Sleep Health 2 (2016) 211–218

Contents lists available at ScienceDirect

Sleep Health
Journal of the National Sleep Foundation

journal homepage: sleephealthjournal.org

Sleep duration and patterns in adolescents: correlates and the role

of daily stressors
S.V. Bauducco, PhD Student a,⁎, I.K. Flink, PhD a, M. Jansson-Fröjmark, PhD b, S.J. Linton, PhD a
a
Örebro University, Fakultetsgatan 1, Örebro, Sweden
b
Stockholm University, Universitetsvägen 10, Stockholm, Sweden

a r t i c l e i n f o a b s t r a c t

Article history: Objectives: The first aim of this study was to assess the prevalence of sleep deficit in a large sample of ado-
Received 17 April 2016 lescents. Second, the study aimed to assess whether short sleep duration in the sample was associated with
Received in revised form 18 May 2016 emotional and behavioral problems. Lastly, the study aimed to investigate the association between daily
Accepted 23 May 2016 stressors–bedtime activities and sleep duration.
Design: Cross-sectional survey.
Keywords:
Setting: The questionnaires were completed during school hours in 17 municipal junior high schools
National Sleep Foundation (NSF)
Sleep duration recommendations
in Sweden.
Sleep deficit Participants: A total of 2767 adolescents aged 12 to 16 years, 48% girls.
Sleep patterns Measurements and Results: Sleep measures included total sleep time (TST) for schooldays and weekends,
Emotional and behavioral problems obtained as combined measures of self-reported bed-time, wake-time, and sleep onset latency. We used
Adolescent sleep the new National Sleep Foundation's guidelines to operationalize sleep duration. Overall 12% of younger ad-
Daily stressors olescents (age 12-13 years) and 18% of older adolescents (14-16 years) slept less than recommended (TST b
Electronic media 7 hours). Adolescents reporting nonrecommended TST also reported more behavioral (ie, norm-breaking
Information and communication technology (ICT)
behaviors) and emotional problems (ie, depression, anxiety, and anger), with effects in the small-
Sleep hygiene
medium range. Finally, adolescents reporting bedtime arousal and use of information and communication
Bedtime arousal
technology in bed were more likely to report TST b 7 hours. Stress at home (for younger adolescents) and
stress of school performance (for older adolescents) were also associated with TST less than 7 hours.
Conclusions: The new National Sleep Foundation's recommendations were informative in this context. Fu-
ture sleep interventions need to target barriers to good sleep practices, such as use of information and
communication technology, stress, and worry that may contribute to arousal at bedtime.
© 2016 National Sleep Foundation. Published by Elsevier Inc. All rights reserved.

Sleep deficit is common in adolescence, and research has shown
that it can have serious consequences for wellbeing. From a develop-
mental perspective, sleep deficit is in fact normal, and most adoles-
cents will experience it at some time point as they all go through
biological and psychosocial changes that deeply affect their sleep
patterns. 1,2 Several studies involving a range of cultural groups have
shown that adolescent sleep deficit is common. These studies show
that 24% to 73% of adolescents sleep less than 7 hours per school
night, 3–6 which is 2 hours less than the generally recommended 9
hours.7 This adolescent sleep deficit appears to have increased over
the past 20 years, possibly due to the expanded availability of activi-
ties late in the evening.8,9
⁎ Corresponding author at: Fakultetsgatan 1, 70182 Örebro, Sweden. Tel.: +46
19303618; fax: +46 19303484.
E-mail address: serena.bauducco@oru.se (S.V. Bauducco).
Although activities that compete with sleep such as homework, rec-
reational activities, socializing, and use of information and communica-
tion technology (ICT) have been shown to interfere with sleep, a more
recent meta-analysis has shown that the actual number of competing
activities did not consistently predict sleep deficit.10 The authors sug-
gested that it might instead be the perceived stressfulness of these ac-
tivities rather than the objective amount of activity that negatively
impacts sleep. This explanation would be consistent with research
that has shown that adolescents who are cognitively and emotionally
aroused at bedtime due to worrying, thinking about things to do the
next day, or ruminating on what happened during the day are at higher
risk for poor sleep.10 However, only a few studies have specifically ex-
amined the association between daily stressors and sleep duration. 11
Furthermore, the impact of ICT use upon sleep duration was inconsis-
tent, suggesting that ICT's influence on sleep might depend on other
features such as the timing of the activity.10

