J Youth Adolescence

DOI 10.1007/s10964-014-0176-x

EMPIRICAL RESEARCH

Adolescents’ Electronic Media Use at Night, Sleep Disturbance,

and Depressive Symptoms in the Smartphone Age
Sakari Lemola • Nadine Perkinson-Gloor •
Serge Brand • Julia F. Dewald-Kaufmann •
Alexander Grob

Received: 30 April 2014 / Accepted: 16 August 2014

 Springer Science+Business Media New York 2014

Abstract Adolescence is a time of increasing vulnera- symptoms. 362 adolescents (12–17 year olds, M = 14.8,
bility for poor mental health, including depression. Sleep SD = 1.3; 44.8 % female) were included and completed
disturbance is an important risk factor for the development questionnaires assessing sleep disturbance (short sleep
of depression during adolescence. Excessive electronic duration and sleep difficulties) and depressive symptoms.
media use at night is a risk factor for both adolescents’ Further, participants reported on their electronic media use
sleep disturbance and depression. To better understand the in bed before sleep such as frequency of watching TV or
interplay between sleep, depressive symptoms, and elec- movies, playing video games, talking or text messaging on
tronic media use at night, this study examined changes in the mobile phone, and spending time online. Smartphone
adolescents’ electronic media use at night and sleep asso- ownership was related to more electronic media use in bed
ciated with smartphone ownership. Also examined was before sleep, particularly calling/sending messages and
whether sleep disturbance mediated the relationship spending time online compared to adolescents with a
between electronic media use at night and depressive conventional mobile phone. Smartphone ownership was
also related to later bedtimes while it was unrelated to sleep
disturbance and symptoms of depression. Sleep disturbance
S. Lemola (&)
N. Perkinson-Gloor
A. Grob partially mediated the relationship between electronic
Department of Psychology, University of Basel, Missionsstrasse media use in bed before sleep and symptoms of depression.
62, 4055 Basel, Switzerland
Electronic media use was negatively related with sleep
e-mail: [email protected]
duration and positively with sleep difficulties, which in turn
N. Perkinson-Gloor
were related to depressive symptoms. Sleep difficulties
e-mail: [email protected]
were the more important mediator than sleep duration. The
A. Grob
results of this study suggest that adolescents might benefit
e-mail: [email protected]
from education regarding sleep hygiene and the risks of
S. Brand electronic media use at night.
Center for Affective, Stress and Sleep Disorders, Psychiatric
Clinics of the University of Basel, Wilhelm Klein-Strasse 27,
Keywords Electronic media use in bed before sleep

4012 Basel, Switzerland
e-mail: [email protected] Smartphone
Sleep
Sleep duration
Sleep difficulties

Depressive symptoms
Sleep hygiene
Adolescence
S. Brand
Division of Sport Science, Department of Sport, Exercise and
Health, Faculty of Medicine, University of Basel, Basel,
Switzerland Introduction

J. F. Dewald-Kaufmann The aim of this study is to examine changes in electronic

Department of Paediatric Endocrinology, Dr. von Hauner
media use in bed and adolescents’ sleep associated with
Children’s Hospital, Ludwig Maximilian University,
Geschwister-Scholl-Platz 1, 80539 Munich, Germany smartphone ownership. Moreover, we test whether elec-
e-mail: [email protected] tronic media use in bed before sleep is associated with

123
 J Youth Adolescence

depressive symptoms and whether this relationship is studies have been identified which consistently reported
mediated by sleep disturbance including short sleep dura- later bed times, shorter sleep duration, and longer sleep
tion and sleep difficulties. To do so, we follow a theoretical latency to be related to greater use of these electronic
model proposed by Cain and Gradisar (2010) suggesting media. Two out of three studies also reported more daytime
that increased electronic media use—particularly in the sleepiness/tiredness (Van den Bulck 2004; Eggermont and
bedroom before sleep—is related to sleep disturbance. Van den Bulck 2006), while one study could not confirm
Moreover this model suggests that sleep disturbance in turn this relationship (Li et al. 2007). With regard to mobile
is associated with impaired daytime functioning. Despite phone use and sleep, seven studies had been conducted till
important work in this area, research has not kept pace with 2010 (Cain and Gradisar 2010). Two studies found shorter
changes in technology and the broad use of such electronic sleep duration being related to greater mobile phone use
devices as smartphones. Therefore, to the authors’ best (Harada et al. 2002; Punamäki et al. 2007), while one study
knowledge this study is the first that addresses changes in could not confirm this link (Yen et al. 2008); three studies
adolescents’ electronic media use at night, sleep, and found that greater mobile phone use was associated with
mental health associated with the use of smartphones. increased daytime sleepiness/tiredness (Van den Bulck
2003, 2007; Söderqvist et al. 2008), while no relationship
Electronic Media Use and Adolescents’ Sleep was found with sleep latency (Gaina et al. 2005) or sleep
difficulties (Söderqvist et al. 2008; Yen et al. 2008).
Adolescence is a time of changing sleep patterns with a More recent studies regarding adolescents’ bedtime
pronounced shift of bedtimes to later in the evening. For mobile phone use include two surveys conducted in Japan
many adolescent students, this results in sleep deprivation examining the relation of mobile phone communication
during the school week and sleeping in on weekends after lights out with sleep (Munezawa et al. 2011; Oshima
(Crowley et al. 2007; National Sleep Foundation 2006; et al. 2012). Munezawa et al. (2011) found both texting and
Perkinson-Gloor et al. 2013; Lemola et al. 2012). The calling after lights were turned off to be related with more
delay of bedtime during adolescence is considered to be sleep disturbance (including short sleep duration, sub-
due to both biological maturation and environmental fac- jective poor sleep quality, excessive daytime sleepiness,
tors (Crowley et al. 2007). Among the environmental fac- and insomnia symptoms) controlling several confounders
tors that may delay bedtimes, electronic media use in the (including gender, grade-level, alcohol-drinking, smoking,
evening has been suggested to play an outstanding role eating breakfast, extracurricular activities, and mental
(Cain and Gradisar 2010). health). Oshima et al. (2012) reported mobile phone use
During the last several decades, there has been a par- after lights were turned off to be negatively related to sleep
ticularly pronounced increase in the use of electronic media duration only in younger adolescents ages 13–15 years,
during leisure time among children and adolescents. while no such relation was found in 16–18 year olds.
Moreover, there is clear increase in media use from Taken together, there is a large body of evidence that
childhood to adolescence (Rideout et al. 2010) and many electronic media use is related to disturbed sleep. However,
adolescents consider the various kinds of electronic media the published studies to date have all been conducted
use as their favorite leisure time activity (Willemse et al. before the latest electronic revolution affecting adoles-
2012). Nowadays, more than half of adolescents from cents’ lives has taken place.
technologically advanced countries report to consume
electronic media on most evenings during the last hour The Possible Role of Smartphones for Adolescents’
before they go to bed (National Sleep Foundation 2011) Electronic Media Use
and for more than two thirds the last activity of the day was
related to electronic media use at least 3 times per week The availability of smartphones may change adolescents’
(Kubiszewski et al. 2013). patterns of electronic media consumption profoundly. First,
There is a growing body of evidence that electronic when having access to wireless Internet, smartphones allow
media use during adolescence is related to later bedtimes, communication with peers without charge, by using for
shorter sleep duration, and sleep disturbance (Cain and instance Internet based messenger applications such as
Gradisar 2010, for a systematic review). The greatest WhatsApp. Particularly the cost of calling and sending text
attention had been addressed towards the relationship messages was a limiting factor of adolescents’ mobile
between TV-consumption and sleep; of 20 studies exam- phone communication before smartphones became easily
ining this link, which were identified by Cain and Gradisar available. During that time adolescents applied various
(2010), 17 have found a significant relationship between tricks to avoid charges related to calling or texting (Prezza
the amount of watching TV and poor sleep. With regard to et al. 2004). For instance, ‘‘ringing’’ was a procedure
the use of computers, internet, or video game playing, 15 involving one adolescent dialing a friend but interrupting

