Background
A recent examination of sedentary behaviour among Australian young people reported that most youth exceeded the daily recommendation of ≤2 h of screen time per day, and stated that limiting screen-based media use among children and adolescents was ‘
virtually impossible’ [
1]. Internet use represents one component of the broader category leisure screen – time. It has been suggested that screen-use, typically measured as specifically for leisure (not for school or work-related purposes) may interact with emotional and mental health [
2‐
4]. With a large proportion of adolescents reporting mental health disorders or associated symptomatology [
5], and increasing popularity for online activities for entertainment, socialising and information sourcing, quantifying the strength of any possible associations between Internet use and mental health is critical.
The screen-time recommendation of two hours or less, advocated in Australia [
6] and abroad [
7], has been criticised as unrealistic, due to screens being deeply integrated into modern life. The health impact of using screens has also been questioned, with some researchers suggesting such recommendations being offered despite being counterproductive based on inconsistent research findings to date [
8,
9]. In addition, screen use has been suggested to provide opportunity for individuals to experience meaningful entertainment [
10]. Inconsistencies in the impact of Internet use specifically was highlighted by the clinical report from the American Academy of Pediatrics [
11], and subsequent contradictory research findings [
12]. The report identified ‘Facebook Depression’ as a potential outcome of overuse of social media, however subsequent evidence found no direct link between social media use and symptoms of depression. The continued examination of the health correlates of screen-use through epidemiological studies is imperative.
Previous systematic reviews have predominately focused on the relationship between adolescent sedentary behaviour and physical health outcomes such as body composition, physical activity levels and general well-being indicators [
13‐
15]; however mental health outcomes have mostly been overlooked. Our recent systematic review examined evidence for the relationship between sedentary behaviour and mental health problems among adolescents [
16]. We identified a range of cross-sectional and prospective studies and our findings showed consistent evidence for the relationship between sedentary behaviour (most commonly operationalized as hours of screen-time for leisure) and both depressive symptomatology and psychological distress, with significant gender specific associations. However, studies were of poor quality, often using derived mental health indicators based on a single item and therefore vulnerable to validity concerns. Sampling methods often lacked randomisation and representativeness, and weaknesses were identified in statistical methods including failing to account for important confounding variables. The evidence to date is significantly limited by such methodological weaknesses.
We identified the need for a comprehensive examination of mental health status of a large, representative cohort of adolescents, and associations with time spent using Internet for leisure. The impact of sedentary behaviours on health is developing as an important area in research for physical health, and it is critical that the mental health related evidence base is at the same time strengthened. Evidence exists for the unique experiences of mental health for male and females and defining such differences is another critical aspect of this research. The associations between problematic Internet use (addiction) and mental disorders among Australian adolescents has recently been reported [
17]. However the relationship between duration of time spent online (spanning low to high users) and mental health status requires further investigation.
The aim of this study was to determine the cross-sectional associations between depressive symptoms, psychological distress and time spent using Internet among a community-based, representative Australian adolescent group. This study aimed to answer the research question; what is the relationship between time spent using Internet for leisure and mental health outcomes; depressive symptoms and psychological distress, in a large, representative Australian adolescent group?
Methods
This study examined data from the Second Australian Child and Adolescent Survey of Mental Health and Wellbeing, conducted during 2013–2014 [
5]. The methods summarised in this section have previously been reported [
18]. The national household survey aimed to: capture the mental health and well-being of young people across Australia; provide updated national prevalence estimates of common mental disorders; and report current behavioural patterns such as mental health service use and health risk behaviours.
The survey comprised of two parts; face-to-face interviews with the primary carer, and a self-report questionnaire completed by young people aged 11–17 years within participating households. The current study examined data from the adolescents’ self-reported questionnaire responses, and included some demographic information from the carer-reported data.
The national survey involved a random probability-based sample of 5500 young people aged 4–17 years [
18]. This sample size was chosen to meet reliability estimates for prevalence of mental disorders and behavioural patterns for males and females aged 4–11 years and 12–17 years. Areas eligible for participation were selected using a multi-stage, area-based sample selection procedure to ensure social and economic proportional representation across Australia. Participation was voluntary and all participants gave informed written consent prior to survey completion.
