OBJECTIVE: We hypothesized that greater screen use would be associated with greater psychological difficulties and that children with high levels of screen entertainment use and low levels of physical activity would have the most-negative psychological profiles.
METHODS: Participants were 1013 children (age, mean ± SD: 10.95 ± 0.41 years), who self-reported average daily television hours and computer use and completed the Strengths and Difficulties Questionnaire. Sedentary time (minutes per day with <100 cpm) and moderate/vigorous physical activity (MVPA) (minutes with ≥2000 cpm) were measured by using accelerometers. Multivariate regression models examined the association between television viewing, computer use, sedentary time, and Strengths and Difficulties Questionnaire scores, with adjustment for MVPA, age, gender, level of deprivation, and pubertal status.
RESULTS: Greater television and computer use were related to higher psychological difficulty scores after adjustment for MVPA, sedentary time, and confounders. However, sedentary time was inversely related to psychological difficulties after adjustment. Children who spent >2 hours per day watching television or using a computer were at increased risk of high levels of psychological difficulties (television, odds ratio [OR]: 1.61 [95% confidence interval [CI]: 1.20–2.15]; computer, OR: 1.59 [95% CI: 1.32–1.91]), and this risk increased if the children also failed to meet physical activity guidelines (television, OR: 1.70 [95% CI: 1.09–2.61]; computer, OR: 1.81 [95% CI: 1.02–3.20]).
CONCLUSION: Both television viewing and computer use are important independent targets for intervention for optimal well-being for children, irrespective of levels of MVPA or overall sedentary time.
WHAT'S KNOWN ON THIS SUBJECT:
It is not known whether use of specific electronic media (computer use or television viewing) or total sedentary time is most strongly related to psychological difficulties and the extent to which objectively measured physical activity moderates this relationship.
WHAT THIS STUDY ADDS:
Greater television or computer use was related to greater psychological distress for 10- to 11-year-old children, irrespective of objectively measured physical activity or sedentary time. Limiting computer use and television viewing may be important for optimal well-being for young people.
The availability of electronic media has increased markedly in recent years.1 Television viewing has been positively associated with childhood obesity,2 and combined use of television and computers (screen entertainment) was reported to be positively related to psychological difficulties (emotional, peer, concentration, and conduct problems) among young people.3,–,5 Use of electronic media by young people may vary according to both medium (computer use versus television viewing) and purpose (computer use for gaming versus homework), and it is unclear the extent to which these activities contribute to overall sedentary time. Objectively measured sedentary time has been positively related to insulin resistance in young people,6 but its relationship with psychological well-being is not well documented. Traditionally, sedentary time has been assessed through either children's self-reports or parents' proxy reports, but the validity of both of these approaches is limited by recall accuracy. Accelerometers provide an objective method of quantifying sedentary time7,8 but cannot provide information about the activities being conducted while the individual is sedentary. Therefore, it is unclear whether time spent static and/or the amount of time spent using screen entertainment is most closely related to well-being.3,5,9
Although screen viewing has been reported to be inversely related to well-being,3 physical activity has been shown to be positively related to well-being among children and adolescents.10,11 Therefore, any potential negative psychological consequences of sedentary behaviors and/or time may be ameliorated by higher levels of physical activity. A recent study with 4- to 12-year-old Scottish children found that children with the highest scores for screen entertainment and the lowest levels of physical activity had the highest levels of psychological difficulties.5 However, that study and other studies in this area relied on self-reported physical activity.
It is important to account for confounders such as socioeconomic position in this area, because adolescents from a lower socioeconomic position report less participation in sports/exercise and greater screen viewing behavior, compared with those from a higher socioeconomic position.12,13 Because boys report a higher volume of use of electronic media, especially computers,13 and girls report lower levels of activity routinely, it also is important to account for gender when assessing associations between sedentary indicators and well-being.14 Overweight also is a potential confounder, because lower levels of physical activity have been associated with higher levels of overweight15 and overweight has been positively associated with psychological difficulties.16
This study addressed weaknesses in the literature by assessing associations between objectively measured physical activity levels, sedentary time, television/computer use, and psychological difficulties in a large sample of 10- to 11-year-old UK children. We hypothesized that frequency of use of screen entertainment and sedentary time would be positively associated with psychological difficulties and that moderate/vigorous physical activity (MVPA) would be inversely associated, after adjustment for confounders. We also hypothesized that participants with high screen entertainment use combined with low levels of MVPA would have the most negative psychological profiles.