123
J Youth Adolescence

the call before the other person could take the phone. other possible activities and sleep (Kubey 1986; Van den
Hence,—and without charge—the other mobile phone Bulck 2004).
displayed who was calling, whereby telling that person ‘‘I As a second mechanism, it has been proposed that
was thinking of you.’’ Another more elaborate trick was to electronic media use before sleep may increase mental,
answer ‘‘yes’’ or ‘‘no’’ questions by the number of phone emotional, or physiological arousal. This possible mecha-
ring signals (e.g., one signal meant ‘‘yes’’ and two signals nism has been examined by six experimental studies that
meant ‘‘no’’). The availability of smartphones and wireless all examined the effects of playing video games. In a first
Internet access, however, make such practices obsolete. study, Higuchi et al. (2005) studied young male adults who
Beyond replying to simple questions, smartphones and the played an exciting video game late at night or performed
access to wireless Internet offer several new opportunities simple tasks with low mental load in front of a screen as a
to communicate free of charge such as, for instance, control condition. Playing video games slightly increased
sharing pictures and short movies with a predefined group sleep latency and heart rate, and it decreased subjective
of friends. Beside the decrease of costs related to com- sleepiness compared to a control condition. In a similar
munication, smartphones are also more convenient to use vein, Dworak et al. (2007) found that video game playing
while lying in bed to surf the Internet or to watch videos resulted in longer sleep latency and poorer memory per-
from Youtube as smartphones have the advantage of being formance on the following day in adolescent boys. More-
lighter and handier than for instance notebook computers. over, an Australian study found that playing video games in
Their superior handiness compared to other electronic the evening as compared to more passive watching of a
media thus make smartphones particularly practical to use movie decreased male adolescents’ sleepiness and
when already lying in bed. While being convenient, these increased sleep latency (although only to a slight degree;
points also bear the risk that adolescents increase their use Weaver et al. 2010). However, physiological arousal was
of the mobile phone in general and when lying in the bed. not affected in this study. A second Australian study also
The rate of smartphone ownership varies by Nation and found moderately reduced sleep quality but no effect on
age group, and is currently rapidly increasing. In the United heart rate among adolescents with previous video game
States, for instance, the percentage of the population who experience (King et al. 2013). Finally, in two Swedish
owned a smartphone increased from 35 % in 2011 to 56 % studies, violent video game playing was compared with
in 2013 (Smith 2013). In Switzerland, the number of the non-violent video game playing. The first study, which
12–19 year olds who owned a smartphone increased from included male adolescents with rather low experience with
47 % in 2010 to 79 % in 2012 (Willemse et al. 2012); the violent video games, revealed that violent video game
same representative survey study revealed that the number playing induced stronger autonomic responses than non-
of adolescents using the mobile phone to surf the Internet violent gaming (while not affecting subjective sleep quality
increased from 16 % in 2010 to 66 % in 2012. Similar rates or cortisol secretion; Ivarsson et al. 2009a, b). The second
have been reported for Germany, where 25 % of the experiment found differential effects of experimental
12–19 year olds owned a smartphone in 2011 while this exposure to violent versus non-violent video games
number has increased to 72 % in 2013 (Medienpädagogi- according to the previous gaming experience such that
scher Forschungsverbund Südwest 2013); 60 % of the adolescents with low gaming experience showed more
adolescents in Germany also reported to subscribe to a flat- negative effects related to heart rate variability, sleep
rate data plan for their mobile phones. However, scientific quality, and mood after the violent game pointing to the
knowledge whether smartphone ownership affects adoles- importance of the previous experience with video games
cents’ electronic media use in the bed before sleep, and (Ivarsson et al. 2013). Taken together, experimental studies
whether it increases the risk of poor sleep and daytime are in line with a causal role of playing video games for
functioning, is still missing. poor sleep. However, physiological arousal (including
heart rate variability) and neuroendocrine responses
Possible Mechanisms Linking Electronic Media Use (including cortisol secretion) were not confirmed as the
with Poor Sleep mediators of this effect—at least not in experienced video
game players. It is possible that mental and/or emotional
Cain and Gradisar (2010) propose several mechanisms arousal which was not assessed in these studies mediated
through which electronic media use in the evening may the effects on sleep. With regard to other types of elec-
reduce sleep duration and interfere with sleep quality. As a tronic media use, electromagnetic radiation emitted by
first mechanism, electronic media use may displace sleep. mobile phones has been reported to delay melatonin pro-
As an unstructured leisure activity with no fixed starting duction and could therefore be related to later sleep onset
and stopping point, it involves an increased risk of (Wood et al. 2006). Moreover, communication via mobile
expanding and taking up more time, and thus displacing phone when lying in the bed before sleep has also been

123
 J Youth Adolescence

suggested to increase emotional and/or cognitive arousal and protective factors, and to develop efficacious treatment
and might therefore affect sleep (Munezawa et al. 2011). and prevention approaches.
However, this latter mechanism has not yet been tested The comorbidity of depression with sleep problems is
experimentally. very high with 73 % of adolescents with depressive dis-
As a third mechanism, Cain and Gradisar (2010) sug- order also suffering from a sleep disorder (Lui et al. 2007).
gested that light emission of the screens of electronic media Moreover, most studies examining correlations between
devices might affect sleep. In line with this notion, there is sleep disturbance and depressive symptoms in adolescents
evidence from an experimental study with young adults found significant correlations (e.g., Short et al. 2013, Le-
that particularly light emission of modern flat screens with mola et al. 2011). Often researchers assumed a bi-direc-
LED back light technology may interfere with sleep. LED tional relationship between the sleep disturbance and
back light screens emit an increased amount of light in the mental health (e.g., Cortese et al. 2013). However, in a
short wave length spectrum of around 460 nm which recent meta-analysis summarizing longitudinal and treat-
suppresses melatonin secretion in the evening and reduces ment studies that examined the prospective role of sleep
subjective and objective signs of sleepiness (Cajochen et al. disturbance in the development of depression and vice
2011). However, one recent study comparing the impact of versa during adolescence, Lovato and Gradisar (2014)
the three conditions (a) 1 hour of bright tablet screen concluded that sleep disturbance rather acts as a precursor
exposure, (b) 1 h of short-wavelength filtered tablet screen to the development of depression while little support was
exposure, and (c) 1 h of dim tablet screen exposure on found for a predictive role of depressive symptoms in the
adolescents’ pre-bedtime alertness, sleep, and daytime development of sleep disturbance. These findings point to
functioning found only minimal differences (Heath et al. the importance of sleep disturbance in the etiology of
2014). No evidence is yet available indicating whether depression during adolescence.
more than 1 h of exposure to a tablet screen might have a In a related vein, experimental studies have tested the
stronger effect on adolescents’ sleep. causal role of sleep restriction for vulnerability factors that
A fourth mechanism by which particularly mobile are known to be related to depression. A recent experi-
phones in the bedroom may disturb sleep is that incoming mental study showed that restricted sleep to 6.5 h for
messages may wake adolescents up at night. Recently, a 5 days—a regimen mimicking common sleep curtailment
representative survey of the US population revealed that during a school week—resulted in increased self-reports of
18 % of adolescents aged between 13 and 18 years are tension, anxiety, and fatigue as well as greater parent rated
woken up by text messages after sleep onset at least few oppositionality/irritability and poorer emotional regulation
times per week compared to only 10 % of individuals aged in adolescents aged 12–17 years (Baum et al. 2014).
between 30 and 45 years (National Sleep Foundation Likewise, curtailing sleep to 6.5 h on a first night and to
2011). Similar rates of being woken up by incoming text \2 h on a second night showed similar effects involving
messages and calls after lights out have been reported by increased negative affect and decreased positive affect in
adolescents in Belgium (Van den Bulck 2007). In this latter adolescents aged 10–16 years (Dagys et al. 2012; McG-
study, the odds of being very tired during the day strongly linchey et al. 2011; Talbot et al. 2010). In a computerized
increased with the frequency of mobile phone use after analysis of emotional markers in speech, adolescents
lights out. Moreover, also a nation-wide study with Japa- appeared to be even more vulnerable to the effects of sleep
nese adolescents showed that the use of mobile phones deprivation compared to adults (McGlinchey et al. 2011).
after lights out was associated with poor sleep and exces- The mechanism through which short sleep has an impact
sive daytime sleepiness (Munezawa et al. 2011). on emotional and behavioral functioning in adolescents
may involve an increase in negative mood and a decrease
The Role of Sleep for Depression in Adolescence in the ability to regulate emotions (Baum et al. 2014). In
adulthood, a neuroimaging study showed increased amyg-
While severe cases of depression are comparably rare dala activity in sleep deprived individuals as well as
before puberty, the prevalence rate drastically increases decreased connectivity between the prefrontal cortex and
until the end of adolescence (Hankin et al. 1998; Kessler the amygdala (Yoo et al. 2007). Amygdala activity is
et al. 2001). Around 20 % of the population has experi- known to reflect processes related to negative affectivity
enced a depressive episode when turning 18 years old, including anxiety. The increase in amygdala activity and
which also involves a highly increased risk for further decrease in prefrontal control of the amygdala related to
depressive episodes in adulthood (Lewinsohn et al. 1993). sleep deprivation may indicate proneness to negative affect
It is therefore of major interest to understand why this rate and lower ability to regulate negative feelings (Yoo et al.
is increasing during adolescence, to identify possible risk 2007).