http://dx.doi.org/10.1016/j.sleh.2016.05.006
2352-7218/© 2016 National Sleep Foundation. Published by Elsevier Inc. All rights reserved.
212 S.V. Bauducco et al. / Sleep Health 2 (2016) 211–218
Adolescents may also inadvertently be contributing to their own
sleep deficit by attempting to compensate for insufficient sleep dur-
ing the week. By delaying both bed and wake times during weekends
and holidays, adolescents may place themselves at a higher risk for
developing a delayed sleep phase 12; this, in turn, can sustain later
weekday bedtimes and exacerbate sleep deficit. 7
Despite adolescent sleep deficit being developmentally typical,
this deficit in required sleep is alarming because sufficient sleep is es-
sential for adolescents' everyday functioning. 13 There is a growing
body of evidence linking emotional and behavioral problems to
sleep deficit in adolescents.14 Sleep plays an important role in emo-
tion regulation15 and poor emotion regulation in turn can contribute
to psychological problems such as anxiety, depression, and anger.16
Longitudinal and cross-sectional studies have also shown a relation-
ship between short sleep duration and behavioral problems such as
aggression and norm-breaking behaviors. 17–20 Sleep duration has
been shown to be related to physical health problems such as obesity,
elevated blood pressure, and pain,13 and there is even a strong rela-
tionship between sleep and students' academic achievement and
school absenteeism.21,22 Given that sleep appears to impact upon a va-
riety of areas of adolescents' functioning, designing effective interven-
tions to promote sleep is a pressing issue. 23 To begin to address this
issue, improved knowledge of adolescents' sleep patterns is required
so that effective targets for preventive interventions can be identified.
To date, the development and implementation of preventive sleep
interventions for adolescents has achieved limited success. Previous
sleep interventions conducted in a universal arena such as a school
have shown that teaching sleep hygiene alone is not enough to induce
behavioral change. For example, improving adolescents' knowledge of
good sleep practices does not alter sleep behavior. 23 The addition of
motivational techniques, such as motivational interviewing, has also
proved disappointing,24 with some exceptions.25 One possible reason
for the failure to achieve actual behavioral change may be that these
studies do not address the daytime activities competing with sleep
and the arousal associated with them. The aim of this study was there-
fore to assess for this possibility.
The National Sleep Foundation (NSF) has recently updated its
sleep duration recommendations across the life span. 26 These up-
dates include recognition of excessive sleep duration as a risk, recom-
mendations for specific age groups, and a relaxing of the criteria for
insufficient sleep duration through the addition of an intermediate
level, “may be appropriate for some.” These guidelines are therefore
likely to be more informative than dichotomized measures of sleep
duration and may advance our understanding of the problem. These
new guidelines were adopted when operationalizing sleep duration
in this study.
The first aim of this study was to assess the prevalence of sleep
deficit in a sample of Swedish adolescents using these newly updated
guidelines. To our knowledge, the sleep behavior of Swedish adoles-
cence has never previously been described in the literature. Second,
this study aimed to assess whether short sleep duration in the sample
was associated with emotional and behavioral problems. Lastly, this
study aimed to assess whether there was any association between
daily stressors–bedtime activities and sleep duration.
Method
Participants and procedure
The participants were high school students in seventh and eighth
grades (age range, 12-16 years) from 17 public schools in 3 commu-
nities in middle Sweden. The target sample included a total of 3336
students. The consent procedure required active consent from stu-
dents and passive consent from parents due to the fact that passive
consent can increase participation rate and limit sampling bias. 27,28
All parents received a letter informing them about the study; overall,
122 parents declined having their child in the study. A total of 2768
eligible students (48% girls) were present at data collection (response
rate, 83%). Most of the students were born in Sweden (89%) and had
at least one biological parent born in Scandinavia (71%). The majority
lived with both biological parents (71%). We divided students into 2
age groups according to the NSF guidelines; 43% of our sample was
12 to 13 years old (school-aged children) and 57% was 14 to
16 years old (adolescents). This study was approved by the Regional
Ethical Board in Uppsala, Sweden.
Students filled out the questionnaires in their classrooms, and
teachers were asked to leave in order to ensure confidentiality.
Trained test leaders informed the students about confidentiality and
that participation was voluntary, helped the students if necessary
(eg, explaining difficult questions), and collected the completed
questionnaires. Students had 180 minutes to complete the question-
naire and received a snack during data collection. In addition, each
class received 300 Swedish crowns (35 US dollars) as a thank you
for participating.
Measures
The questionnaire included both established instruments and
questions developed specifically for this study that assessed
sociodemographics, sleep, stress, use of electronic media, emotional
and behavioral problems, and other variables not included as part
of this study.
Sociodemographics
Items included age, sex, country of birth, family situation, and par-
ents' country of birth.
Sleep duration
Weekday sleep duration was estimated by calculating the interval
between students' self-reported bed time (“What time do you usually
go to bed on school days?”) and wake time (“What time do you usu-
ally wake up on school days?”), subtracting sleep onset latency (“On
school days, after you got to bed at night, about how long does it usu-
ally take for you to fall asleep?”). Weekend sleep duration was calcu-
lated in an identical way. These items were taken from the School
Sleep Habits Survey. 29 Weekday sleep duration was then trans-
formed into a categorical variable according to the NSF guidelines
for “recommended,” “may be appropriate,” and “not recommended”
sleep duration (in this study the terms “optimal,” “borderline,” and
“poor” will be used). Namely, for school-aged children (6-13 years),
recommended sleep duration is 9 to 11, 7 to 8, or 12 hours may be
appropriate, whereas less than 7 or more than 12 hours is not recom-
mended; for adolescents (13-17 years), a sleep duration of 8 to
10 hours is recommended, 7 or 11 hours may be appropriate, and
less than 7 or more than 11 hours is not recommended.26
Stressors and bedtime activities/arousal
Daily stressors. Stress was measured using a short version (27 items)
of the Adolescent Stress Questionnaire.30 In this study, we used 4 sub-
scales “stress of home life” (4 items, eg, “arguments at home), “stress
of school performance” (3 items, eg, “keeping up with school work”),
“stress of school/leisure conflict” (3 items, eg, “not having enough
time for activities outside school”), and “stress of peer pressure” (4
items, eg, “pressure to fit in with peers”). 31 Responses were on a
5-point scale (0-4) from “not stressed at all” to “extremely stressed.”
Reliability was Cronbach α = .84 for “stress of home life,” α = .79
for “school performance,” α = .84 for “school/leisure conflict,” and
α = .84 for “stress of peer pressure.”
 S.V. Bauducco et al. / Sleep Health 2 (2016) 211–218 213