Measures
Participating young people completed the questionnaire privately on a tablet computer. Questionnaires comprised various modules, with those relevant to this study reported below. Details on other information contained in the questionnaire have been reported elsewhere [
18].
Ethics and consent to participate
Ethical approval for this study was granted by Deakin University Human Research Ethics Committee (2016–035). Original ethics for the Second Australian Child and Adolescent Survey of Mental Health and Wellbeing, was provided by the Australian Government Department of Health Ethics Committee and the University of Western Australia Human Research Ethics Committee.
Participation in the survey was voluntary and written consent was required from all participants. Initial verbal consent was obtained from parents or carers for their child’s participation, which was then followed up with paper consent forms from both parents/carers and the young person. Participation was voluntary and all young people were informed that they had the right to withdraw their consent for study participation at any time. Specific protocols were developed to ensure that if any issues arose for participants in response to the survey, there were services available to assist in receiving information and support they may require.
Depressive symptoms
Depressive symptoms were measured by the youth self-report module from the Diagnostic and Statistical Manual of Mental Disorders Version IV (DISC-IV) criteria [
19]. The DSM-IV criteria specify that: at least five symptoms of depression must be present for a minimum of a two-week period; the symptoms cause clinically significant distress; and the symptoms interfere with the child or adolescent’s normal functioning at school, at home or in social settings. Symptoms of depression measured included low mood, loss of pleasure in daily activities, irritable, significant weight loss or gain, loss of appetite, insomnia or hypersomnia, restlessness, fatigue and loss of energy, feelings of worthlessness and inability to concentrate. The major depressive module of the DSM-IV has shown moderate to good diagnostic reliability, and moderate to very good validity in relation to clinically meaningful symptomatology [
19]. Items used to detect depressive disorder showed high internal consistency (Cronbach’s α > 0.70).
Psychological distress
Kessler Psychological Distress scale (K10) [
20,
21] was used to measure distress based on symptoms of anxiety or depressive symptoms experienced in the previous month. Ten items asking about an individual’s emotional experiences such as ‘
during the last 30 days, about how often did you feel tired out for no good reason?’, with the following response options; ‘
none of the time’, ‘
a little of the time’, ‘
some of the time’, ‘
most of the time’ or ‘
all of the time’. Responses were scored 0 to 4, where 0 represented lowest severity (none of the time) and 4 represented highest severity (all of the time). Participants received a score out of 40 with higher scores representing higher levels of psychological distress. Scores of psychological distress were categorised as; 0–5 = low, 6–11 = moderate, 12–19 = high, and 20–40 = very high, as adopted by the Australian Bureau of Statistics based on previous population-based research [
22].
The K10 has been shown to be comparative to other mental health instruments such as mental health component of the General Health Questionnaire, Short Form 12 questionnaire, and the Composite International Diagnostic Interview [
23]. The K10 has shown moderate reliability in the general Australian population [
24]. In this survey data, the K10 Psychological Distress scale showed high internal consistency (Cronbach’s α > 0.70).
Internet use for leisure
Internet use indicators were included in the survey in response to developments in technology as a common means for young peoples’ communication, entertainment and for information sourcing. Internet use excluded time spent on school or work related activities. Internet use was described as Internet accessed on a computer, mobile phone or tablet, and including using social media such as Facebook or Twitter, emailing, looking at websites or chatting online. Questions were focused on time using Internet ‘
not related to schoolwork or for work purposes’. One item asked ‘
on an average weekday approximately how much time do you spend on the internet?’ and another item asked ‘
On an average day on the weekend approximately how much time do you spend on the internet?’ Response categories for both items were; less than 1 h, 1–2 h, 3–4 h, 5–6 h, 7–8 h, 9–10 h, 11 h or more. Responses were categorised into 2 h or less, 3 to 6 h, and 7 h or more, for weekday and weekend responses separately. This was based on Australian guidelines for daily screen-time recommendations (two hours or less per day) [
6] and the growing prevalence of Internet related disorders that are characterised by a large proportion of the day spent online [
25].