123
J Youth Adolescence

The Role of Sleep Disturbance for the Relationship The Current Study
Between Electronic Media Use and Depression
In the current study, we examine whether adolescents who
While it is generally accepted that electronic media use own a smartphone differ from their peers who do not own a
during adolescence has a predictive role in the devel- smartphone regarding their electronic media use in bed
opment of depression (Primack et al. 2009) and that before sleep. Moreover, we examine whether electronic
electronic media use is related to sleep disturbance (Cain media use in the bed is related to sleep disturbance (including
and Gradisar 2010), studies that explicitly examine the short sleep duration and sleep difficulties), and whether short
mediating role of sleep disturbance for the relationship sleep duration and sleep difficulties mediate the relationship
between electronic media use and depressive symptoms between electronic media use in bed and depressive symp-
are rare. Most studies analyzing relationships between toms. In particular we test the following hypotheses.
electronic media use, sleep disturbance, and depressive Possession of a smartphone is related to more electronic
symptoms simultaneously in adolescents (Munezawa media use in general and particularly in bed before sleep
et al. 2011; Oshima et al. 2012) or in young adults (Hypothesis 1). We expect that adolescents with a smart-
(Brunborg et al. 2011) defined one of these variables as a phone more often use electronic media as smartphones allow
covariate rather than a mediator. One study treated to communicate with peers without charge (e.g., by using
depressive symptoms as a covariate when studying the smartphone applications such as WhatsApp or the Facebook
relationship between mobile phone use after lights out application for smartphones). Moreover, due to their small
and sleep disturbance finding that the raw association size smartphones are more convenient to use than for
was attenuated when the covariates (including depressive instance notebook computers when lying in the bed. Due to
symptoms) were controlled (Munezawa et al. 2011). A these reasons, we assume that the advent of smartphones in
second study treated sleep duration as a covariate when adolescents’ bedrooms profoundly change electronic media
studying the relationship between mobile phone use after use in the bed and thus might also affect sleep.
lights out and poor mental health, suicidal ideation, and Electronic media use in bed before sleep is related to
self-injury (Oshima et al. 2012). Similarly as in the higher levels of depressive symptoms (Hypothesis 2). In
former study, the raw associations were attenuated when line with previous research (Brunborg et al. 2011; Lemola
controlling for covariates. Furthermore, also Brunborg et al. 2011; Oshima et al. 2012) we expect that higher
et al. (2011) controlled for depressive symptoms when levels of depressive symptoms in adolescents who more
assessing the relationship between using the mobile often use electronic media in bed before sleep.
phone in the bedroom at night and sleep disturbance in Electronic media use in bed before sleep is related to
young adults. A formal test of the mediation hypothesis shorter sleep on weekday nights and/or sleep difficulties on
that sleep disturbance may mediate the relationship weekday nights (Hypothesis 3). We expect electronic media
between electronic media use and depressive symptoms use in bed before sleep to be related to shorter sleep and/or
was conducted by two studies (Adams and Kisler 2013; sleep difficulties in line with previous research (Cain and
Lemola et al. 2011). Results from Adams and Kisler Gradisar 2010 for a review; Brunborg et al. 2011; Dworak
(2013) support the mediation hypothesis in a sample of et al. 2007; Eggermont and Van den Bulck 2006; Fossum
college students. In contrast, Lemola et al. (2011) could et al. 2014; Higuchi et al. 2005; King et al. 2013; Munezawa
not find mediation of the relationship between habitual et al. 2011; Oshima et al. 2012; Van den Bulck 2007).
video game playing at night and depressive symptoms by Several factors may lead to less sleep in adolescents who use
sleep disturbance in a sample of adolescents and young electronic media in bed before sleep, including that it may
adults who played the online multiplayer role play game replace the time for sleep and it may increase arousal due to
World of Warcraft. the media contents or due to alerting features of the screens
Taken together, there is a wealth of evidence that including brightness and the specific wave-length of LED-
electronic media use before bedtime, sleep disturbance, and backlight screens (Cain and Gradisar 2010).
depressive symptoms in adolescence are interrelated. The relationship between electronic media use in bed
However, studies assessing how the availability of smart- before sleep and depressive symptoms is partly mediated
phones—including the opportunity to cheaply and conve- by sleep duration and/or sleep difficulties (Hypothesis 4).
niently communicate with peers when already lying in the We expect that the relationship between electronic media
bed—changes the use of electronic media in adolescents’ use in bed and depressive symptoms is at least partly due to
bedrooms are missing. Moreover, there is only little sleep disturbance. This expectation is based on the theo-
research studying the hypothesis that disturbed sleep acts retical model proposed by Cain and Gradisar (2010) sug-
as a mediator of the relationship between electronic media gesting that electronic media use has an impact on daytime
use and depressive symptoms in adolescence. functioning via sleep disturbance. Moreover, there is now a

123
 J Youth Adolescence

large body of evidence indicating that sleep disturbance in 362 (92.8 %) were underage, i.e. between 12 and 17 years
adolescence predicts the development of depressive old. Young adults (ages 18–20; n = 28; 7.2 %) were
symptoms (Lovato and Gradisar 2014). excluded from the analyses of the current study in order to
In further analyses, we explore whether the relations address the research questions within underage adoles-
between electronic media use in bed before sleep, sleep cents. Among the remaining 362 students, 200 (55.2 %)
duration, sleep difficulties, and depressive symptoms are were male and 162 (44.8 %) were female. A majority
moderated by age (Additional Research Question 1). indicated to speak only German/Swiss German at home
Moreover, we explore (Additional Research Question 2) (55.8 %), 21.3 % indicated to speak German/Swiss Ger-
which type of electronic media use is most strongly related man and another language at home, and 22.9 % indicated
with sleep duration, sleep difficulties, and depressive to speak only another language at home.
symptoms. We do not propose hypotheses regarding which
media device would be most strongly related to sleep dis- Measures
turbance and depressive symptoms.
Sleep Duration

