Content uploaded by Markus Gerber
Author content
All content in this area was uploaded by Markus Gerber on Jul 10, 2024
Content may be subject to copyright.
Original article
Daily Morning Running for 3 Weeks Improved Sleep and Psychological
Functioning in Healthy Adolescents Compared With Controls
Nadeem Kalak, M.Sc.
a
, Markus Gerber, Ph.D.
b
, Roumen Kirov, M.D., Ph.D.
c
, Thorsten Mikoteit, M.D.
a
,
Juliana Yordanova, M.D., Ph.D.
c
, Uwe Pu¨ hse, Ph.D.
b
, Edith Holsboer-Trachsler, M.D.
a
, and
Serge Brand, Ph.D.
a,
*
a
Center for Affective, Stress and Sleep Disorders, Psychiatric Hospital of the University of Basel, Basel, Switzerland
b
Institute of Exercise and Health Sciences, University of Basel, Basel, Switzerland
c
Institute of Neurobiology, Bulgarian Academy of Sciences, Sofia, Bulgaria
Article history: Received September 20, 2011; Accepted February 28, 2012
Keywords: Intervention study; Objective and subjective sleep improvements; Psychological functioning; Exercising; Adolescents
ABSTRACT
Purpose: To compare sleep electroencephalographic patterns and psychological functioning of
healthy adolescents running regularly in the mornings with those of control subjects. Although
several studies have shown that regular moderate-to-vigorous exercise is related to favorable sleep
and psychological functioning in adolescents, research on the effectiveness of short interventions is
more limited.
Methods: Fifty-one adolescents (mean age ⫽18.30 years; 27 female [53%]) took part in the study;
they were randomly assigned either to a running or to a control group. The running group went
running every morning for 30 minutes at moderate intensity during weekdays for 3 consecutive
weeks. Sleep electroencephalographic patterns and psychological functioning were assessed in
both groups before and after the 3-week period. All participants also kept a sleep log for 3 weeks.
Results: Objective sleep improved (slow-wave sleep increased; sleep onset latency decreased) in
the running group compared with the control group. Subjective sleep quality, mood, and concen-
tration during the day improved, whereas sleepiness during the day decreased.
Conclusions: Thirty minutes of running in the morning during weekdays for 3 consecutive weeks
impacted positively on sleep and psychological functioning in healthy adolescents compared with
control subjects. Running is inexpensive and easy to implement during school schedules, and as
both objective and subjective improvements were observed within 3 weeks, regular physical
exercise should be promoted.
䉷2012 Society for Adolescent Health and Medicine. All rights reserved.
IMPLICATIONS AND
CONTRIBUTION
Moderate running in the
morning for 3 consecutive
weeks impacted positively
on objective and subjective
sleep and psychological
functioning among healthy
adolescents. Regular exer-
cise, such as running should
be promoted as both rem-
edy and preventative mea-
sure for poor sleep and poor
psychological functioning.
Adolescence is a critical period for both neural and psycho-
logical [1] development, in which sleep plays an important func-
tional role [2,3]. Owing to a variety of different factors, such as
physical maturation (e.g., dramatic increase in secretion of
growth hormones), psychological factors (e.g., identity forma-
tion), social factors (relative dependence from parents and
peers), and extracurricular factors (e.g., sports and leisure activ-
ities, academic achievements), total sleep time (TST) decreases
[2–5], although adolescents seem to still require 9 or more hours
of sleep per night [4,5]. Daytime sleepiness is common among
adolescents, and acute [6] and chronic [7] sleep disturbances
have been related to poor physical and psychological function-
ing. In a cross-sectional survey [8],⬎60% of respondents were
categorized as poor-quality sleepers, and shorter sleep duration,
poor sleep quality, and sleepiness were independently associated
with poor school achievement in children and adolescents [9].
* Address correspondence to: Serge Brand, Ph.D., Center for Affective, Stress
and Sleep Disorders, Psychiatric Hospital of the University of Basel, Wilhelm
Klein-Strasse 27, 4012 Basel, Switzerland.
E-mail address: serge.brand@upkbs.ch (S. Brand).
