TV > 04 h is negatively associated with life satisfaction (p = .012), with large to medium effect size.

Khan & Burton

[58]

Bangladesh

Cross-sectional

N: 505

M Age: 14.3

53% girls

Assess the association of moderate to vigorous PA (MVPA) with depressive symptoms in adolescents with recreational high ST.

Depression: Center for Epidemiological Studies Depression Scale (CESD 10)

Self-reported: recreational ST (watching TV, DVDs/videos, PC use for entertainment; use of social media - Facebook, Twitter), Adolescent Sedentary Activity Questionnaire (ASAQ).

Adolescents with high recreational ST: a quarter (24.6%) reported depression. A significant amount more girls with high ST than boys reported depression (29% vs. 20%). There were no significant associations between meeting MVPA recommendations and depression for those with low recreational ST (< 2 h/d). Medium effect size.

Przybylski & Weinstein [59]

UK

Cross-sectional

N: 120,115

M Age: 15,0

Assess how ST is linked to mental well-being and empirically quantify a moderate activity level in digital activities.

Mental well-being: Warwick-Edinburgh Mental Well-Being Scale.

Self-reported: throughout the week, how much of your free time is spent on TV, console, and computer games, internet, email, smartphones, online chats and social networks.

Relationship to impaired mental well-being and watching movies/TV, playing games, and using the PC during the week. Small effect size. For smartphone use, this relationship was observed only on weekdays. Girls spent more time using smartphones, PC and watching videos, and boys spent more time on the PC and console games (p < .001).

Babic et al. [60]

Australia

Longitudinal

N: 322

M Age: 14.4

66% girls

To examine

associations between changes in ST

(total and device-specific) and MH indicators (well-being and malaise).

Physical self-concept: Physical self-concept subscale from Marsh’s Physical Self-Description Questionnaire.

Psychological well-being: Diener and colleagues’ Flourishing Scale.

Psychological distress: Strength and Difficulties Questionnaire.

Self-reported: Adolescent Sedentary Activity Questionnaire ASAQ:

Recreational time spent using screen devices (TV, PC, tablet/cellphone) each weekday and weekend.

Non-recreational ST when a PC was used for homework.

Total recreational ST (p = .048) and tablet/cellphone use (p < .001) were negatively associated with physical self-concept. Total recreational ST (p = .001) and PC use (p = .003) were negatively associated with psychological well-being. There was a positive association between TV/DVD use and psychological difficulties (p = .015). Large effect size.

Goldfield et al. [61]

Canada

Cross-sectional

N: 358

M Age: 15.6

72.9% girls

To examine the association between duration and types of ST and depressive symptomatology in overweight and obese adolescents.

Depressive symptoms: Childhood Depression Inventory (CDI)

Self-reported: How many hours a day spent watching TV, playing video games, and using the computer for recreational purposes.

Computer games excluded.

ST is associated with more severe depressive symptomatology (p = .001). Playing video games and recreational PC use was associated with depression (p = .05 and p = .006) respectively, but watching TV showed no significant relationship (p = .09). ST may represent a risk factor for depressive symptomatology. Small effect size.

Gunnell et al. [62]

Canada

Longitudinal

N: 1,160

M Age: 13.54

60.5% girls

To examine changes in PA and ST, depression, and anxiety. Bidirectional relationships between PA, ST, and depression and anxiety.

ST: Self-report questionnaire

Depression: Children’s Depression Inventory (CDI).

Anxiety: Multidimensional Anxiety Scale for Children-10 (MASC-10).

Self-reported

Questions: How many hours per day did the adolescents spend watching TV, playing video games, and using the computer on weekdays and weekends?

Symptoms of depression, anxiety, and ST increased significantly (p < .05) over time. Higher initial symptoms of anxiety are associated with initial symptoms of depression (p < .05) and higher ST (p < .05).

Hayward et al. [63]

Australia

Cross-sectional

N: 3,295

45.7% girls

M Age: 15.14

To establish associations and relative contributions between diet, PA, ST behaviors, and depressive symptomatology.

Depressive symptomatology: Moods and Feelings Questionnaire - Short Form (SMFQ).

ST: Key Indicators and Measures from the Youth Health Survey.

Questions: How many hours during the 07 days before the survey did adolescents have ST (less than 01 h, 1-2 h, 2-5 h, and 5 + h).

Achieving the recommended ST guidelines was associated with reduction in the likelihood of

depressive symptoms (OR = 0.90 and 95% CI= [0.87, 0.93] in girls. Small effect size.

Trinh et al. [64]

Canada

Cross-sectional

N: 2,660

52.5% girls

M Age: 15.8

To examine the independent effects of PA and ST on MH, school connectedness, and academic performance, and to identify possible interactions between PA and ST in such associations.