Method To assess sleep duration, students were asked to indicate

the time they turn off the lights to go to sleep on regular
Procedure school nights (Monday to Thursday night) and the time
they get up in the morning of regular school days (Monday
Participants were recruited from public high schools in to Friday morning). Sleep duration was defined as the time
northwestern Switzerland. Principals of all 42 high schools between the indicated ‘‘lights off’’-time and rise time.
within the area received a letter with information on the
study aims and procedures. Approximately 1 week later, Sleep Difficulties
the principals of the 42 high schools were contacted by
phone to ask if they were interested in participation. Seven Sleep difficulties were assessed with five items from the
high school principals agreed that school classes of their German translation of the Insomnia Severity Index (ISI;
school participated in the study. In total, the seven partic- Bastien et al. 2001; translated by the 3rd author of the
ipating schools consisted of 82 school classes and teachers present article). Reliability and validity of the German
of 32 of those agreed to participate. The 32 school classes version were established in a previous study (Gerber et al.
consisted of 646 students, of which 390 agreed to partici- 2010). The items were answered on a 5-point Likert scale
pate and if underage, had consent from their parents. (0 = not at all/very satisfied and 4 = very much/very dis-
Trained study personnel visited the school classes and satisfied) and assessed difficulties falling asleep and
administered questionnaires on sleep, media consumption maintaining sleep, satisfaction with the current sleep pat-
before going to sleep, and psychological health. Complet- tern and feeling rested after awakening (e.g. ‘‘In the last
ing the questionnaire took approximately 25 min. After 2 weeks, did you have difficulties falling asleep?’’). A
completing the questionnaires students either received an higher mean score represents more sleep difficulties
interventional lesson on sleep hygiene or general infor- (Cronbach’s alpha = .71).
mation on sleep related topics such as dreaming. One
month later, students were visited a second time and Electronic Media Use in Bed Before Sleep
completed the same questionnaires. The present paper
reports data collected at the first school visit and from all To assess electronic media use in bed before sleep stu-
participating students. Data for the present study was dents were asked about their behavior regarding media
gathered between October 2012 and February 2013. Stu- consumption in bed before going to sleep on a regular
dents and parents of underage students gave written school night. Media consumption in bed was assessed
informed consent to participate in the study. The study was with four items assessing how often participants watch
approved by the Ethics Committee of Basel and performed TV or movies, play video games, talk on the phone or
in accordance with the ethical standards laid down in the text, and spend time online on Facebook or in chat rooms
Declaration of Helsinki. or surf the Internet before going to sleep and while
already in bed. Answer categories ranged from 1 (never)
Participants to 5 (most of the time to always; at 5–7 days per week). A
higher sum score represents more electronic media con-
In total, 390 adolescents aged 12–20 years completed the sumption in bed before going to sleep (Cronbach’s
questionnaires at the first data assessment time of whom alpha = .70).

123
J Youth Adolescence

Daily Duration of Electronic Media Use in General additionally employed the SPSS procedure Indirect by
Preacher and Hayes (2008), assessing the models with both
To assess the duration of electronic media use in general mediators simultaneously via bootstrapping. All regression
the students indicated on four items how many minutes and models controlled adolescents’ age and gender. Fourth, we
hours they (1) watch TV, (2) play video games, (3) spend tested whether any of the regression models of the previous
time online on Facebook, (4) spend time on the Internet in steps were moderated by adolescents’ age (test of Addi-
total on an ordinary school day. Moreover, they indicated tional Research Question 1) following the procedure sug-
how many text messages they send per day (including short gested by Aiken and West (1991). Finally, further
messages, WhatsApp messages etc.). regression models specified simultaneous entry of the four
different types of electronic media consumption in bed
Depressive Symptoms before sleep and the variable whether the mobile phone
was switched off during the night according to the ‘‘step-
Depressive symptoms were assessed with six items taken wise’’ algorithm (probability-of-F-to-enter: 0.05; prob-
from the short version of the ‘‘Allgemeine Depressionss- ability-of-F-to-remove: 0.10; test of Additional Research
kala’’ (ADS-K; Hautzinger and Bailer 1993), the German Question 2). Again all regression models controlled ado-
version of the Center of Epidemiological Studies Depres- lescents’ age and gender.
sion Scale (CES-D; Radloff 1977), including ‘‘feeling
depressed’’, ‘‘feeling everything one does is an effort’’,
‘‘feeling fearful’’, ‘‘feeling sad’’, ‘‘that one could not get Results
going’’, and ‘‘that one enjoyed life’’ (reverse coded). Thus,
no item assessing sleep disturbance was included. The Table 1 shows descriptive statistics of the study variables
items were answered on a 4-point Likert scale ranging from and the comparisons of the electronic media use between
0 (occurred never or rarely) to 3 (occurred most of the time the group owning a smartphone (n = 299; 82.6 % of the
or always) reflecting how often the symptoms were expe- sample) and the group owning a conventional mobile
rienced during the preceding week. Higher scale mean phone (n = 51; 14.4 % of the sample) (test of Hypothesis
scores reflect higher levels of depressive symptoms 1). Only a minority of 10 adolescents (2.8 %) indicated not
(Cronbach’s alpha = .73). to own a mobile phone and were therefore excluded from
this comparison. Adolescents owning a smartphone were
Statistical Analysis somewhat older and more often spoke a different language
than German at their homes than their peers who had a
First, mean value comparisons and frequency comparisons conventional mobile phone. With regard to their electronic
of the study variables between smartphone owners and media consumption in general, they spent more time on the
owners of a conventional mobile phone were conducted by Internet and on Facebook, and they sent several times more
analysis of variance (ANOVA) and v2 statistics (test of text messages per day, while they did not spend more time
Hypothesis 1). Second, Pearson correlations were calcu- watching TV or playing video games. Regarding their
lated. The Pearson correlations represent preliminary electronic media use in bed before sleep, they more often
analyses to present the zero-order relations between the watched videos, communicated by phone or text message,
study variables. Third, regression analyses were conducted more often spent time online, and more often had their
as suggested by Baron and Kenny (1986) to test mediation. phones switched on during the night, while they did not
This included the following steps: (a) Prediction of the significantly differ regarding video game playing in the
dependent variable (depressive symptoms) by the inde- bed. Furthermore, smartphone owners switched off the
pendent variable (electronic media use in bed; test of lights later at night than their peers with conventional
Hypothesis 2). (b) Prediction of the mediators (sleep mobile phones, while they did not differ significantly
duration and sleep difficulties) by the independent variable regarding sleep duration, sleep difficulties, and symptoms
(electronic media use in bed; test of Hypothesis 3). of depression.
(c) Prediction of the dependent variable (depressive Table 2 presents Pearson correlations for participants’
symptoms) by both the independent variable (electronic age, gender, sleep, depressive symptoms, and variables of
media use in bed) and the mediators (sleep duration and electronic media use at night. Age was negatively related to
sleep difficulties; test of Hypothesis 4; for partial mediation sleep duration and it was positively related to depressive
it is required that the direct relation between the indepen- symptoms and all types of electronic media use at night
dent variable and the dependent variable is reduced by except for playing video games. Girls had higher levels of
inclusion of the mediator to the model). In order to also depressive symptoms and reported more calling and/or text
formally test the significance of the indirect effects we then messaging in bed before sleep, while they also reported less

123
 J Youth Adolescence

Table 1 Description of the study variables and comparisons between adolescents with a smartphone and with a conventional mobile phone
Total sample Adolescents with a Adolescents with a Statisticsa
(N = 362) smartphone (n = 299) conventional mobile phone
(n = 51)

n/mean (%/SD) n/mean (%/SD) n/mean (%/SD)