Cognitive-emotional arousal. Cognitive-emotional arousal was mea- Norm-breaking behavior

sured through one subdimension of the Adolescents Sleep Hygiene
Scale 32: the cognitive-emotional factor, which includes 6 items
(eg, “I go to bed and think about things I need to do,” “During the
1 hour before bedtime, things happen that make me feel strong emo-
tions“). Students reported how often sleep-related thoughts and
emotions had occurred in the past months on a 6-point scale (0-5),
with higher scores indicating lower arousal. The cognitive-
emotional subscale showed good reliability, Cronbach α = .83.
Electronic media in bed. One question assessed whether students used
electronic media “after lights out,” including TV, computer, tablet, or mo-
bile. Responses were on a 4-point scale from “never” to “almost always.”
Depression
Depressive symptoms during the last week were measured
through the Centre for Epidemiological Studies Depression Scale, 20
items.42 Higher scores indicate greater depressed mood. For consis-
tency with the rest of the questionnaire, we used a response scale
ranging from 0 “never” to 4 “very often.” The scale showed good reli-
ability, Cronbach α = .95.
Anxiety
General anxiety during the last week was assessed through the
Overall Anxiety Severity and Impairment Scale (OASIS).33 The OASIS
has 5 items with responses ranging from 0 “none” to 4 “all the time.”
In this study, the OASIS showed good reliability, Cronbach α = .87.
We asked students 13 questions about norm-breaking behaviors
during the last year, including stealing, violence, contact with police
and social services, and use of illegal drugs. 34 Responses ranged be-
tween “no, it has not happened” (0) to “more than 10 times” (4).
The reliability for norm-breaking behavior was Cronbach α = .82.
Data analysis
Because only a small portion (n = 18) of adolescents reported
sleeping more than recommended, it was not possible to look at exces-
sive sleep duration separately. Therefore, we removed these individ-
uals from the analyses. Independent-samples t tests were used to test
for age and sex differences, whereas paired t tests were used to test
for differences between week and weekend sleep patterns. Multivari-
ate analysis of variance (MANOVA) was used to test for differences in
emotional and behavioral outcomes (ie, depression, anxiety, delin-
quency, and anger) at the different levels of total sleep time (TST; ie,
optimal, borderline, and poor). Finally, multinomial logistic regression
was used to determine the role of ICT after lights out, cognitive-
emotional sleep hygiene, and stress for borderline and poor TST as
compared with optimal TST. The TST referred to in the MANOVA and
regression analyses is weekday TST. Multivariate analysis of variance
and multinomial logistic regression analysis were performed separate-
ly for students aged 12 to 13 years and students aged 14 to 16 years,
based on the NSF recommendations,26 and controlling for sex because
of potential sex differences in sleep and its correlates.35,36

Anger
Feelings of anger during the past week were assessed through an
adapted version of the OASIS (5 items), where “anxiety” was replaced by
“anger,” with the exception of one item: “How often do you avoid situa-
tions, places, objects, or activities because of anxiety or fear?” which was
substituted with “How often have you had an outburst (eg, hit, fought,
shouted) during the last week?” Responses for the whole scale were as fol-
lows: “not at all,” “sometimes,” “from time to time,” “often,” “all the time.”
This scale showed good reliability, Cronbach α = .85.
Results
Sample characteristics
The sample included 2768 adolescents, 48% of were girls and
43.3% were 12 to 13 years old. For the distribution of depressive
symptoms, anxiety, anger, norm-breaking behaviors, sleep hygiene,
ICT in bed, and daily stressors among boys/girls and adolescents
aged 12-13/14-16 years (see Table 1).

Table 1
Descriptives of sleep variables, emotional and behavioral problems, and stressors-bedtime activities in the sample.