Participants were year 6 children (10–11 years of age) recruited from 23 primary schools within Bristol (England) as part of the Personal and Environmental Associations with Children's Health project between September 2006 and July 2008. Ethical approval for the study was provided by a University of Bristol ethics committee, and informed parental consent was obtained for all children. Complete information on recruitment and data collection procedures was reported elsewhere.17
Sedentary time was measured with accelerometers (ActiGraph GT1M [ActiGraph, Pensacola, FL]).18 Children were instructed to wear the accelerometer on a belt around their waist during waking hours for 7 consecutive days, and the instruments were programmed to record data every 10 seconds. Data were collapsed into 1-minute epochs, and sedentary time was defined as any minute in which <100 counts were recorded.8,19
Children were asked to report, through single items on a computerized questionnaire, “How many hours a day do you usually spend watching TV?” and “How many hours a day do you usually spend playing on a computer (not for homework)?” Responses ranged from 0 to ≥5 hours per day.
MVPA was used in analyses because it has been related consistently to psychological well-being.20 For consistency with other related studies, MVPA was defined as any minute in which ≥2000 accelerometer counts were recorded.8,21
Psychological difficulties were assessed by using the self-report Strengths and Difficulties Questionnaire (SDQ),22 one of the most widely used inventories of young people's psychological well-being.23 Participants responded to 25 items on a 3-point scale (scored from 0 to 2), from which 20 items were summed to create a total difficulty score (range: 0–40). On the basis of this score, categories were created to represent low (scores of 0–14), borderline (scores of 15–19), and high (scores of 20–40) levels of psychological difficulties.
Level of Deprivation
The UK Index of Multiple Deprivation (IMD) 2007 score was obtained for each participant's full postal code. The IMD score is based on 6 categories of deprivation (income, employment, health and disability, educational skills and training, housing, and geographical access to services),24 and a higher score indicates a higher level of deprivation.
Pubertal status was measured by using the scale developed by Petersen et al,25 and 5 derived stages (equivalent to Tanner stages) were used in analyses.
Height (in meters) was measured by using a stadiometer, and weight (in kilograms) was measured by using digital scales (Seca, Hamburg, Germany), with children wearing indoor clothing and no shoes. BMI SD scores were derived from standard tables.26
Data Reduction and Analyses
Means, SDs, and checks for normality were calculated for all variables. Participants were included in analyses only if they provided ≥3 days of accelerometer data, with ≥600 minutes of registered time per day. Periods of ≥60 epochs of 0 counts were interpreted as time when the monitor was not worn and were discarded, as were data from midnight to 7 am, to minimize the influence of sleep on sedentary time. Minutes with <100 cpm and ≥2000 cpm were used to represent sedentary time and MVPA, respectively.8 We acknowledge the debate concerning the appropriate thresholds of accelerometer counts to define sedentary time and MVPA in children; therefore, analyses also were conducted for alternative thresholds for sedentary time (<800 cpm) and moderate physical activity (≥3200 cpm).27 Results are reported in the tables for <100 cpm and ≥2000 cpm, but any differences in associations with alternative cutoff points are reported in the text.
Differences in television viewing, computer use, and pubertal status between boys and girls were assessed by using Pearson χ2 tests, with independent t tests being used for all continuous variables. Spearman correlations between sedentary behaviors, sedentary time, and MVPA were determined. Multivariate linear regression models were used to assess the associations between television viewing, computer use, sedentary time, MVPA, and SDQ total scores. Baseline models were unadjusted (model 1). Models then were adjusted for confounders (age, gender, deprivation, and pubertal status [model 2]) and then MVPA (model 3). Screen viewing and sedentary time also were adjusted for each other (model 4). When confounders were significantly related to psychological difficulties, terms for interactions with other significant exposures were entered into the model and F tests were used to test the significance of the interactions. Robust SEs were used to take into account clustering of participants in schools (ie, nonindependence of pupils from the same school).