N.K. and M.G. contributed equally to this work.
Journal of Adolescent Health 51 (2012) 615–622
www.jahonline.org
1054-139X/$ - see front matter 䉷2012 Society for Adolescent Health and Medicine. All rights reserved.
http://dx.doi.org/10.1016/j.jadohealth.2012.02.020
Exercise appears to be a simple and inexpensive method for
addressing sleep loss and daytime sleepiness. Although there is
empirical evidence for young and elderly adults [10], relevant
research on adolescents is scarce, and studies have predomi-
nantly involved cross-sectional designs [11,12]. In previous stud-
ies, we have been able to show a relation between exercising and
improved subjective and objective sleep in adolescent elite ath-
letes [13,14] and in moderately exercising adolescents [15], com-
pared with control subjects. A limitation was that findings were
derived from cross-sectional group-comparison designs, whereas
intervention studies do allow stronger conclusions regarding the
direct effects of exercising. In this respect, Dworak et al [16] were
able to show that acute bouts of exercise increased objectively
recorded sleep.
Additionally, there is evidence that regular exercise is associ-
ated with improved psychological functioning in adolescents
[14,15,17,18]; exercising buffered the effects of family conflict on
depressed mood [19], and sports participation has been shown to
be a protective factor against depression and suicidal ideation,
mediated by its impact on increased self-esteem and social sup-
port [20]. Additionally, there is also evidence that the implemen-
tation of regular exercise as a therapeutic intervention leads to
positive psychological outcomes: regular exercising improved
self-esteem in children and adolescents [21], and walking regu-
larly for 30– 45 minutes during weekdays for 12 consecutive
weeks led to complete psychiatric remission in half of patients
suffering from therapy-resistant major depressive disorders, as
compared with control subjects [22].
Thus, there is reason to anticipate that exercise interventions
can improve sleep and psychological functioning. Therefore, the
following two hypotheses were formulated. First, we expected
a positive impact of a moderate-to-vigorous exercise (here,
moderate-to-vigorous exercise was defined as planned and con-
tinuous running without interruption at a speed such that con-
versation is not possible) training program on sleep [13,16] as
compared with a control condition. Second, following previous
research [13–18], we anticipated an improvement in psycholog-
ical functioning (such as stress perception, curiosity, somatosen-
sory amplification, mood, concentration, and sleepiness) in exer-
cising adolescents as compared with control subjects.
Methods
Sample
Participants were recruited from a high school in the canton
of Basel-Landschaft, a district of the German-speaking north-
western part of Switzerland. Figure 1 shows the study flowchart
and dropout rates. Of the 60 adolescents originally approached,
51 (85%) completed the study (see Figure 1; age: mean [M] ⫽
18.30 years; standard deviation [SD] ⫽.89; range: 17.5–19.5
years): 27 were female (age: M ⫽18.11 years, SD ⫽.80) and 24
were male (age: M ⫽18.13 years, SD ⫽1.00). Participants were
randomly assigned to either the running or the control group
(see later in the text). Neither gender distribution (
2
(1) ⫽.30,
p⫽.87) nor age (analysis of variance [ANOVA]: group: F(1, 47) ⫽
.01, p⫽.97; gender: F(1, 47) ⫽.02, p⫽.97) differed significantly
between the two groups. Body mass index differed significantly
between male and female participants (male: M ⫽23.29, SD ⫽2.27;
female: M ⫽20.74, SD ⫽1.54; gender: F(1, 47) ⫽18.62, p⫽.000)
but not between the two groups (group: F(1, 47) ⫽.44, p⫽.51).
As in previous studies [14,15], mean weekly vigorous exercise
was assessed through the following question: “For how many
hours do you do vigorous exercise? Vigorous exercise means:
You are playing sports at such a level as to have a markedly
increased heart rate and to sweat.” Answers indicated the num-
ber of hours over which intense exercise was undertaken for
each of the 7 consecutive days. These values were then summa-
rized to generate a total weekly exercise index (hrs/wk). At the
beginning of the study, mean vigorous exercise did not statisti-
cally differ between the two groups (running group [RG]: M ⫽
2.19 hours (SD ⫽1.56); control group [CG]: M ⫽2.24 hours (SD ⫽
1.78); t(49) ⫽.53, p⫽.60, d⫽.15).