Age (years), mean (SD) 14.82 (1.26) 14.9 (1.2) 14.3 (1.3) F = 12.57, p \ 0.001
Gender (female) n (%) 162 (44.8) 136 (45.5) 23 (44.2) v2 = 0.03, p = 0.87
Language spoken at home (German only) n (%) 202 (55.8) 155 (51.8) 37 (71.2) v2 = 6.67, p = 0.01
Sleep on weekdays
Time of switching lights off, mean (SD) 10:26 (0:55) 10:30 (0:55) 10:10 (0:48) F = 5.75, p = 0.02
Time of getting up, mean (SD) 06:30 (0:23) 06:31 (0:24) 06:25 (0:20) F = 3.32, p = 0.07
Sleep duration (hours), mean (SD) 8:04 (0:54) 8:01 (0:55) 8:14 (0:52) F = 2.54, p = 0.11
Sleep difficulties, mean (SD) 2.20 (0.68) 2.21 (0.68) 2.13 (0.61) F = 0.73, p = 0.39
Depressive symptoms, mean (SD) 1.58 (0.48) 1.59 (0.48) 1.53 (0.44) F = 0.79, p = 0.37
Amount of media use on weekdays, hours, mean (SD)
Watching TV 1:41 (1:22) 1:44 (1:23) 1:32 (1:22) F = 0.83, p = 0.36
Playing video games 0:53 (1:00) 0:54 (1:01) 0:46 (0:46) F = 0.79, p = 0.37
To be on the Internet (total) 1:52 (1:39) 2:00 (1:42) 1:08 (0:59) F = 12.36, p \ 0.001
To be on Facebook 0:48 (1:07) 0:54 (1:11) 0:17 (0:28) F = 13.54, p \ 0.001
Number of text messages sent per day, mean (SD) 71.11 (89.53) 84.63 (92.89) 7.06 (10.49) F = 36.10, p \ 0.001
During the night mobile phone is…b n (%)
Switched on and at the bed 204 (57.6) 181 (60.5) 21 (41.2) v2 = 24.28, p \ 0.001
Switched on but not at the bed 75 (21.2) 68 (22.7) 6 (11.8)
Switched off or not in bedroom 75 (21.2) 50 (16.7) 24 (47.1)
Media use in bed before sleep, mean (SD) 9.49 (4.18) 10.10 (4.18) 6.76 (2.76) F = 30.20, p \ 0.001
Media use in bed before sleep n (%)
Watching TV/videos
Never 183 (50.8) 144 (48.3) 34 (66.7) U = 6,004.5, p = 0.01
Once a week or less 61 (16.9) 49 (16.4) 7 (13.7)
Twice a week 43 (11.9) 37 (12.4) 5 (9.8)
3–4 times a week 29 (8.1) 27 (9.1) 2 (3.9)
5–7 times a week 44 (12.2) 41 (13.8) 3 (5.9)
Playing video games
Never 245 (68.1) 200 (67.1) 36 (70.6) U = 7,225.5, p = 0.50
Once a week or less 47 (13.1) 40 (13.4) 6 (11.8)
Twice a week 36 (10.0) 27 (9.1) 8 (15.7)
3–4 times a week 22 (6.1) 21 (7.0) 1 (2.0)
5–7 times a week 10 (2.8) 10 (3.4) 0 (0.0)
Communication by phone or text message
Never 92 (25.6) 57 (19.1) 26 (51.0) U = 3,743.0, p \ 0.001
Once a week or less 48 (13.3) 39 (13.1) 9 (17.6)
Twice a week 42 (11.7) 31 (10.4) 10 (19.6)
3–4 times a week 64 (17.8) 60 (20.1) 3 (5.9)
5–7 times a week 114 (31.7) 111 (37.2) 3 (5.9)
To be online (on Facebook, Chatroom, etc.)
Never 146 (40.6) 100 (33.6) 37 (72.5) U = 4,359.0 p \ 0.001
Once a week or less 52 (14.4) 49 (16.4) 3 (5.9)
Twice a week 41 (11.4) 34 (11.4) 5 (9.8)
3–4 times a week 55 (15.3) 50 (16.8) 5 (9.8)
5–7 times a week 66 (18.3) 65 (21.8) 1 (2.0)
a
Statistics refer to the comparison between adolescents who own a smartphone versus adolescents who own a conventional mobile phone
b
The question whether the mobile phone was switched on during the night was also replied by three of the individuals who indicated not to own a mobile phone

123
J Youth Adolescence

Table 2 Pearson correlations between study variables

1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

1. Age
2. Gender 2.14
3. Sleep duration on weekdays 2.42 2.05
4. Sleep difficulties .01 .08 2.16
5. Depressive symptoms .13 .12 2.22 .44
6. Mobile switched on after lights out .42 2.07 2.39 .14 .15
7. Media use in bed (scale) .31 2.02 2.40 .20 .28 .44
8. Watching TV/videos in bed .19 -.14 2.24 .12 .19 .25 .70
9. Playing video games in bed .07 2.21 2.13 .02 .16 .16 .55 .38
10. Calling/text messaging in bed .34 .13 2.37 .18 .19 .45 .81 .33 .23
11. To be online (Facebook, Chat etc.) in bed .27 .08 2.37 .22 .25 .39 .81 .36 .21 .68
Pairwise exclusion of missing values. Gender coding: male = 1, female = 2. Coding of ‘‘Mobile switched on after lights out’’: mobile switched
off after lights out (or not in the bedroom) = 1, mobile switched on but not at the bed = 2, mobile switched on and at the bed = 3. Media use in
bed (scale) represents the mean score of the four variables watching TV/videos in bed, playing video games in bed, calling/text messaging in bed,
and to be online (Facebook, Chat etc.) in bed
Italics p \ 0.05; italics and bold p \ 0.01; italics, bold, and underlined p \ 0.001

Fig. 1 Mediation of the relationship between electronic media use in controlled for age and gender. The coefficient in brackets represents
bed before sleep and depressive symptoms by sleep duration. the standardized regression coefficient when the mediator (sleep
Coefficients are standardized regression coefficients that are duration) is also in the model. **p \ 0.01; ***p \ 0.001

watching of TV and videos and less often played video age and gender. The direct effect of electronic media use in
games in bed compared to boys. Sleep duration on week- bed on depressive symptoms remained significant after
days was negatively related to depressive symptoms and all inclusion of the mediator to the regression model, but was
types of electronic media use at night, while sleep diffi- substantially reduced when sleep difficulties as the medi-
culties were positively related to these variables (except for ator came to the model (from b = .26, t = 4.84,
a non-significant relation to video game playing in bed). DR2 = 0.060; p \ 0.001 to b = .17, t = 3.49, DR2 =
Depressive symptoms were positively related to all types of 0.026; p \ 0.001) and somewhat reduced when sleep
electronic media use at night. duration as the mediator came to the model (from b = .26,
Figure 1 represents the regression models testing the t = 4.84, DR2 = 0.060; p \ 0.001 to b = .23, t = 4.10,
relations between electronic media use in bed before sleep DR2 = 0.043; p \ 0.001) indicating partial mediation (test
(independent variable), sleep duration (mediator), and of Hypothesis 4). Testing the indirect effect with both
depressive symptoms (dependent variable), while Fig. 2 mediators in the model revealed only significance of the
shows the same model with sleep difficulties as the medi- indirect path via sleep difficulties (B = 0.08, SE = 0.02,
ator. The regression models revealed that electronic media bias corrected CI 0.04–0.13 based on 3,000 bootstrap
use in bed before sleep was related to higher levels of samples) but not sleep duration (B = 0.02, SE = 0.02, bias
depressive symptoms (b = .26, t = 4.84, p \ 0.001; test corrected CI -0.01 to 0.05).
of Hypothesis 2). Moreover, regression models revealed Testing the age-moderation of the associations between
that electronic media use in bed before sleep was also electronic media use in bed before sleep, sleep duration,
related to shorter sleep on weekday nights (b = -.29, sleep difficulties, and depressive symptoms showed no
t = -6.00, p \ 0.001) and sleep difficulties (b = .21, significant interactions with age (all p values C0.10;
t = 3.91, p \ 0.001; test of Hypothesis 3) controlling for additional Research Question 1).

123
 J Youth Adolescence

Fig. 2 Mediation of the relationship between electronic media use in controlled for age and gender. The coefficient in brackets represents
bed before sleep and depressive symptoms by sleep difficulties. the standardized regression coefficient when the mediator (sleep
Coefficients are standardized regression coefficients that are difficulties) is also in the model. ***p \ 0.001