To interpret data in the context of current public health guidelines, participants were grouped according to whether they met or exceeded screen entertainment guidelines of ≤2 hours per day28,29 and/or physical activity guidelines of ≥60 minutes of MVPA per day.30 Data for these groups were entered into multivariate logistic regression analyses (with adjustment for confounders) to determine whether meeting screen entertainment and/or physical guidelines was related to risk of belonging to the group with borderline/high SDQ scores, which have been related to clinically diagnosed disorders.31 All analyses were conducted by using Stata 10 (Stata Corp, College Station, TX), and significance was set at P < .05.
Of the 1899 year 6 children from the 23 schools who were invited to take part in the study, 1340 (70.5%) provided parental consent. Of these, 33 were absent on the days of measurement. Seven participants did not complete the computerized questionnaire, 34 accelerometers were not returned or failed to download data, 24 accelerometers had timing errors, and 229 children did not meet the inclusion criteria for valid accelerometer data, which left data for 1013 children for analyses. There were no differences in age, gender, BMI SD scores, or pubertal status for the children included in analyses, but those who were included had lower IMD scores and spent significantly less time watching television or using a computer, compared with those who were excluded.
Girls had significantly less MVPA time and more sedentary time, compared with boys (Table 1). There were no gender differences in total SDQ scores or time spent watching television, but boys spent significantly more time using computers, compared with girls. Girls were significantly more advanced in pubertal status and were heavier than boys. Sixty-nine percent of children (n = 899) were in the low category, 17.2% (n = 221) were in the borderline category, and 13.8% (n = 180) were in the high category for psychological difficulties, on the basis of SDQ scores. Hours of computer and television use were weakly but significantly positively related to sedentary time (computer, ρ = 0.065; P = .023; television, ρ = 0.076; P = .04) but not to MVPA (computer, ρ = −0.025; P = .383; television, ρ = −0.006; P = .827). The correlation between television viewing and computer use was weak to moderate (ρ = 0.275; P < .01), whereas the association between sedentary time and MVPA was strong (ρ = 0.726; P < .01). For potential confounders, higher SDQ scores were significantly associated with more-advanced maturational status (ρ = 0.82 [95% confidence interval [CI]: 0.40–1.22]) and greater deprivation (ρ = 0.04 [95% CI: 0.01–0.07]). Younger children also had higher SDQ scores (ρ = −1.41 [95% CI: −2.17 to −0.11]), but there was no significant association between gender (ρ = −0.40 [95% CI: −1.11 to 0.29]) or BMI SD scores (ρ = 0.07 [95% CI: −0.23 to 0.37]) and SDQ scores.
Table 2 shows that greater television viewing and computer use were associated with higher SDQ scores in all models. Conversely, greater sedentary time (minutes with <100 cpm) was associated with lower (more-positive) SDQ scores in all models. Greater MVPA was significantly related to lower SDQ total scores only when sedentary minutes were taken into account (model 3). Further investigation indicated that this might be attributable to the association in different directions between MVPA and the 4 subscales that constitute the SDQ total score. MVPA was significantly inversely correlated with internalizing difficulties (emotional, ρ = −0.99; P = .001) and peer problems (ρ = −0.059; P = .039) but was positively correlated with externalizing difficulties (hyperactivity, ρ = 0.090; P = .002) and conduct problems (ρ = 0.127; P < .001), which potentially neutralized any association with total SDQ scores. Associations were similar in direction, magnitude, and significance for sedentary time assessed as minutes with <800 cpm, MVPA assessed as minutes with ≥3200 cpm, and mean cpm values (data not shown), which suggests that the findings were not a function of the cutoff points selected. The only significant (P < .05) interactions between confounders and sedentary indicators were between gender and computer use (ρ = −0.59 [95% CI: −1.16 to −0.02]) and between pubertal status and television use (β = 0.30 [95% CI: 0.19–0.41]), but entering these into the models did not alter any associations.