Main variables All (n = 2466) Boys (n = 1293) Girls (n = 1173) Age 12-13 y (n = 1057) Age 14-16 y (n = 1408)

Sleep variables (h/min; SD)

TSTschool-week 7:56 (1:16) 8:10 (1:12) 7:41 (1:19) 8:06 (1:13) 7:49 (1:18)
Bedtimeschool-week 22:19 (1:01) 22:16 (0:56) 22:21 (1:06) 22:07 (0:56) 22:28 (1:03)
SOLschool-week 0:30 (0:29) 0:27 (0:26) 0:33 (0:31) 0:29 (0:28) 0:31 (0:30)
Wake timeschool-week 6:45 (0:40) 6:54 (0:40) 6:35 (0:39) 6:43 (0:40) 6:47 (0:41)
TSTweekend 9:32 (1:31) 9:29 (1:35) 9:35 (1:26) 9:37 (1:30) 9:28 (1:32)
Bedtimeweekend 00:34 (1:41) 00:42 (1:45) 00:24 (1:35) 00:21 (1:42) 00:43 (1:39)
SOLweekend 0:21 (0:26) 0:19 (0:24) 0:24 (0:28) 0:20 (0:23) 0:22 (0:28)
Wake timeweekend 10:27 (1:34) 10:31 (1:38) 10:24 (1:29) 10:19 (1:30) 10:34 (1:36)
Emotional/behavioral problems, mean (SD)
Depressive symptoms (0-80) 16.8 (14.7) 11.8 (10.8) 22.4 (16.5) 15.3 (13.8) 18.0 (15.3)
Norm-breaking (0-52) 1.1 (2.9) 1.3 (3.2) 0.9 (2.6) 0.9 (2.5) 1.27 (3.2)
Anxiety (0-20) 3.0 (3.5) 2.0 (2.8) 4.2 (3.9) 2.7 (3.2) 3.2 (3.7)
Anger (0-20) 3.4 (3.4) 2.8 (3.1) 4.1 (3.7) 3.2 (3.2) 3.6 (3.6)
Sleep hygiene (0-30), mean (SD) 21.7 (6.2) 23.8 (5.5) 20.1 (6.5) 22.2 (5.9) 21.3 (6.4)
ICT in bed (%)
Never 22.0 25.5 18.4 22.3 22.0
Sometimes 29.2 31.7 26.6 31.8 27.3
Often 16.0 15.9 15.8 15.5 16.2
Almost always 32.8 26.9 39.2 30.4 34.5
Daily stressors, mean (SD)
School (0-12) 4.7 (3.3) 3.7 (2.9) 5.8 (3.2) 4.4 (3.1) 5.0 (3.4)
Home (0-16) 3.0 (3.2) 2.2 (2.7) 3.1 (3.4) 2.8 (3.0) 3.1 (3.3)
Peers (0-16) 1.4 (2.4) 1. 1 (2.0) 1.8 (2.6) 1.4 (2.3) 1.4 (2.4)
School/leisure conflict (0-12) 3.5 (3.2) 2.8 (3.0) 4.2 (3.2) 3.2 (3.0) 3.7 (3.3)

TST, total sleep time; SOL, sleep onset latency; ICT, information and communication technology.
214 S.V. Bauducco et al. / Sleep Health 2 (2016) 211–218

Sleep patterns
Age and sex differences in weekdays sleep patterns
Of the younger adolescents (age 12-13 years), 58.1% reported opti-
mal TST, 29.6% reported borderline TST, and 12.3% reported sleeping
more or less than recommended (N12 hours [0.2%] and b7 hours
[12.1%], respectively). In the older age group (age 14-16 years), 48.2%
of students reported optimal sleep duration, 33.0% reported borderline
TST, and 18.8% reported a TST more or less than recommended
(N11 hours [0.8%] and b 7 hours [18.1%], respectively).
Girls and older students generally reported shorter sleep duration.
More specifically, girls reported on average shorter sleep duration than
did boys, both in the younger and older groups (mean differences, 26 mi-
nutes [t1151 = −6.28, P b .001, Cohen d = 0.37] and 30 minutes [t1402.7 =
−7.67, P b .001, Cohen d = 0.41], respectively), and when looking into
bed and wake times specifically, it emerged that although bedtimes
were similar, girls woke up earlier than boys both in the younger and
older groups (mean differences, 14 minutes [t1170 = −6.03, P b .001,
Cohen d = 0.35] 21 minutes [t1395.7 = −10.70, P b .001, Cohen d =
0.57], respectively).
Weekday-weekend differences
Sleep patterns during the week and during the weekend were sig-
nificantly different, with later bed and wake times for both the
younger and older groups (younger adolescents' bedtime mean dif-
ference, 2 hours 13 minutes [ t1169 = − 55.45, P b .001, Cohen d =
1.6], and younger adolescents' wake time mean difference, 3 hours
36 minutes [t1142 = −77.16, P b .001, Cohen d = 2.3]; older adoles-
cents' bedtime mean difference, 2 hours 15 minutes [t1522 =
−61.41, P b .001, Cohen d = 1.6], and older adolescents' wake time
mean difference, 3 hours 47 minutes [t1504 = − 88.79, P b .001,
Cohen d = 2.3], respectively). Both groups reported longer average
sleep duration during weekends (younger adolescents mean differ-
ences, 1 hour 30 minutes [t1102 = −30.63, P b .001, Cohen d = .92];
older adolescents mean difference, 1 hour 39 minutes [t1460 =
−38.79, P b .001, Cohen d = 1.02]).
Differences in emotional and behavioral problems by sleep duration
We performed MANOVA analyses to look at whether emotional
and behavioral outcomes including depression, anxiety, norm-
breaking behaviors, and anger differed at different levels of TST
(optimal, borderline, and poor), controlling for sex.
Age 12-13 years (n = 1113)
There was a significant difference in the emotional and behavioral
outcomes (depression, anxiety, delinquency, and anger) based on
TST (V = 0.12, F8,2210 = 17.4, P b .001, ηp2 = 0.06). Separate univariate