All students were informed about the purpose of the study
and about the voluntary basis of participation. Participants were
assured of the confidentiality of their responses and gave written
informed consent. Of the 51 participants, 10 were younger than 18
years. For these participants, parents’ written informed consent
was requested. For participation, they received a voucher of 30.00
Swiss francs for a sports shop. The study was approved by the local
ethics committee of Basel (Switzerland; trial number: 72/10).
Procedure
Figure 1 depicts the study structure, assessments, randomiza-
tion, and total sample sizes. First, a psychiatric interview [23]
ensured that only participants without psychiatric disorders
(e.g., affective disorders, eating disorders, substance abuse disor-
ders, sleep disorders, or others) were enrolled in the study. Ad-
ditionally, brief questions related to physical health state en-
sured that only participants without medical illnesses, allergies,
and cardiovascular, pulmonary, or orthopedic diseases took part
in the study. Thereafter, participants were asked to refrain from
any intake of psychoactive or sleep-altering substances (alcohol,
cannabis, nicotine, mood- or energy-enhancing drinks) for 2
weeks before commencement of and during the study itself.
Participants kept a sleep log (see later in the text) for 3 weeks
(21 consecutive days), covering 3 ⫻5 weekdays and 3⫻2 week-
end days. To compare possible effects of regular running, at the
beginning and at the end of the study, participants completed a
series of questionnaires related to psychological functioning and
sleep (see later in the text). Additionally, at the beginning and at
the end of the study, objective sleep assessment was executed
(which will be described further).
The study was conducted during a school term from mid-
August to the end of September 2010, that is, during the summer
season with high light exposure from early morning.
Participants were randomly assigned to one of the two study
conditions, namely, the RG or the CG. For 3 consecutive weeks,
during the 5 school days per week, participants assigned to the
RG or the CG met every morning at 7
AM
at school. Afterward, the
RG went running for between 30 and 37 minutes. All participants in
the RG had 3 ⫻5 running sessions. Running was cross-country;
after two laps on the school’s running track, running continued in
the forest close to school. Participants were allowed to maintain
their own pace though while running without interruption in
groups of at least four people. Running differed from jogging in
that running speed was such that talking was more difficult. All
participants had to pass a checkpoint and a turning point, and the
track did not allow shortcuts. After participants completed the
session, they got ready for school, and a breakfast was provided
before school commenced.
In contrast to the RG participants, those in the CG remained
on the school’s running track, remained seated, followed
N. Kalak et al. / Journal of Adolescent Health 51 (2012) 615–622616
school activities, and completed homework. They could speak
and interact with each other, but they were not allowed to use
electronic devices such as mobile phones or electronic note-
books. Participants in the CG had 3 ⫻5 sessions of resting
activity. They ceased their activities when the last of the
runners returned. Next, after all participants in the RG had
prepared for school, a breakfast was provided for both groups.
By arranging the same study conditions (with the exception of
running itself) for both the RG and the CG, potential confound-
ers, such as morning schedule, exposure to daylight and arti-
ficial light, social interaction, and eating at school, were rigor-
ously controlled for. Moreover, meeting at 7
AM
at school did
not interfere with participants’ habitual sleep/wake pattern.