In logistic regression analyses with borderline to high SDQ scores as the dependent variable, children who exceeded guidelines for television viewing (n = 288 [28.6%]) or computer use (n = 141 [14.0%]) had higher SDQ scores (television, odds ratio [OR]: 1.61 [95% CI: 1.20–2.15]; computer, OR: 1.59 [95% CI: 1.32–1.91]). Risk of high SDQ scores was greatest for the small proportion of children (n = 69 [6.8%]) who spent ≥2 hours watching television and ≥2 hours per day playing on a computer (OR: 2.27 [95% CI: 1.56–3.13]). Children who met physical activity guidelines (n = 352 [34.7%]) were not at significantly increased risk of psychological difficulties (OR: 1.03 [95% CI: −0.08 to 1.19]) unless they also failed to meet screen guidelines for either television viewing (OR: 1.69 [95% CI: 1.09–2.61]) or computer use (OR: 1.81 [95% CI: 1.02–3.20]). However, those who met activity guidelines but exceeded screen entertainment guidelines for either television viewing (OR: 1.54 [95% CI: 1.07–2.31]) or computer use (OR: 1.48 [95% CI: 1.08–2.04]) were at increased risk of psychological distress.
This study found that greater television and computer use was related to greater psychological difficulties, independent of gender, age, level of deprivation, pubertal status, and objectively measured physical activity and sedentary time. In addition, children who reported >2 hours of daily computer or television use but engaged in <60 minutes MVPA per day were at increased risk for psychological difficulties. This is consistent with other studies that demonstrated an enhanced negative effect for children with both high screen-viewing levels and low physical activity levels.3,5
Contrary to our original hypothesis, MVPA was not significantly related to SDQ scores except when sedentary time was taken into account. This lack of association may be attributable in part to the fact that SDQ subscales were related in different directions to physical activity, a finding supported by 2 other longitudinal studies in which higher levels of self-reported physical activity were associated with lesser emotional and peer problems32,33 and greater hyperactivity problems.32
One study with 1397 US adolescents 12 years of age also investigated the association between objectively measured physical activity and depression and, similar to this study, found no significant association between MVPA and psychological outcomes.19 However, other studies showed an independent relationship between higher levels of self-reported physical activity and higher total SDQ scores, even when sedentary behaviors were included in the analysis.3,5 Taken together, these studies suggest that the relationship between physical activity and well-being may depend on both the measures of exposure and the outcomes used. Additional studies are required to determine whether conflicting findings are a function of different measures of physical activity (objective versus self-reported) and/or whether physical activity is related differently to different psychological outcomes in young people.
The finding that less-objectively measured sedentary time was related to increased psychological difficulties was contrary to the hypothesized result. It may be that the threshold used here to define sedentary time (<100 cpm) was inappropriate, but analyses using an alternative recommended threshold (<800 cpm)27 yielded the same results (data not shown). One other study did report a very weak but significant inverse relationship between sedentary time (minutes with <100 cpm) and depression scores.19 The weak association between sedentary time and screen entertainment found here and in other studies8,34 supports the view that sedentary time and sedentary behaviors are different constructs.35 Additional studies are necessary to confirm the relationship between sedentary time and psychological difficulties and should include sedentary activities that may have a beneficial impact on psychological well-being (for example, socializing and reading), as well as those that have a negative impact.
The main limitation of this study is its cross-sectional design, because we could not determine whether higher levels of screen entertainment were associated with the development of negative well-being or negative well-being preceded higher levels of screen entertainment. Other limitations include the sample, which, although relatively large, was limited to 1 predominantly urban area. However, levels of screen entertainment and average sedentary time reported here were similar to those reported for other UK samples.3,5,8 The mean SDQ scores in this study were similar to those reported for 2 UK studies with older adolescents (11–14 years32 and 13–16 years3 of age), but mean SDQ scores and the proportions of children with borderline/high SDQ scores were significantly greater in this study, compared with a younger sample of Scottish 4- to 12-year-old children.5
Other limitations of this study include the use of single self-report items to measure screen entertainment, which is influenced by children's recall ability. However, single-item measures have been shown to have high validity, compared with other methods used currently.36 Accelerometers may not capture sedentary time when the child is not moving sufficiently to register a count, although it is unlikely that children of this age remain that static for long periods. To maintain a sufficient number of children with borderline/high SDQ scores in the analyses, a 3-day inclusion criterion for accelerometer data was selected, which is consistent with some childhood studies, although others recommended a 4- to 5-day minimal inclusion criterion at this age.37 However, use of a 5-day inclusion criterion did not yield different findings for any of the analyses (data not shown) and would have meant excluding 163 of the children (34.1%) with borderline/high levels of psychological difficulties. Finally, although this study adjusted for important confounders, other factors that may influence children's psychological difficulties, such as parental psychological well-being and screen viewing content, were not included in these data.