Testing which type of electronic media use in bed was In addition to the increase in the prevalence of depres-
most strongly associated with sleep duration, sleep diffi- sion with the transition from childhood to adolescence,
culties, and depressive symptoms stepwise regression was there is also a secular trend of an increasing incidence of
conducted (additional Research Question 2). Being online depression during adolescence since the 1960s (Kessler
(Facebook, Chat etc.) in bed (b = -.21, t = - 4.13, et al. 2001). A similar but decreasing secular trend has been
p \ 0.001) and having the mobile phone switched on at reported for adolescents’ sleep duration (Dollman et al.
night (b = .20, t = 3.71, p \ 0.001) (but not watching TV, 2007; Iglowstein et al. 2003). Although speculative, it is
playing video games, and calling/text messaging in bed) possible that these secular trends have a common cause.
were related with sleep duration on weekdays. Being online One such possible cause is the increasing availability of
(Facebook, Chat etc.) in bed before sleep (b = .22, electronic devices that has changed adolescents’ every day
t = 4.10, p \ 0.001) (but not watching TV, playing video lives profoundly since the 1960s. There is now a large body
games, calling/text messaging in bed, and having the of evidence confirming that electronic media use particu-
mobile switched on at night) was related with sleep diffi- larly is related and may cause sleep disturbance (Cain and
culties. Moreover, being online (Facebook, Chat etc.) in Gradisar 2010). Moreover, there is also evidence of a
bed (b = .19, t = 3.43, p \ 0.001) and playing video predictive role of media use for the development of
games (b = .15, t = 2.86, p = 0.005) (but not watching depression (Primack et al. 2009). A theoretical model
TV, calling/text messaging in bed, and having the mobile proposed by Cain and Gradisar (2010) suggests that elec-
switched on at night) were related with depressive tronic media use may cause sleep disturbance, which in
symptoms. turn may cause daytime dysfunction such as increased
depressive symptoms. One aim of the current study was to
test whether adolescents’ sleep disturbance mediates the
Discussion relationship between electronic media use at night and
symptoms of depression. A further aim of the current study
Adolescence is a time of increasing vulnerability for was to examine how smartphones have changed adoles-
severe mental health disorders such as depression. Epi- cents’ electronic media use at night as during the latest
demiological studies show that the incidence of new years the prevalence of smartphones has escalated among
cases of depression drastically increases with puberty adolescents in Switzerland (Willemse et al. 2012). Due to
(Hankin et al. 1998; Kessler et al. 2001). As the recur- the large functional range of smartphones, their handiness,
rence rate during adulthood is high, prevention of and the inexpensiveness of accessing the Internet, we
depression with onset during adolescence is of major hypothesized that owning a smartphone would change
importance (Lewinsohn et al. 1993). During adolescence, electronic media use at night and increase sleep disturbance
sleep patterns change with a pronounced shift of bed- in adolescents.
times to later in the evening, increasing sleep deprivation Our results show that adolescents who owned a smart-
during the school week and sleeping in on weekends phone sent a lot more text messages and spent more time
(Crowley et al. 2007). Importantly, there is growing on the Internet and on Facebook per day than their peers
evidence that sleep disturbance in adolescence may pre- with conventional mobile phones, while they did not watch
dict the development of depression (Lovato and Gradisar more TV or play video games. At night when lying in the
2014). In order to prevent the burden of depression with bed, adolescents who owned a smartphone were a lot more
onset during adolescence, it appears to be a promising likely to communicate by calling or text messaging or to be
venue to identify and address sleep disturbance (Lovato online (e.g., on Facebook or chat), and somewhat more
and Gradisar 2014). likely to watch TV or videos (which may also be done on

123
J Youth Adolescence

the smartphone screen), while they were not more likely to or text messaging, and spending time online were corre-
play video games. Furthermore, adolescents who owned a lated with sleep difficulties. However, no such correlation
smartphone were more likely to go to bed later, although was found between playing video games in bed before
their sleep duration was not shorter on average, and they sleep and sleep difficulties. The latter finding is surprising,
did not report more sleep difficulties or symptoms of although there are also other reports that adolescents who
depression than their peers who owned a conventional are experienced in playing video games may not neces-
mobile phone. These findings are in line with our first sarily have poor sleep when playing before sleep (Lemola
hypothesis that adolescents with smartphones report more et al. 2011). The electronic media use type that was most
electronic media use when lying in the bed and are con- strongly related with sleep disturbance and depressive
sistent with the interpretation that the availability of new symptoms was spending time online (e.g., on Facebook or
electronic devices may strongly change adolescents’ habits. chatrooms) when lying in the bed before sleep.
However and interestingly, the effect did not generalize to Depression during adolescence is a severe condition
more sleep disturbance or symptoms of depression. This involving a high recurrence rate during young adulthood
points to the interpretation that it is more important when, (Lewinsohn et al. 1993). As the incidence is drastically
how much, and how often adolescents use their electronic increasing during adolescence and prevalence rates have
devices than whether they use a smartphone or a conven- also risen across the last few decades (Kessler et al. 2001)
tional mobile phone. it is of major interest to provide ways of prevention during
Consistent with our second hypothesis, we found that adolescence. A promising venue for prevention of depres-
electronic media use in the bed before sleep was related to sion during adolescence is by improving adolescents’ sleep
higher levels of depressive symptoms. This is in line with quality as there is evidence that disturbed sleep is a vul-
several studies showing that electronic media use in gen- nerability factor for the development of depressive symp-
eral (e.g., Primack et al. 2009) and at night are related to toms during adolescence (Lovato and Gradisar 2014).
depressive symptoms in adolescents (Lemola et al. 2011; Recommendations to prevent sleep disturbance in ado-
Oshima et al. 2012). Moreover, these results are thus also lescents might involve the following. First, as electronic
in line with similar reports concerning young adults media use before sleep is related with sleep disturbance and
(Brunborg et al. 2011; Thomée et al. 2011). Consistent with depressive symptoms many adolescents might benefit from
our third hypothesis and in line with a large body of evi- improved sleep hygiene involving a reduction in electronic
dence (e.g., Cain and Gradisar 2010), we found that elec- media use before and at bedtime. In our study, more than
tronic media use in the bed before sleep was related to 80 % of the adolescents owned a smartphone and more
shorter sleep duration and more sleep difficulties. More- than a third of these adolescents reported to write text
over, consistent with our fourth hypothesis our findings are messages or were still online most of the nights when they
in line with the theoretical model involving the mediation were already in bed before sleep. This reflects the endemic
of the relationship between electronic media use and of electronic media use among adolescents nowadays and
depressive symptoms by sleep disturbance (Cain and the decrease of parental control of their children’s media
Gradisar 2010). Particularly sleep difficulties substantially consumption. Sleep hygiene education for adolescents
decreased the relationship between electronic media use might be taught in classrooms and parents could be
and depressive symptoms when added to the regression informed regarding the risk and possible strategies to
model (the reduction of the criterion variance explained by reduce adolescents’ electronic media use at night. Second,
electronic media use was from 6 to 2.6 % when entering and in addition to sleep hygiene education, adolescents’
sleep difficulties to the regression model indicating that smartphones might feature specific applications supporting
more than half of the effect was mediated by sleep diffi- adolescents to maintain sleep hygiene. For instance, such
culties). Less strong was the mediating role of sleep applications might track the time when adolescents use
duration, which even disappeared when tested in a model their phone (or other electronic media), at what time there
simultaneously with sleep difficulties. The relationships is noise in their bedroom, and at what time they go to sleep
between electronic media use, sleep disturbance, and and when they get up. Such sleep hygiene applications for
depressive symptoms appeared relatively consistent across smartphones might also actively provide reminders when
the studied age range as no moderation by age was found. sleep hygiene rules are violated. Finally, adolescents suf-
All types of electronic media use in the bed before sleep fering from excessive media use and a sleep disorder
were moderately to strongly correlated with each other. should be identified by school psychologists and referred to
Moreover, they were correlated with shorter sleep and cognitive behavioral treatment of insomnia in adolescence
more symptoms of depression, which is consistent with as there is now evidence that maintaining healthy sleep
earlier reports (e.g., Cain and Gradisar 2010). Also con- patterns in adolescents might possibly reduce the incidence
sistent with earlier studies watching TV or videos, calling of depression in adolescence (Lovato and Gradisar 2014).