Assessing psychological functioning
Daily log. Participants filled out the sleep and mood log in the
evening and in the morning. For evenings, participants answered
questions on an 8-point visual analog scale about sleepiness
during the day (1 ⫽very sleepy; 8 ⫽not at all sleepy), concen-
tration during the day (1 ⫽very bad concentration; 8 ⫽very high
concentration), and mood at bedtime (1 ⫽very bad mood; 8 ⫽
very good mood). For mornings, the questionnaire asked about
sleep quality (1 ⫽very bad sleep quality; 8 ⫽very good sleep
quality) and mood on awaking, using the same analog scale
(Cronbach
␣
⫽.89). Nights were defined as weekday nights if the
participant went to school the next day; weekend nights were
Assessed for eligibility (n= 60)
Excluded (n= 4)
Declined to participate (n= 4)
Analysed (n= 27)
Excluded from analysis (n= 0)
Lost to follow-up (n= 0)
Allocated to intervention (n= 28)
Received allocated intervention (n= 28)
Did not receive allocated intervention
(accident) (n= 1)
Lost to follow-up (n= 0)
Allocated to intervention (n=28)
Received allocated intervention (n= 28)
Did not receive allocated intervention
(illness) (n= 4)
Analysed (n= 24)
Excluded from analysis (n= 0)
Anal
y
p
Randomized (n= 56)
Sleep-EEG-recordings
Questionnaires; beginning of log
Sleep-EEG-recordings
Questionnaires; beginning of log
Sleep-EEG-recordings
Questionnaires; end of log
Sleep-EEG-recordings
Questionnaires; end of log
Figure 1. Flow diagram for recruitment and analysis of participants.
N. Kalak et al. / Journal of Adolescent Health 51 (2012) 615–622 617
Friday and Saturday nights. To compute data, weekdays of a
single week and weekend days of a single weekend were aggre-
gated, resulting in three composite variables for weekdays
(weeks 1, 2, and 3) and three composite variables for weekends
(weekends 1, 2, and 3).
Similar to a previous study [15], we tested whether perceived
stress, coping strategies, somatosensory perception, and curios-
ity and exploratory behavior (see later in the text) might change
in a positive direction over time.
Perceived stress. The Perceived Stress Scale [24] consists of 10
items and was used to determine perceived overall stress occur-
ring over the previous month. Answers were given on a 5-point
rating scale ranging from 1 (never) to 5 (very often), with higher
scores reflecting greater perceived stress (Cronbach
␣
⫽.89).
Coping with stress. The questionnaire consists of 18 items and
assesses positive and negative coping strategies [25]. Positive
coping strategies are those that reduce tension in both the short-
and the long-term, including minimizing the situation, control-
ling the situation, and self-instruction. Negative coping strate-
gies are those that reduce tension in the short-term but increase
stress in the long-term, including social withdrawal, rumination,
and resignation. Answers were given on a 5-point rating scale
ranging from 1 (very unlikely) to 5 (very likely). The higher the
score, the more pronounced is the coping strategy (Cronbach
␣
⫽
.82). Two composite mean scores were computed reflecting pos-
itive and negative coping strategies.
Somatosensory amplification. Somatosensory amplification re-
fers to a tendency to experience somatic and visceral sensations
as unusually intense, noxious, and disturbing. Assessment was
through the Somatosensory Amplification Scale [26]. The ques-
tionnaire consists of 10 items relating to body hypervigilance
and the predisposition to focus on certain weak and infrequent
body sensations. Answers were given on a 5-point rating scale
ranging from 1 (not at all true) to 5 (completely true). Higher
scores reflect an increased tendency to somatosensory amplifi-
cation (Cronbach
␣
⫽.88).
Curiosity and exploratory behavior. Kashdan et al’s [27] Curiosity
and Exploration Inventory was used to assess this dimension.
Curiosity is conceptualized as a positive emotional–motivational
system associated with the recognition, pursuit, and self-regulation
of novelty and challenge. The inventory consists of seven items,
and answers were given on a 7-point rating scale with the anchor
points 1 (not at all true) to 7 (completely true). Higher scores
indicate greater curiosity/exploration (Cronbach
␣
⫽.79).
Sleep evaluation
Objective sleep electroencephalographic recordings. Sleep was ob-
jectively assessed at the beginning and at the end of the study. At
the beginning of the study, before starting the intervention, the
sleep electroencephalographic (EEG) device was applied twice.
With the first application, participants slept with the sleep EEG
device to avoid possible unfavorable “first night effects.” No
registration was performed. The following night, sleep registra-
tion was performed.
On the day of the recording night, participants had to attend
regular schedules, but without evening exercise so as to avoid
possible effects of acute bouts of exercise on sleep [16]. Partici-
pants were requested to go to bed at the usual time, which was
between 9 and 10.30
PM
and to get up between 6 and 6.30
AM
.