In addition, research has indicated both positive and negative mental health impact of screen-use, dependent on time and frequency of use [
26]. We therefore recognised the need to uniquely examine the group reporting 7 h or more to gauge differences between low users, those exceeding screen-time based recommendations currently endorsed by the Australian Government, and those considered very high users. We examined Internet use separately for an average weekday and for an average weekend day as done in previous research examining physical activity and sedentary behavioural research among adolescents [
27], and for increased potential for future practical applications of findings (e.g, school or home based interventions).
Demographics and confounding variables
Participants were asked basic demographics including age and gender as a part of the questionnaire completion. Socio-economic status was taken from the 2011 Australian Bureau of Statistics (ABS) Index of Relative Socio-Economic Disadvantage for the Statistical Area in which the family was living at the time of the survey [
28]. The ABS defines socio-economic advantage and disadvantage as access to material and social resources, and their ability to participate in society [
29]. The index forms a collection of variables designed to assess level of advantage/disadvantage, including; deprivation, poverty, human capital, and economic, social and political opportunities [
29].
Participants were asked to self-report height in metres and weight in kilograms, from which body mass index (BMI) was derived (divided kilograms by height in metres squared). Weight status (thinness/normal weight, overweight/obesity) was calculated using criteria detailed by Cole et al. (2000) [
30]. Cole et al. define age and sex specific cut-points that derive from adult BMI equivalent of 25 kg/m
2 for overweight and 30 kg/m
2 for obesity. Thinness was combined with normal weight due to the low proportion of adolescents categorised as thin (<6%).
Statistical analysis
Analyses were conducted using STATA release V.14.1 (Stata Corp., College Station, Texas, USA, 2015). All variables were checked for missing data. In all cases there were little missing data and case-wise deletion was used where relevant. Descriptive statistics were used to analyse differences between males and females using independent sample Student t-tests and Pearson’s χ2 test where appropriate. Separate multivariate logistic regression models were used to examine the relationship between Internet use and dichotomous mental health outcomes; depressive symptoms (symptoms suggesting depression compared to no depression) and psychological distress (high/very high levels of distress compared to low/moderate levels), stratified by gender. Models were adjusted for potential confounders: age; relative level of socio-economic disadvantage (henceforth referred to as socio-economic status [SES]); and body mass index (BMI). Results were considered statistically significant at p < 0.05.
Results
The survey ran during May 2013 and April 2014. Participant characteristics are reported in Table
1. A total of 2967 adolescents completed the self-report questionnaire and were subsequently included in this study. Adolescents were aged 11–17 years (M = 14.5 years, SD = 2.04 years) and females formed approximately half the sample (48.4%). Overall mean BMI was 21.2 (SD = 4.5) and approximately a quarter (25.9%) of participating adolescents were classified as overweight/obese. Other demographic characteristics indicated that this sample was broadly representative of the wider Australian population when compared to data from the Census of Population and Housing [
18].
Table 1
Characteristics of survey sample
Gender |
Male | 1530 | 51.6 |
Female | 1437 | 48.4 |
Age |
11–15 years | 1615 | 54.4 |
16–17 years | 1352 | 45.6 |
Mean age in years (SD) | 14.6 | (2.0) |
Index of relative socio-economic disadvantage |
Lowest quintile (least advantaged) | 473 | 15.9 |
Second quintile | 554 | 18.7 |
Third quintile | 543 | 18.3 |
Fourth quintile | 676 | 22.8 |
Highest quintile (most advantaged) | 721 | 24.3 |
Parent or carer education |
Bachelor degree of higher | 911 | 32.4 |
Diploma or certificate | 1268 | 45.2 |
Secondary education | 631 | 22.4 |
Weight status |
Normal | 2030 | 74.1 |
Overweight/obesity | 711 | 25.9 |
BMI Mean SD |
Mean (SD) | 21.2 | (4.5) |
Self-reported mental health findings indicated that 9.7% of youths experienced depressive symptoms, with a significantly greater proportion of females (13.9%) compared to males (5.8%) classified with symptoms indicating depression (
p < 0.05) (Table
2). Gender differences were found in average depressive symptomatology scores with females reporting greater total number of symptoms (M = 4.5, SD = 5.9) compared to males (M = 2.8, SD = 4.4). One fifth (21.7%) of adolescents experienced high/very high levels of psychological distress, however a greater proportion of females (28.7%) reported high/very high levels of psychological distress compared to males (15.3%).