Fig. 1. Emotional and behavioral problems for younger adolescents (age 12-13 years, n = 1115) reporting optimal, borderline, and poor sleep duration. Score range for anger (0-20),
anxiety (0-20), depression (0-80), and norm-breaking behaviors (0-52). TST, total sleep time.
 S.V. Bauducco et al. / Sleep Health 2 (2016) 211–218 215

analyses of variance showed that scores for depression (F2 = 44.5, P b
.001, ηp2 = 0.07), anxiety (F2 = 17.7, P b .001, ηp2 = 0.03), and anger
(F2 = 29.3, P b .001, ηp2 = 0.05) were significantly higher for border-
line TST compared with optimal and for poor TST compared with border-
line, and scores for norm-breaking behaviors were significantly higher
for poor TST compared with borderline (F2 = 35.2, P b .001, ηp2 =
0.06; see Fig. 1).
Finally, there was a significant interaction of TST by sex (V = 0.03,
F8,2214 = 4.21, P b .001, ηp2 = 0.015) on the total score of norm-
breaking behaviors and depression. Simple effects analysis showed
that norm-breaking behaviors did not differ for boys and girls
sleeping in the optimal range, but they increased more for boys
than for girls for borderline and poor sleep duration (F2 = 10.5, P b
.001, ηp2 = 0.02). Girls' scores of depression were higher than boys'
at each level of TST (F2 = 4.1, P = .02, ηp2 b 0.01).
Age 14-16 years (n = 1426)
There was a significant difference in the emotional and behavioral
outcomes based on TST (V = 0.12, F8,2836 = 22, P b .001, ηp2 = 0.06).
Separate univariate analyses of variance showed that scores for de-
pression (F2 = 65.8, P b .001, ηp2 = 0.08), anxiety (F2 = 27.8, P b
.001, ηp2 = 0.04), anger (F2 = 36.6, P b .001, ηp2 = 0.06), and norm-
breaking behaviors (F2 = 36.3, P b .001, ηp2 = 0.05) were significantly
higher for borderline TST compared with optimal, and for poor TST
compared with borderline. Results were similar to those for younger
adolescents and, therefore, no figure is presented.
Similar to the younger group, there was a significant interaction of
TST by sex (V = 0.02, F8,2894 = 3.78, P = .001, ηp2 = 0.010), so that
girls' scores for depression (F2 = 6.8, P b .001, ηp2 b 0.01), anxiety
(F2 = 6.5, P = .002, ηp2 b 0.01), and anger (F2 = 3.2, P = .04, ηp2 b
0.01) were higher than boys' at each level of TST. Norm-breaking be-
haviors did not differ for boys and girls sleeping in the optimal range,
but they increased more for boys than for girls for borderline and
poor sleep duration (F2 = 3.2, P = .04, ηp2 b 0.01; Fig. 2).
Stressors, ICT in bed, and sleep hygiene and sleep duration
We performed separate analyses for the younger and older group
testing for associations between TST (optimal, borderline, and poor)
and common stressors and behaviors including stress of school,
school/leisure conflict, home and peer pressure, use of electronic
media after lights out, and cognitive-emotional sleep hygiene (Table 2).
Independent of age, adolescents who used ICT in bed more often
were also more likely to report borderline (odds ratio [OR] = 1.37-
1.30) and poor sleep duration (OR = 1.43-1.61). Similarly, adoles-
cents who reported better sleep hygiene at bedtime were more likely
to report optimal sleep duration as compared with poor (OR = 0.64-
0.58) or borderline TST (OR = 0.80-0.75).

Fig. 2. Interaction between sex and emotional and behavioral problems for older adolescents (age 14-16 years, n = 1426) reporting optimal, borderline, and poor sleep duration.
Score range for depression (0-80), anxiety (0-20), anger (0-20), and norm-breaking behaviors (0-52).TST, total sleep time.
216 S.V. Bauducco et al. / Sleep Health 2 (2016) 211–218

Table 2
Multinomial logistic regression exploring how stressors, ICT in bed, and sleep hygiene relate to borderline and poor TST for the 2 age groups.

Variable Age 12-13 y (n = 1069) Age 14-16 y (n = 1386)

Odds ratio (optimal is the referent) 95% CI Odds ratio (optimal is the referent) 95% CI