123
 J Youth Adolescence

Our study also has limitations. First, media use, sleep References
disturbance, and depressive symptoms were self-reported
by adolescents using questionnaires. This may involve bias Adams, S. K., & Kisler, T. S. (2013). Sleep quality as a mediator
between technology-related sleep quality, depression, and anx-
due to common method (for a discussion of this issue iety. Cyberpsychology, Behavior, and Social Networking, 16,
related to the link between subjective sleep disturbance and 25–30.
mental health problems see e.g., Lemola et al. 2013). Aiken, L. S., & West, S. G. (1991). Multiple regression: Testing and
Moreover, questionnaire measures of behavior may be interpreting interactions. Newbury Park, London: Sage.
Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator
biased by memory difficulties. While sleep could be variable distinction in social psychological research: Conceptual,
assessed objectively with actigraphy in future studies, sleep strategic, and statistical considerations. Journal of Personality
hygiene and electronic media use could be measured by and Social Psychology, 51, 1173–1182.
experience sampling methods in order to achieve higher Bastien, C. H., Vallières, A., & Morin, C. M. (2001). Validation of the
Insomnia Severity Index (ISI) as an outcome measure for
validity of measurement. Second, the correlational nature insomnia research. Sleep Medicine, 2, 297–307.
of the present study precludes causal conclusions regarding Baum, K. T., Desai, A., Field, J., Miller, L. E., Rausch, J., & Beebe,
the relationship between electronic media use, sleep dis- D. W. (2014). Sleep restriction worsens mood and emotion
turbance, and depressive symptoms. While experimental regulation in adolescents. Journal of Child Psychology and
Psychiatry, 55, 180–190.
work with adolescents has convincingly shown that sleep Brunborg, G. S., Mentzoni, R. A., Molde, H., Myrseth, H., Skouveroe,
restriction compromises emotional regulation (e.g., Baum K. J. M., Bjorvatn, B., et al. (2011). The relationship between
et al. 2014), future studies might also try to experimentally media use in the bedroom, sleep habits and symptoms of
restrict electronic media use at night to test whether ado- insomnia. Journal of Sleep Research, 20, 569–575.
Cain, N., & Gradisar, M. (2010). Electronic media use and sleep in
lescents are willing and able to reduce their electronic school-aged children and adolescents: A review. Sleep Medicine,
media use for a certain time for research purposes and 11, 735–742.
whether sleep and psychological adjustment could be Cajochen, C., Frey, S., Anders, D., Späti, J., Bues, M., Pross, A., et al.
positively influenced by such a procedure. Finally, our (2011). Evening exposure to a light-emitting diodes (LED)-
backlit computer screen affects circadian physiology and cog-
study cannot advance our knowledge which mechanisms nitive performance. Journal of Applied Physiology, 110,
might be involved in possible effects of electronic media 1432–1438.
use on sleep disturbance. Cortese, S., Brown, T. E., Corkum, P., Gruber, R., O’Brien, L. M.,
Stein, M., et al. (2013). Assessment and management of sleep
problems in youths with attention-deficit/hyperactivity disorder.
Journal of the American Academy of Child and Adolescent
Conclusion Psychiatry, 52, 784–796.
Crowley, S. J., Acebo, C., & Carskadon, M. A. (2007). Sleep,
Our findings suggest that the availability of smartphones is circadian rhythms, and delayed phase in adolescence. Sleep
Medicine, 8, 602–612.
related to increased electronic media use of adolescents at Dagys, N., McGlinchey, E. L., Talbot, L. S., Kaplan, K. A., Dahl, R.
night. Further, our findings confirm earlier reports that E., & Harvey, A. G. (2012). Double trouble? The effects of sleep
electronic media use at night is related to sleep disturbance. deprivation and chronotype on adolescent affect. Journal of
Sleep disturbance in turn appears to be a partial mediator of Child Psychology and Psychiatry, 53, 660–667.
Dollman, J., Ridley, K., Olds, T., & Lowe, E. (2007). Trends in the
the relationship between electronic media use at night and duration of school-day sleep among 10- to 15-year-old South
depressive symptoms. Adolescents might benefit from Australians between 1985 and 2004. Acta Paediatrica, 96,
education regarding sleep hygiene and the risk of electronic 1011–1014.
media use at night as improving sleep quality may be a key Dworak, M., Schierl, T., Bruns, T., & Strüder, H. K. (2007). Impact of
singular excessive computer game and television exposure on
factor in the prevention of depression. sleep patterns and memory performance of school-aged children.
Pediatrics, 120, 978–985.
Acknowledgments The study was supported by the Stiftung Eggermont, S., & Van den Bulck, J. (2006). Nodding off or switching
Suzanne und Hans Biäsch zur Förderung der Angewandten Psy- off? The use of popular media as a sleep aid in secondary-school
chologie. The authors would like to thank all the participating children. Journal of Paediatrics and Child Health, 42, 428–433.
schools, parents, and adolescents who agreed to participate in the Fossum, I. N., Nordnes, L. T., Storemark, S. S., Bjorvatn, B., &
present study. Pallesen, S. (2014). The association between use of electronic
media in bed before going to sleep and insomnia symptoms,
Author contributions SL conceived of the study, participated in its daytime sleepiness, morningness, and chronotype. Behavioral
design and coordination, performed the statistical analysis, and Sleep Medicine, 12, 343–357.
drafted the manuscript; NPG participated in the design and coordi- Gaina, A., Sekine, M., Kanayama, H., Sengoku, K., Yamagami, T., &
nation of the study, performed the measurement, and helped to draft Kagamimori, S. (2005). Short-long sleep latency and associated
the manuscript; SB participated in the design and interpretation of the factors in Japanese junior high school children. Sleep and
data; JKD participated in the design and interpretation of the data; AG Biological Rhythms, 3, 162–165.
participated in the design and interpretation of the data. All authors Gerber, M., Brand, S., Holsboer-Trachsler, E., & Pühse, U. (2010).
read and approved the final manuscript. Fitness and exercise as correlates of sleep complaints: Is it all in