The strengths of this study include the separate investigation of television and computer use, objective measurement of sedentary time and physical activity, and the use of a multidimensional, widely used index of well-being that has shown good validity and reliability for children. Important confounders (age, pubertal status, level of deprivation, weight status, and gender) were accounted for in analyses. Associations for screen entertainment media and sedentary and physical activity were presented independently and in combination. This investigation of combined physical activity and sedentary behaviors is important to consider for intervention, because it has been reported that both parents and children perceive that >2 hours of use of electronic media is acceptable as part of a “balanced lifestyle” and that excessive use of electronic media is not a concern if children are physically active.38,–,40 These data support some restriction of screen entertainment use irrespective of levels of physical activity and indicate that guidelines for both sedentary and physical activity behavior are warranted.41 The current guidelines from the United States28 and Australia29 for screen entertainment of no >2 hours per day were able to differentiate between children with normal versus borderline/high levels of psychological difficulties.
This work was supported by the National Prevention Research Initiative (grant G0501311) and the World Cancer Research Fund. This report involves research arising from a Career Development Fellowship (to Dr Jago), supported by the National Institute for Health Research.
We are very grateful to the participants and their families who gave their time to the study. We also acknowledge all members of the Personal and Environmental Associations with Children's Health Project team not listed as coauthors of this article.
- Accepted August 2, 2010.
- Address correspondence to Angie S. Page, PhD, Centre for Exercise, Nutrition, and Health Sciences, School for Policy Studies, University of Bristol, Tyndall Avenue, Bristol BS8 1TH, United Kingdom. E-mail:
The views expressed in this publication are those of the authors and not necessarily those of the National Health Service, the National Institute for Health Research, or the Department of Health. The funding agencies were not involved in the design, analysis, or writing of the study.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
- CI =
- confidence interval •
- IMD =
- Index of Multiple Deprivation •
- MVPA =
- moderate/vigorous physical activity •
- OR =
- odds ratio •
- SDQ =
- Strengths and Difficulties Questionnaire
- 1.↵Office for National Statistics. Internet access. Available at: www.statistics.gov.uk/cci/nugget.asp?id=8. Accessed September 15, 2010
- Tripodaki E,
- Kormas G,
- Konstantoulaki E,
- et al
- Hamer M,
- Stamatakis E,
- Mishra G
- Sardinha LB,
- Andersen LB,
- Anderssen SA,
- et al
- Ekelund U,
- Brage S,
- Besson H,
- Sharp S,
- Wareham NJ
- Steele RM,
- van Sluijs EM,
- Cassidy A,
- Griffin SJ,
- Ekelund U
- 11.↵US Department of Health and Human Services. Physical Activity Guidelines Advisory Committee Report. Washington, DC: US Department of Health and Human Services; 2008
- 20.↵Chief Medical Officer. At Least Five a Week: Evidence on the Impact of Physical Activity and Its Relationship to Health. London, England: Department of Health; 2004
- Ekelund U,
- Sardinha LB,
- Anderssen SA,
- et al
- Cole TJ,
- Freeman JV,
- Preece MA
- 28.↵American Academy of Pediatrics, Committee on Public Education. Children, adolescents, and television. Pediatrics. 2001;107(2):423–426
- 29.↵Department of Health and Ageing. Active Kids Are Healthy Kids: Australia's Physical Activity Recommendations for 5–12 Year Olds. Canberra, Australia: Department of Health and Ageing; 2004
- 30.↵US Department of Health and Human Services. Physical Activity Guidelines for Americans. Washington, DC: US Department of Health and Human Services; 2008
- van Sluijs EM,
- Page A,
- Ommundsen Y,
- Griffin SJ
- Granich J,
- Rosenberg M,
- Knuiman M,
- Timperio A
- Copyright © 2010 by the American Academy of Pediatrics