After the intervention was completed 3 weeks later, objective
sleep assessment was repeated. Sleep EEG recordings were per-
formed at home using a three-channel EEG device (Fp2-A1, C3-
A2, C4-A1; electrooculogram; electromyogram; SOMNOwatch;
Somnomedics, Randersacker, Germany). Sleep polygraphs were
visually analyzed by two experienced raters according to the
standard procedures [28] (inter-rater reliability:
⫽.91). Raters
were completely blinded to participants’ group assignments. The
SOMNOwatch device provides assessment of TST, sleep period
time, sleep onset latency (SOL), sleep efficiency, stages 1– 4 (min-
utes and %), light sleep (stages 1 ⫹2), slow-wave sleep (stages 3
⫹4), rapid eye movement (REM) sleep, REM sleep latency, and
number and times of awaking after sleep onset.
Subjective assessment of sleep. Participants also completed at the
beginning and at the end of the study the Insomnia Severity
Index [29], a screening tool for insomnia. The seven items, an-
swered on 5-point rating scales (1 ⫽not at all, 5 ⫽very much),
refer to difficulty in falling asleep, difficulties maintaining sleep,
increased daytime sleepiness, and worrying about sleep. The
higher the overall score, the more the respondent is assumed to
suffer from insomnia (Cronbach
␣
⫽.86).
Statistical analyses
To calculate changes on the daily log dimensions (e.g., mood,
concentration) across the 3 weeks, a series of ANOVAs for re-
peated measures with the factors time (six conditions: weekdays
weeks 1–3; weekend days weeks 1–3) and group (RG vs. CG) was
performed. For before to after comparison of objective sleep
variables, ANOVAs for repeated measures were performed with
the factors time (pre vs. post) and group (RG vs. CG) as indepen-
dent variables. In case of deviations from sphericity, statistical
tests were performed using Greenhouse–Geisser-corrected de-
grees of freedom, and the original degrees of freedom are re-
ported with the relevant Greenhouse–Geisser epsilon value ().
Test results with an
␣
level ⬍.05 are reported as significant. Effect
sizes for ANOVAs (partial eta squared [
2
]) were calculated fol-
lowing Cohen [30], with .059 ⱖ
2
ⱖ.01 indicating negligible
practical importance, .139 ⱖ
2
ⱖ.06 indicating moderate prac-
tical importance, and
2
ⱖ.14 indicating crucial practical impor-
tance effect sizes.
Results
Daily log
Tables 1 and 2provide the descriptive and inferential statis-
tical overview of the data from the daily log, separately by groups
(RG vs. CG) and time (weeks 1–3, weekdays and weekend days).
Sleep quality significantly increased over time and was signif-
icantly higher in the RG compared with the CG; the group ⫻time
interaction was also significant, reflecting a significantly greater
increase in sleep quality over time in the RG than in the CG
(Figure 2).
Mood in the morning significantly improved over time and was
significantly higher in the RG than the CG; the group ⫻time inter-
action was also significant; mood in the morning increased signifi-
cantly over time in the RG compared with the CG.
N. Kalak et al. / Journal of Adolescent Health 51 (2012) 615–622618
Concentration during the day did not differ between groups
and did not change over time. However, the group ⫻time inter-
action was significant; concentration increased significantly
over time in the RG, but not in the CG.
Sleepiness during the day did not differ between groups but
decreased significantly over time. The group ⫻time interaction
was significant; sleepiness decreased significantly over time in
the RG compared with the CG.
Mood in the evening did not differ between groups. Over time,
irrespective of group, mood in the evening improved. The group ⫻
time interaction was not significant.
Psychological functioning
Table 3 provides the descriptive and inferential statistical
overview of objective sleep measurements, separately by groups
(RG vs. CG) and time (before vs. after assessment).
Perceived stress,positive and negative coping strategies, and
curiosity and exploratory behavior did not differ significantly be-
tween groups or over time. Moreover, no statistically significant
group ⫻time interactions were observed.