Table 2
Proportion of male and female adolescents with major depressive disorder, depressive symptomatology and psychological distress, and time spent using Internet
Major depressive disordera
|
With depression n (%) | 288 (9.7) | 89 (5.8) | 199 (13.9) |
p < 0.05 |
Total depressive symptomatology score Mean (SD) | 3.6 (5.3) | 2.8 (4.4) | 4.5 (5.9) |
p < 0.05 |
Psychological distressb
|
High/Very high | 645 (21.7) | 233 (15.3) | 412 (28.7) |
p < 0.05 |
Total psychological distress score Mean (SD) | 7.6 (6.8) | 6.4 (5.8) | 8.9 (7.6) |
p < 0.05 |
Internet |
Uses Internet n (%) Yes | 2941 (99.2) | 1516 (99.2) | 1425 (99.2) |
NS
|
Internet use on an average weekday n (%) |
NS
|
2 h or less | 1248 (42.4) | 670 (44.2) | 578 (40.6) | |
3–6 h | 1170 (39.8) | 576 (38.0) | 594 (41.7) | |
7 h or more | 523 (17.8) | 270 (17.8) | 253 (17.8) | |
Internet use on an average weekend day n (%) |
NS
|
2 h or less | 1015 (34.5) | 547 (36.1) | 468 (32.8) | |
3–6 h | 1253 (42.6) | 628 (41.4) | 625 (43.9) | |
7 h or more | 673 (22.9) | 341 (22.5) | 332 (23.3) | |
Almost all adolescents (99.2%) reported using the Internet. Over half (57.6%) of the adolescent group reported using the Internet three or more hours on an average weekday, and 65.5% reported three or more hours of Internet use an average weekend day. Close to one fifth reported 7 or more hours of Internet use on an average weekday (17.8%) and on an average weekend day (22.9%). No significant differences existed between Internet usage patterns between male and female subgroups.
Unadjusted logistic regression models demonstrated a significant relationship between experiencing depressive symptoms and higher reported hours of Internet use among females (Table
3). Compared to those reporting two hours or less, female adolescents reporting 3–6 h of Internet use on a weekday were twice as likely to experience depressive symptoms (OR = 2.11, 95% CI = 1.35–3.29,
p < 0.05). Females reporting seven hours or more on a weekday (OR = 2.33, 95% CI = 1.35–4.02,
p < 0.05) or weekend (OR = 2.25, 95% CI = 1.32–3.83,
p < 0.05) had greater increased odds of depressive symptoms. Female adolescents who used the Internet for seven or more hours on an average weekday were more likely to report high/very high levels of psychological distress (OR = 1.73, 95% CI = 1.15–2.60,
p < 0.05), as were both males and females reporting more than two hours Internet use on an average weekend day.