Stress—school performance
Borderline 1.02 (0.85-1.24) 1.12 (0.95-1.31)
Poor 1.16 (0.90-1.51) 1.26⁎ (1.04-1.53)
Stress—school/leisure
Borderline 1.05 (0.88-1.26) 0.99 (0.85-1.16)
Poor 1.03 (0.81-1.32) 1.10 (0.91-1.32)
Stress—home
Borderline 1.12 (0.90-1.39) 1.14 (0.95-1.38)
Poor 1.40⁎ (1.06-1.84) 1.20 (0.96-1.50)
Stress—peers
Borderline 1.60⁎ (1.07-2.38) 0.88 (0.69-1.12)
Poor 1.04 (0.57-1.90) 0.84 (0.64-1.10)
Sleep hygiene
Borderline 0.80⁎ (0.68-0.95) 0.75⁎⁎⁎ (0.65-0.87)
Poor 0.64⁎⁎⁎ (0.52-0.80) 0.58⁎⁎ (0.49-0.69)
ICT in bed
Borderline 1.37⁎⁎⁎ (1.21-1.55) 1.30⁎⁎ (1.17-1.45)
Poor 1.43⁎⁎⁎ (1.19-1.71) 1.61⁎⁎ (1.40-1.85)
Sex (girl)
Borderline 5.35⁎⁎⁎ (2.53-11.34) 1.51⁎⁎ (1.16-1.96)
Poor 1.52 (0.53-4.41) 1.33 (0.96-1.86)
Sex (girl) × stress—peers
Borderline 0.47⁎⁎ (0.28-0.78) – –
Poor 1.13 (0.58-2.23) – –

Model fit (age 12-13 y): R2 = 0.149 (Cox and Snell), 0.176 (Nagelkerke); model fit (age 14-16 y): R2 = 0.157 (Cox and Snell), 0.181 (Nagelkerke).
ICT, information and communication technology; TST, total sleep time.
⁎ P b .05.
⁎⁎ P b .01.
⁎⁎⁎ P b .001.