Somatosensory amplification scores decreased significantly
over time. Moreover, the group ⫻time interaction was statisti-
cally significant, with decreased scores over time in the RG com-
pared with the CG.
Sleep
Subjective sleep. Insomnia scores decreased significantly over
time. Moreover, the group ⫻time interaction was statistically
significant, with decreased scores over time in the RG compared
with the CG.
Objective sleep recordings
Table 4 provides the descriptive and inferential statistical
overview of objective sleep measurements, separately for groups
(RG vs. CG) and time (before vs. after assessment).
No statistically significant mean differences between groups,
over time (from before to after assessments), or combining group
and time (group ⫻time interactions) were observed for TST,
awakenings after sleep onset (number, time), stage 2 (minutes),
stage 3 (minutes), light sleep (minutes; %), or REM sleep (%).
SOL significantly decreased in the RG compared with the CG
over time.
Sleep efficiency was significantly higher in the RG compared
with the CG. No statistically significant mean differences were
observed over time or for the group ⫻time interaction.
Over time, stage 1 (minutes; %) significantly decreased, whereas
stage 3 (%) and REM sleep (minutes) significantly increased, with
no mean differences for group or for group ⫻time interactions.
Table 1
Daily log: Overview of descriptive values, separately for weeks (week 1, week 2, week 3), for week days (weekdays, weekend days), and for groups (RG, CG)
Variables Group Study weeks
Week 1 Week 2 Week 3
Weekdays
M (SD)
Weekend
M (SD)
Weekdays
M (SD)
Weekend
M (SD)
Weekdays
M (SD)
Weekend
M (SD)
Sleep quality
a
RG 4.44 (1.05) 5.50 (.95) 5.24 (.84) 6.13 (.78) 5.71 (.75) 6.35 (.72)
CG 4.54 (.98) 5.13 (.90) 4.73 (.78) 4.92 (1.15) 4.80 (1.28) 5.15 (1.08)
Mood: Morning
a
RG 4.28 (.76) 4.44 (1.05) 4.64 (.75) 5.11 (.68) 5.21 (.85) 5.53 (.97)
CG 4.36 (.61) 4.54 (.98) 4.36 (.44) 4.53 (.90) 4.34 (.48) 4.91 (.85)
Concentration during the day
a
RG 4.56 (1.40) 4.83 (1.06) 5.29 (1.18) 4.93 (.95) 5.74 (1.09) 5.60 (.88)
CG 5.33 (1.58) 5.52 (1.00) 4.74 (.89) 5.45 (1.25) 4.71 (.85) 5.19 (.78)
Sleepiness during the day
a
RG 3.84 (.67) 3.77 (.78) 3.93 (.75) 4.35 (.69) 4.77 (.70) 5.74 (.94)
CG 3.98 (.69) 3.69 (.96) 3.90 (.89) 4.13 (.86) 4.01 (.67) 4.17 (1.49)
Mood: Evening
a
RG 3.80 (.86) 4.32 (.87) 3.93 (.81) 4.71 (.71) 5.11 (1.50) 5.02 (.76)
CG 4.12 (1.45) 5.05 (1.00) 4.04 (1.16) 4.86 (.72) 5.10 (1.84) 4.94 (.73)
M⫽mean; SD ⫽standard deviation; RG ⫽running group; CG ⫽control group.
a
Higher means reflect a more positive position on the dimension; for example, sleep quality increased in the weekdays from week 1 (RG ⫽4.44; CG ⫽4.54) to week
3 (RG ⫽6.35; CG ⫽5.15). Likewise, sleepiness during the day decreased from week 1 (RG ⫽3.84; CG ⫽3.98) to week 3 (RG ⫽5.74; CG ⫽4.17).