Table 3
Unadjusted (Model 1) and adjusted (Model 2) associations for Internet use for major depressive disorder, depressive symptomatology, and psychological distress (adjusted for age, SES and BMI)
Major depressive disorderb
|
Weekday internet use |
≤ 2 h | Ref | | | Ref | | | Ref | | | Ref | | |
3–6 h | 1.27 | 0.70, 2.33 | 0.429 |
2.11
|
1.35, 3.29
|
0.001
| 1.20 | 0.63, 2.26 | 0.581 |
1.87
|
1.15, 3.02
|
0.011
|
≥ 7 h | 1.99 | 0.97, 4.04 | 0.057 |
2.33
|
1.35, 4.02
|
0.002
| 1.81 | 0.86, 3.83 | 0.120 |
2.09
|
1.16, 3.76
|
0.014
|
Weekend internet use |
≤ 2 h | Ref | | | Ref | | | Ref | | | Ref | | |
3–6 h | 0.95 | 0.51, 1.73 | 0.861 | 1.33 | 0.83, 2.10 | 0.232 | 0.94 | 0.50, 1.78 | 0.847 | 0.93 | 0.57, 1.51 | 0.779 |
≥ 7 h | 1.62 | 0.80, 3.27 | 0.177 |
2.25
|
1.32, 3.83
|
0.003
| 1.47 | 0.70, 3.07 | 0.306 | 1.62 | 0.93, 2.82 | 0.088 |
Age | | | | | | |
1.20
|
1.04, 1.37
|
0.010
|
1.38
|
1.24, 1.53
|
0.000
|
SESc
| | | | | | | 1.02 | 0.87, 1.19 | 0.795 | 0.96 | 0.86, 1.08 | 0.521 |
BMI | | | | | | |
1.06
|
1.00, 1.11
|
0.020
| 1.02 | 0.98, 1.06 | 0.269 |
Psychological distressd
|
Weekday internet use |
≤ 2 h | Ref | | | Ref | | | Ref | | | Ref | | |
3–6 h | 1.00 | 0.69, 1.46 | 0.986 | 1.36 | 0.99, 1.86 | 0.055 | 0.93 | 0.62, 1.39 | 0.725 | 1.16 | 0.83, 1.62 | 0.390 |
≥ 7 h | 1.53 | 0.97, 2.43 | 0.068 |
1.73
|
1.15, 2.60
|
0.008
| 1.45 | 0.89, 2.37 | 0.135 | 1.54 | 1.00, 2.39 | 0.051 |
Weekend internet use |
≤ 2 h | Ref | | | Ref | | | Ref | | | Ref | | |
3–6 h |
1.47
|
1.00, 2.16
|
0.050
|
1.54
|
1.11, 2.13
|
0.009
| 1.39 | 0.93, 2.08 | 0.437 | 1.39 | 0.98, 1.97 | 0.065 |
≥ 7 h |
2.28
|
1.43, 3.68
|
0.001
|
2.67
|
1.78, 4.01
|
0.000
|
2.23
|
1.36, 3.65
|
0.002
|
2.38
|
1.55, 3.67
|
0.000
|
Age | | | | | | | 1.03 | 0.95, 1.12 | 0.437 |
1.20
|
1.11, 1.29
|
0.000
|
SES | | | | | | | 1.06 | 0.95, 1.17 | 0.302 | 0.98 | 0.89, 1.07 | 0.603 |
BMI | | | | | | |
1.03
|
1.00, 1.07
|
0.047
|
1.04
|
1.01, 1.07
|
0.008
|
After adjusting for potential confounders (age, SES, and BMI), the relationship between Internet use and depressive symptoms remained significant for female adolescents reporting 3–6 h (OR = 1.87, 95% CI = 1.15–3.02, p < 0.05) and seven or more hours (OR = 2.09, 95% CI = 1.16–3.76, p < 0.05) on an average weekday. Psychological distress appeared more frequent among female adolescents reporting seven or more hours Internet use on a weekday, however this relationship did not reach significance (p = 0.051). Seven or more hours of Internet use on the weekend doubled the odds of experiencing high/very high psychological distress for males (OR = 2.23, 95% CI = 1.36–3.65, p < 0.05) and females (OR = 2.38, 95% CI = 1.55–3.67, p < 0.05) compared to adolescents reporting two hours or less.
Age increased likelihood of depressive symptoms in both males (OR = 1.20, 95% CI = 1.04–1.37, p < 0.05) and females (OR = 1.38, 95% CI = 1.24–1.53, p < 0.05), and psychological distress in females (OR = 1.20, 95% CI = 1.11–1.29, p < 0.05). BMI was positively associated with depressive symptoms among males (OR = 1.06, 95% CI = 1.00–1.11, p < 0.05) and psychological distress in both males (OR = 1.03, 95% CI = 1.00–1.07, p < 0.05) and females (OR = 1.04, 95% CI = 1.01–1.07, p < 0.05).
Acknowledgements
Data was derived from the second Australian Child and Adolescent Survey of Mental Health and Wellbeing to which over 6000 families generously gave their time.