Concerning the relationship between stress and sleep, slightly dif-
ferent patterns emerged for the younger and the older groups. Youn-
ger adolescents experiencing stress in the home environment were
more likely to report poor sleep duration (OR = 1.40), whereas
older adolescents experiencing stress of school performance were
more likely to report poor TST (OR = 1.26).
Discussion
In this study, we aimed to assess sleep duration and patterns in a
Swedish sample of adolescents. We divided our sample into younger
(age 12-13 years) and older (age 14-16 years) adolescents according
to the NSF's sleep recommendations 26 and found that 12% of the
younger and 18% of the older adolescents reported sleeping less
than 7 hours per night. The prevalence of poor sleep duration was
lower than reported by previous studies, which ranges between
24% and 73%3,4; however, this could be due to the relatively younger
age of the participants in our sample (mean = 13.6). We also found
that very few reported excessive sleep duration, 0.2% of the younger
adolescents and 0.8% of the older, respectively. We also aimed to
assess for the potential relationship between sleep duration and
adolescents' behavioral and emotional problems. We found that
adolescents reporting poor and borderline TST also reported more
problems, with effects in the small-medium range. This finding is
consistent with a growing body of research showing that short
sleep duration is common 7 and intertwined with emotional and
behavioral problems.37
A striking finding was the large difference between weekday and
weekend sleep duration. On average, adolescents delayed their sleep
by 2 hours and woke up almost 4 hours later than during school days,
thus gaining an average of 1.5 hours extra sleep in the weekends. It
may be that those adolescents who are sleep deprived try to compen-
sate for their sleep during the weekends and then struggle to readjust
bed and wake times during the school week. This pattern appears to
be common among adolescents 7 and may be a warning sign for an
irregular circadian rhythm, which, in turn, has negative conse-
quences for sleep quantity, quality, and daytime functioning. 12,38
This study aimed to investigate the potential association between
daily stressors–bedtime activities and sleep duration. We found a
dose-response relationship between sleep duration and cognitive-
emotional arousal and ICT. However, the ORs were small. Higher cog-
nitive and emotional arousal around bedtime was consistently asso-
ciated with poor and borderline TST, which is in line with a recent
meta-analysis.10 Thus, giving adolescents tools to cope with emotion-
al arousal and worry at bedtime might be an effective intervention.
Similarly, using electronic media after lights out was related to
poor TST and is worrisome given that (49%) reported doing this
“often” or “almost always.” Electronic media in bed may impact
sleep duration simply because time spent on ICT would otherwise
be spent sleeping, because ICT-related activities may provoke arousal
and disrupt sleep, or because the bright light from these devices sup-
presses melatonin and delays sleep initiation. 39 Consequently, the
use of ICT at bedtime should be limited if not completely avoided
and future interventions need to address the difficulties adolescents
might have in limiting its usage. Interventions could aim to help ado-
lescents to define new norms about their own access to devices and
their availability to others after bedtime.
Different stressors appear to have an impact on sleep duration in
younger and older adolescents. In particular, stress of school perfor-
mance was related to poor TST for older adolescents, whereas stress
of home arguments was related to poor TST in younger adolescents.
Therefore, sleep interventions might consider including stress man-
agement exercises and problem solving in order to create opportuni-
ties for adolescents to apply good sleep routines at bedtime.
This study has a number of limitations. The cross-sectional design
precludes causal interpretations. Although we suspect that sleep ex-
acerbates emotional and behavioral problems, there is a possibility
that it is the other way around, that emotional and behavioral prob-
lems influence sleep problems. Another possibility is that both are
true and that the relationship is bidirectional. 37 Similarly, adolescents
 S.V. Bauducco et al. / Sleep Health 2 (2016) 211–218
who have difficulties falling asleep might use ICT at bedtime as sleep
aid and get caught in a negative cycle. 40,41 Regardless, longitudinal or
experimental studies are needed to clarify the direction of this rela-
tionship. The cross-sectional design of this study was advantageous,
in that it offered an effective way to get timely information from a
large population. The design allowed us to identify some potentially
important factors for adolescents' sleep that are worth further inves-
tigation. Although we did not perform a nonrespondent analysis, par-
ticipation rate was high (83%). All public schools in the county were
included, which likely included participants from a wide range of so-
cioeconomic backgrounds. Another potential limitation was that all
variables used in this study were self-reported, which might result
in response bias. However, a strength of our sleep duration measure
was that it was a combination of multiple questions (ie, bedtime,
SOL, and wake time), which might be advantageous over a single
item. Another advantage of this study was the large sample size,
which allowed us to test the new NSF's guidelines in 2 age groups,
thus giving a broader insight on their usability.
Conclusions
Our findings suggest that Swedish adolescents report similar
sleep patterns to adolescents from other cultural backgrounds and a
similar association between sleep deficits and emotional and behav-
ioral disturbance. The new NSF recommendations appear useful in
this context, and using these guidelines consistently would facilitate
comparison between studies and improve our knowledge about
sleep in adolescents. Our findings also suggest that there may be an
association between daily stressors and sleep duration. Future sleep
interventions need to take into account barriers to good sleep prac-
tices, such as use of electronic media at bedtime, stress, and, in gener-
al, arousal at bedtime in order to increase sleep duration. Moreover,
they should help adolescents reduce the drastic shift between week-
days' and weekends' sleep patterns, as irregular circadian rhythm can
be a risk factor for further sleep deficits and daytime impairment.
However, longitudinal and experimental research is needed to clarify
further the mechanisms behind these associations.
Disclosures
None of the authors of this paper has a financial or personal rela-
tionship with other people or organizations that could inappropriate-
ly influence or bias the content of the paper.
Acknowledgments
This research was supported by Forskningsrådet Formas,
Forskningsrådet för Arbetsliv och Socialvetenskap (FAS), Vetenskapsrådet,
and Vinnova.