Psychological stress: General Health Questionnaire (GHQ)

Depressive symptoms: four items adapted from the Center for Epidemiologic Studies Depression (CES-D)

Self-esteem: 6 items adapted from the Rosenberg Self-esteem Scale

Self-reported

Question: “In the past seven days, how many hours per day did you spend watching TV, playing games, chatting on the computer, or surfing the Internet?“

ST was significantly associated with poor MH outcomes, including psychological distress (p < .05), low self-esteem (p < .05), and depressive symptoms (p < .05). Boys: greater screen time was associated with psychological distress (p < .05), low self-esteem (p < .05), and depressive symptoms (p < .05). Girls: greater ST was associated with low self-esteem (p < .05). Small effect size.

Maras et al. [65]

Canada

Cross-sectional

N: 2,482

M Age: 14.1

57.8% girls

To examine the relationships between sedentary ST and symptoms of depression and anxiety.

Depression: The Children’s Depression Inventory (CDI) (Self-related).

Anxiety: Multidimensional Anxiety Scale for Children − 10 (MASC-10)

Leisure-Time Sedentary Activities 6-item questionnaire - measuring how many hours per day students spend on activities such as watching TV, playing video games, and using the PC.

Total ST is associated with depression (p < .001) and severity of anxiety (p < .001). ST related to video games and PC use was associated with more severe depression (p < .001). ST related to video games alone was associated with more severe anxiety (p < .001), with medium effect size.

Suchert et al. [66]

Germany

Cross-sectional

N: 1,296

M Age: 13.7

47% girls

Examine the effects of sedentary behavior (SB) on mental well-being and the differences between screen-based SB (sSB) and non-screen-based SB (nSB).

Depression: German version of the Center for Epidemiological Studies Depression Scale for Children (CES-DC);

General self-efficacy: 5 items from Schwarzer and Jerusalem’s general self-efficacy scale;

Self-esteem: KINDL-R.

Physical attractiveness self-image: physical self-concept scales by Stiller, Würth, and Alfermann.

Self-reported: Time spent on activities involving screens during the last school day and on the last Sunday: watching TV/DVDs, playing video/PC games, and other leisure activities on PC/mobile phone, except active electronic games.

Girls: screen-based sedentary behavior (sSB) is associated with higher depression (p = .032), lower self-esteem (p = .039), lower self-concept for physical attractiveness (p = .003) and lower overall self-efficacy (p = .002).

Boys: significant positive association between sSB and self-esteem (p = .014).

Small effect sizes.

Nihill et al. [67]

Australia

Cross-sectional

N: 357

100% girls

M Age: 13.2

To examine the association between SB and self-esteem among adolescent girls.

SB: Adolescent Sedentary Activity Questionnaire (ASAQ);

Physical self-concept and self-esteem: Marsh’s Physical Self-Description Questionnaire (PSDQ).

ST was created by adding time spent watching TV, videos, DVDs and using the PC for non-school purposes.

Multilevel models did not reveal any association. After corrections, there were inverse associations between time spent watching DVDs (p < .05), playing PC games (p < .05), and total ST (p < .05) and self-esteem. Small effect size.

Straker et al. [68]

Australia

Longitudinal

N: 643

M Age: 14,0

To examine the relationships between sedentary behavior (including ST) and self-esteem.

MH: Cowen’s Perceived Self-Efficacy Scale.

BDI for Youth

CBCL - Youth Self-Report version of the Child Behavior Checklist.

Time spent watching TV, playing electronic games, video games and other different PC usages for graphics, text, email, internet and general gaming. Categories: C1 instrumental PC users; C2 multimodal electronic gamers; C3 PC e-gamers.

Instrumental PC users girls had higher self-efficacy than e-gamers girls (p = .028).

Instrumental PC boys had lower depression scores than multimodal and e-gamers boys (p = .046), having also less symptoms. Small effect sizes.

Arbour-Nicitopoulos et al. [69]

Canada

Cross-sectional

N: 2,935

M Age: 15.9

49% girls

Investigate the prevalence of psychological distress and its associations with health risk behaviors.

Stress: General Health Questionnaire (GHQ);

Substance Use (alcohol, tobacco, and cannabis): Self-report

Self-reported: In the last 7 days, how many hours per day you spend in front of screens.

Students who did not comply with ST recommendations were at higher risk for psychological distress (p < .001).

Girls are approximately 2 times more likely than boys to experience psychological distress (p < .001).

Tobacco use was significantly associated with the risk of psychological distress (p < .001).

Small effect sizes.

Cao et al. [70]

China

Cross-sectional

N: 5,003

47.9% girls

M Age: 13.13

To test the association between ST, PA and psychological problems in adolescents.

Depression: 18-item version of the Self-Regulatory Depression Scale for Children (DSRSC);

Anxiety: 41-item Screen for Child Anxiety Related Emotional Disturbances (SCARED);

Satisfaction with school life: 12-item School Life Satisfaction Assessment Questionnaire for Adolescents.

Self-reported: How much time, on average, I spend on weekdays and weekends on sedentary activities, such as watching TV and using a PC.

Positive associations of ST with depression, anxiety, and dissatisfaction with school life (p < .001). High ST was a risk factor for: Depression (OR = 1.52–95%CI: 1.31–1.76); Anxiety (OR = 1.36, 95%CI : 1.18–1.57); Dissatisfaction with school life (OR = 2.07, 95% CI : 1.79–2.38). Small effect sizes.