123
J Youth Adolescence

our minds? Medicine and Science in Sports and Exercise, 42, Lovato, N., & Gradisar, M. (2014). A meta-analysis and model of the
893–901. relationship between sleep and depression in adolescents.
Hankin, B. L., Abramson, L. Y., Moffitt, T. E., Silva, P. A., McGee, Recommendations for future research and clinical practice.
R., & Angell, K. E. (1998). Development of depression from Sleep Medicine Reviews. doi:10.1016/j.smrv.2014.03.006.
preadolescence to young adulthood: Emerging gender differ- Lui, X., Buysse, D. J., Gentzler, A. L., Kiss, E., Mayer, L., Kepornai,
ences in a 10-year longitudinal study. Journal of Abnormal K., et al. (2007). Insomnia and hypersomnia associated with
Psychology, 107, 128–140. phenomenology and comorbidity in childhood depression. Sleep,
Harada, T., Morisane, H., & Takeuchi, H. (2002). Effect of daytime 30, 83–90.
light conditions on sleep habits and morningness–eveningness McGlinchey, E. L., Talbot, L. S., Chang, K. H., Kaplan, K. A., Dahl,
preference of Japanese students aged 12–15 years. Psychiatry R. E., & Harvey, A. G. (2011). The effect of sleep deprivation on
and Clinical Neurosciences, 56, 225–226. vocal expression of emotion in adolescents and adults. Sleep, 34,
Hautzinger, M., & Bailer, M. (1993). Allgemeine Depressions Skala. 1233.
Manual. Göttingen: Beltz Test GmbH. Medienpädagogischer Forschungsverbund Südwest (2013). JIM 2013.
Heath, M., Sutherland, C., Bartel, K., Gradisar, M., Williamson, P., Jugend, Information, (Multi-) Media. Basisstudie zum Medie-
Lovato, N., et al. (2014). Does one hour of bright or short- numgang 12- Bis 19-Jähriger in Deutschland.
wavelength filtered tablet screenlight have a meaningful effect Munezawa, T., Kaneita, Y., Osaki, Y., Kanda, H., Minowa, M., Suzuki,
on adolescents’ pre-bedtime alertness, sleep, and daytime K., et al. (2011). The association between use of mobile phones
functioning? Chronobiology International, 31, 496–505. after lights out and sleep disturbances among Japanese adoles-
Higuchi, S., Motohashi, Y., Liu, Y., & Maeda, A. (2005). Effects of cents: A nationwide cross-sectional survey. Sleep, 34, 1013–1020.
playing a computer game using a bright display on presleep National Sleep Foundation. (2006). Sleep in America poll. Washing-
physiological variables, sleep latency, slow wave sleep and REM ton, DC: National Sleep Foundation.
sleep. Journal of Sleep Research, 14, 267–273. National Sleep Foundation. (2011). Sleep in America poll. Exploring
Iglowstein, I., Jenni, O. G., Molinari, L., & Largo, R. H. (2003). Sleep connections with communications technology use and sleep.
duration from infancy to adolescence: Reference values and National Sleep Foundation: Washington, DC.
generational trends. Pediatrics, 111, 302–307. Oshima, N., Nishida, A., Shimodera, S., Tochigi, M., Ando, S.,
Ivarsson, M., Anderson, M., Åkerstedt, T., & Lindblad, F. (2009a). Yamasaki, S., et al. (2012). The suicidal feelings, self-injury, and
Playing a violent television game affects heart rate variability. mobile phone use after lights out in adolescents. Journal of
Acta Paediatrica, 98, 166–172. Pediatric Psychology, 37, 1023–1030.
Ivarsson, M., Anderson, M., Åkerstedt, T., & Lindblad, F. (2009b). Perkinson-Gloor, N., Lemola, S., & Grob, A. (2013). Sleep duration,
Playing a violent television game does not affect saliva cortisol. positive attitude toward life, and academic achievement: The
Acta Paediatrica, 98, 1052–1053. role of daytime tiredness, behavioral persistence, and school start
Ivarsson, M., Anderson, M., Åkerstedt, T., & Lindblad, F. (2013). The times. Journal of Adolescence, 36, 311–318.
effect of violent and nonviolent video games on heart rate Preacher, K. J., & Hayes, A. F. (2008). Asymptotic and resampling
variability, sleep, and emotions in adolescents with different strategies for assessing and comparing indirect effects in
violent gaming habits. Psychosomatic Medicine, 75, 390–396. multiple mediator models. Behavior Research Methods, 40,
Kessler, R. C., Avenevoli, S., & Ries Merikangas, K. (2001). Mood 879–891.
disorders in children and adolescents: An epidemiologic per- Prezza, M., Pacilli, M. G., & Dinelli, S. (2004). Loneliness and new
spective. Biological Psychiatry, 49, 1002–1014. technologies in a group of Roman adolescents. Computers in
King, D. L., Gradisar, M., Drummond, A., Lovato, N., Wessel, J., Human Behavior, 20, 691–709.
Micic, G., et al. (2013). The impact of prolonged violent video- Primack, B. A., Swanier, B., Georgiopoulos, A. M., Land, S. R., &
gaming on adolescent sleep: An experimental study. Journal of Fine, M. J. (2009). Association between media use in adoles-
Sleep Research, 22, 137–143. cence and depression in young adulthood: A longitudinal study.
Kubey, R. W. (1986). Television use in everyday life: Coping with Archives of General Psychiatry, 66, 181–188.
unstructured time. Journal of Communication, 36, 108–123. Punamäki, R. L., Wallenius, M., Nygård, C. H., Saarni, L., &
Kubiszewski, V., Fontaine, R., Rusch, E., & Hazouard, E. (2013). Rimpelä, A. (2007). Use of information and communication
Association between electronic media use and sleep habits: An technology (ICT) and perceived health in adolescence: The role
eight-day follow-up study. International Journal of Adolescence of sleeping habits and waking-time tiredness. Journal of
and Youth,. doi:10.1080/02673843.2012.751039. Adolescence, 30, 569–585.
Lemola, S., Brand, S., Vogler, N., Perkinson-Gloor, N., Allemand, Radloff, L. S. (1977). The CES-D scale a self-report depression scale
M., & Grob, A. (2011). Habitual computer game playing at night for research in the general population. Applied Psychological
is related to depressive symptoms. Personality and Individual Measurement, 1, 385–401.
Differences, 51, 117–122. Rideout, V. J., Foehr, U. G., & Roberts, D. F. (2010). Generation M2:
Lemola, S., Ledermann, T., & Friedman, E. M. (2013). Variability of Media in the lives of 8- to 18-year-olds. Menlo Park: Henry J.
sleep duration is related to subjective sleep quality and Kaiser Family Foundation.
subjective well-being: An actigraphy study. PLoS ONE, 8, Short, M. A., Gradisar, M., Lack, L. C., Wright, H. R., & Dohnt, H.
e71292. (2013). The sleep patterns and well-being of Australian adoles-
Lemola, S., Schwarz, B., & Siffert, A. (2012). Interparental conflict cents. Journal of Adolescence, 36, 103–110.
and early adolescents’ aggression: Is irregular sleep a vulnera- Smith, A. (2013). Smartphone ownership—2013 update. Washington,
bility factor? Journal of Adolescence, 35, 97–105. DC: Pew Research Center.
Lewinsohn, P. M., Hops, H., Roberts, R. E., Seeley, J. R., & Andrews, Söderqvist, F., Carlberg, M., & Hardell, L. (2008). Use of wireless
J. A. (1993). Adolescent psychopathology: I. Prevalence and telephones and self-reported health symptoms: A population-
incidence of depression and other DSM-III-R disorders in high based study among Swedish adolescents aged 15–19 years.
school students. Journal of Abnormal Psychology, 102, 133. Environmental Health, 7, 18.
Li, S., Jin, X., Wu, S., Jiang, F., Yan, C., & Shen, X. (2007). The Talbot, L. S., McGlinchey, E. L., Kaplan, K. A., Dahl, R. E., &
impact of media use on sleep patterns and sleep disorders among Harvey, A. G. (2010). Sleep deprivation in adolescents and
school-aged children in China. Sleep, 30, 361–367. adults: Changes in affect. Emotion, 10, 831–841.

123
 J Youth Adolescence

Thomée, S., Härenstam, A., & Hagberg, M. (2011). Mobile phone use include positive development in relation with sleep, pre- and perinatal
and stress, sleep disturbances, and symptoms of depression adversities, emotional support in the family and consequences of
among young adults-a prospective cohort study. BMC Public insufficient and poor sleep, particularly during adolescence.
Health, 11, 66.
Van den Bulck, J. (2003). Text messaging as a cause of sleep Nadine Perkinson-Gloor, MSc is a PhD student at the University of
interruption in adolescents, evidence from a cross-sectional Basel, Switzerland. She received her Masters degree in Psychology
study. Journal of Sleep Research, 12, 263. from the University of Basel, Switzerland in 2011. Her major research
Van den Bulck, J. (2004). Television viewing, computer game interests include positive development in relation with sleep, pre- and
playing, and internet use and self-reported time to bed and time perinatal adversities, and consequences of insufficient and poor sleep,
out of bed in secondary-school children. Sleep, 27, 101–104. particularly during adolescence.
Van den Bulck, J. (2007). Adolescent use of mobile phones for calling
and for sending text messages after lights out: Results from a Serge Brand, PhD is a research psychologist and psychotherapist at
prospective cohort study with a one-year follow-up. Sleep, 30, the Center for Affective, Stress and Sleep Disorders, Psychiatric
1220–1223. Clinics of the University of Basel, Switzerland and a research
Weaver, E., Gradisar, M., Dohnt, H., Lovato, N., & Douglas, P. assistant at the Department of Sport, Exercise and Health, Division of
(2010). The effect of presleep video-game playing on adolescent Sport Science at the University of Basel, Switzerland. He received his
sleep. Journal of Clinical Sleep Medicine, 6, 184–189. doctorate in Psychology from the University of Basel, Switzerland in
Willemse, I., Waller, G., Süss, D., Genner, S., & Huber, A.-L. (2012). 2006. His major research interests include relations of sleep and
JAMES – Jugend, Aktivitäten, Medien – Erhebung Schweiz. psychological health from infancy to adulthood and the interaction of
Zürich: Zürcher Hochschule für Angewandte Wissenschaften. well-being, sleep, psychophysiology, and physical exercise.
Wood, A. W., Loughran, S. P., & Stough, C. (2006). Does evening
exposure to mobile phone radiation affect subsequent melatonin Julia F. Dewald-Kaufmann, PhD works at the Dr. von Hauner
production? International Journal of Radiation Biology, 82, Children’s Hospital of the Ludwig Maximilian University, Munich,
69–76. Germany. She received her doctorate in psychology at the University
Yen, C. F., Ko, C. H., Yen, J. Y., & Cheng, C. P. (2008). The of Amsterdam in 2012. Her major research interests include outcomes
multidimensional correlates associated with short nocturnal sleep of chronic sleep reduction in adolescents and the role of sleep
duration and subjective insomnia among Taiwanese adolescents. hygiene.
Sleep, 31, 1515.
Yoo, S. S., Gujar, N., Hu, P., Jolesz, F. A., & Walker, M. P. (2007).
Alexander Grob, PhD is a professor for Personality and Develop-
The human emotional brain without sleep—a prefrontal amyg-
mental Psychology at the University of Basel, Switzerland. He
dala disconnect. Current Biology, 17, 877–878.
received his doctorate in psychology at the University of Berne,
Switzerland in 1990. His major research interests include develop-
Sakari Lemola, PhD is an Assistant Professor at the University of mental and intelligence diagnostics in childhood and adolescence,
Basel, Switzerland. He received his doctorate in Psychology from the early intervention for disadvantaged children and adolescents, and
University of Basel, Switzerland in 2007. His major research interests personality development in close relationships across the life-span.

123