We thank Dr John Barnes for providing help in language editing.
References
1. Dahl RE, Lewin DS. Pathways to adolescent health sleep regulation and behavior.
J Adolesc Health. 2002;31(6):175–184.
2. Colrain IM, Baker FC. Changes in sleep as a function of adolescent development.
Neuropsychol Rev. 2011;21(1):5–21.
3. Hysing M, Pallesen S, Stormark KM, Lundervold AJ, Sivertsen B. Sleep patterns and
insomnia among adolescents: a population‐based study. J Sleep Res. 2013;22(5):
549–556.
4. Meldrum RC, Restivo E. The behavioral and health consequences of sleep depriva-
tion among US high school students: relative deprivation matters. Prev Med. 2014;
63:24–28.
5. Do YK, Shin E, Bautista MA, Foo K. The associations between self-reported sleep
duration and adolescent health outcomes: what is the role of time spent on Inter-
net use? Sleep Med. 2013;14(2):195–200.
217
6. Garaulet M, Ortega F, Ruiz J, et al. Short sleep duration is associated with increased
obesity markers in European adolescents: effect of physical activity and dietary
habits. The HELENA study. Int J Obes. 2011;35(10):1308–1317.
7. Gradisar M, Gardner G, Dohnt H. Recent worldwide sleep patterns and problems
during adolescence: a review and meta-analysis of age, region, and sleep. Sleep
Med. 2011;12(2):110–118.
8. Pallesen S, Hetland J, Sivertsen B, Samdal O, Torsheim T, Nordhus IH. Time trends
in sleep-onset difficulties among Norwegian adolescents: 1983-2005. Scand J Pub-
lic Health. 2008;36(8):889–895.
9. Keyes KM, Maslowsky J, Hamilton A, Schulenberg J. The great sleep recession:
changes in sleep duration among US adolescents, 1991-2012. Pediatrics. 2015;
135(3):460–468.
10. Bartel KA, Gradisar M, Williamson P. Protective and risk factors for adolescent
sleep: a meta-analytic review. Sleep medicine reviews. 2015;21:72–85.
11. Chung KF, Cheung MM. Sleep-wake patterns and sleep disturbance among Hong
Kong Chinese adolescents. Sleep. 2008;31(2):185–194.
12. Crowley SJ, Acebo C, Carskadon MA. Sleep, circadian rhythms, and delayed phase
in adolescence. Sleep Med. 2007;8(6):602–612.
13. Shochat T, Cohen-Zion M, Tzischinsky O. Functional consequences of inadequate
sleep in adolescents: a systematic review. Sleep Med Rev. 2014;18(1):75–87.
14. Owens J, Au R, Carskadon M, et al. Insufficient sleep in adolescents and young adults:
an update on causes and consequences. Pediatrics. 2014;134(3):e921–e932.
15. Yoo S-S, Gujar N, Hu P, Jolesz FA, Walker MP. The human emotional brain without
sleep—a prefrontal amygdala disconnect. Curr Biol. 2007;17(20):R877–R878.
16. Jenni OG, Dahl RE. Sleep, cognition, and emotion: a developmental view; 2008.
17. Clinkinbeard SS, Simi P, Evans MK, Anderson AL. Sleep and delinquency: does the
amount of sleep matter? J Youth Adolesc. 2011;40(7):916–930.
18. McGlinchey EL, Harvey AG. Risk behaviors and negative health outcomes for ado-
lescents with late bedtimes. J Youth Adolesc. 2015;44(2):478–488.
19. Peach HD, Gaultney JF. Sleep, impulse control, and sensation-seeking predict de-
linquent behavior in adolescents, emerging adults, and adults. J Adolesc Health.
2013;53(2):293–299.
20. McKnight-Eily LR, Eaton DK, Lowry R, Croft JB, Presley-Cantrell L, Perry GS. Rela-
tionships between hours of sleep and health-risk behaviors in US adolescent stu-
dents. Prev Med. 2011;53(4):271–273.
21. Bauducco S, Tillfors M, Özdemir M, Flink I, Linton S. Too tired for school? The effects of
insomnia on absenteeism in adolescence. Sleep Health. 2015;1(3):205–210.
22. Hysing M, Harvey AG, Linton SJ, Askeland KG, Sivertsen B. Sleep and academic perfor-
mance in later adolescence: results from a large population‐based study. J Sleep Res.
2016;25:318–324.
23. Cassoff J, Knäuper B, Michaelsen S, Gruber R. School-based sleep promotion pro-
grams: effectiveness, feasibility and insights for future research. Sleep Med Rev.
2013;17(3):207–214.
24. Cain N, Gradisar M, Moseley L. A motivational school-based intervention for ado-
lescent sleep problems. Sleep Med. 2011;12(3):246–251.
25. Bonnar D, Gradisar M, Moseley L, Coughlin A-M, Cain N, Short MA. Evaluation of
novel school-based interventions for adolescent sleep problems: does parental in-
volvement and bright light improve outcomes? Sleep Health. 2015;1(1):66–74.
26. Hirshkowitz M, Whiton K, Albert SM, et al. National Sleep Foundation's sleep time
duration recommendations: methodology and results summary. Sleep Health.
2015;1(1):40–43.
27. Pokorny SB, Jason LA, Schoeny ME, Townsend SM, Curie CJ. Do participation rates
change when active consent procedures replace passive consent. Eval Rev. 2001;
25(5):567–580.
28. Shaw T, Cross D, Thomas LT, Zubrick SR. Bias in student survey findings from ac-
tive parental consent procedures. Br Educ Res J. 2015;41(2):229–243.
29. Wolfson AR, Carskadon MA. Sleep schedules and daytime functioning in adoles-
cents. Child Dev. 1998:875–887.
30. Byrne D, Davenport S, Mazanov J. Profiles of adolescent stress: the development of
the Adolescent Stress Questionnaire (ASQ). J Adolesc. 2007;30(3):393–416.
31. Moksnes UK, Espnes GA. Evaluation of the Norwegian version of the Adolescent
Stress Questionnaire (ASQ‐N): factorial validity across samples. Scand J Psychol.
2011;52(6):601–608.
32. Storfer‐Isser A, Lebourgeois MK, Harsh J, Tompsett CJ, Redline S. Psychometric
properties of the adolescent sleep hygiene scale. J Sleep Res. 2013;22(6):707–716.
33. Norman SB, Hami Cissell S, Means‐Christensen AJ, Stein MB. Development and
validation of an Overall Anxiety Severity and Impairment Scale (OASIS). Depress
Anxiety. 2006;23(4):245–249.
34. Kerr M, Stattin H. What parents know, how they know it, and several forms of ad-
olescent adjustment: further support for a reinterpretation of monitoring. Dev
Psychol. 2000;36(3):366–380.
35. Zahn-Waxler C, Shirtcliff EA, Marceau K. Disorders of childhood and adolescence:
gender and psychopathology. Annu Rev Clin Psychol. 2008;4:275–303.
36. Johnson EO, Roth T, Schultz L, Breslau N. Epidemiology of DSM-IV insomnia in ad-
olescence: lifetime prevalence, chronicity, and an emergent gender difference. Pe-
diatrics. 2006;117(2):e247–e256.
37. Kahn M, Sheppes G, Sadeh A. Sleep and emotions: bidirectional links and underly-
ing mechanisms. Int J Psychophysiol. 2013;89(2):218–228.
38. Saxvig IW, Pallesen S, Wilhelmsen-Langeland A, Molde H, Bjorvatn B. Prevalence and cor-
relates of delayed sleep phase in high school students. Sleep Med. 2012;13(2):193–199.
39. Cain N, Gradisar M. Electronic media use and sleep in school-aged children and
adolescents: a review. Sleep Med. 2010;11(8):735–742.
40. Tavernier R, Willoughby T. Sleep problems: predictor or outcome of media use
among emerging adults at university? J Sleep Res. 2014;23(4):389–396.
218 S.V. Bauducco et al. / Sleep Health 2 (2016) 211–218

41. Eggermont S, Van den Bulck J. Nodding off or switching off? The use of popular 42. Radloff LS. The CES-D Scale: A self-report depression scale for research in
media as a sleep aid in secondary‐school children. J Paediatr Child Health. 2006; the general population. Applied Psychological Measurement. 1977;1(3):
42(7–8):428–433. 385–401.