Table 2
Daily log: Overview of inferential statistics separately for group (RG vs. CG), and time (3 weeks; blocs of weekdays
and weekend days)
Daily log variables Greenhouse–Geisser
epsilon value (ε)
Group Time Group ⫻time
interaction
F
2
F
2
F
2
Sleep quality .611 14.02* .223 17.28* .261 5.98** .11
Mood: Morning .560 4.42*** .083 16.08* .247 6.32** .114
Concentration during the day .687 .00 .000 3.02 .058 17.13* .259
Sleepiness during the day .692 2.68 .052 38.01* .437 2.82*** .054
Mood: Evening .492 1.05 .021 15.50* .240 1.36 .027
Degrees of freedom: Always ⫽(5, 245).
*p⬍.05.
** p⬍.01.
*** p⬍.001.
N. Kalak et al. / Journal of Adolescent Health 51 (2012) 615–622 619
Stage 2 (%) was significantly higher in the control compared
with the RG, with no significant time or group ⫻time interac-
tions.
Stage 4 (minutes, %) was significantly higher in the RG com-
pared with the CG; the significant group ⫻time interaction
showed that stage 4 (minutes, %) increased in the RG compared
with the CG from before to after assessment. Likewise, deep sleep
(minutes; %) was significantly higher in the RG compared with
the CG; moreover, the significant group ⫻time interaction
showed that deep sleep (minutes) increased in the RG compared
with the CG from before to after assessment.
REM sleep latency (minutes) was significantly prolonged in
the RG compared with the CG; moreover, the significant group ⫻
time interaction showed that REM sleep latency (minutes) was
longer in the RG compared with the CG from before to after
assessment.
Discussion
The key finding of the present study is that, compared with a
control condition, an intervention involving running for 30 min-
utes in the morning daily during weekdays for 3 consecutive
weeks improved sleep (objectively and subjectively) and psy-
chological functioning. The results add to the existing literature
in showing that even a short-term intervention of regular run-
ning in the morning does have a favorable impact on the sleep
and psychological functioning of healthy adolescents.
Two hypotheses were formulated and each of these is now
considered in turn.
With the first hypothesis, we expected a favorable impact of
moderate-to-vigorous exercise on adolescents’ sleep [13–16],
and findings fully confirmed this; compared with control sub-
jects, participants in the RG reported improved sleep quality, and
objectively assessed sleep improved, in that deep sleep in-
creased, SOL decreased, and REM sleep latency became longer
over time. Thus, the present findings echo those of the numerous
studies that have confirmed an association between exercise and
sleep [13–16,31]. Importantly, the present results add to the
literature in demonstrating the impact of moderate exercise,
namely, running for 30 minutes in the morning for only 3 weeks.
With the second hypotheses, we anticipated a favorable influ-
ence of regular exercise on psychological functioning, but this
was only partly confirmed. Whereas no changes were observed
in perceived stress and coping or in curiosity and exploratory
behavior, running was related to decreased pain perception. The
lack of any association between exercising and stress is in
marked contrast with the many studies demonstrating a positive
influence of exercise on stress management [32–34]. The follow-
ing are possible explanations for this difference: (a) The time
interval of 3 weeks was too short to induce the relevant changes;
(b) trait (as compared with state) characteristics, such as curios-
ity, perception of stress, and coping strategies, are relatively
stable over time and therefore more difficult to modify; (c) per-
ception of stress and coping did change but in ways too subtle to
be captured by the questionnaires used; (d) participants were
particularly healthy and this would result in ceiling effects.
By contrast, the experience and appraisal of bodily sensations
and its cognitive– emotional elaboration (i.e., the somatosensory
amplification or pain) is much more adaptable and sensitive to
situation-specific conditions [35]. Our results suggest that pain
threshold increased as a function of exercise, and this pattern of
results is in accord with previous findings [15,35].
Despite the clarity of the findings, several issues warrant
against overgeneralization. First, as in all studies with exercising,
confounders, such as exposure to daylight and social contacts,
may bias results. However, we note that social contacts, expo-
0
1
2
3
4
5
6
7
8
W1; WD W1;WE W2; WD W2;WE W3; WD W3;WE
Sleep quality
Weeks
RG
CG
Figure 2. Subjective sleep quality improved over time in both the running group
and the control group; however, improvement was more pronounced in the
running group (for statistics see Tables 1 and 2). W1 ⫽week 1; W2 ⫽week 2;
W3 ⫽week 3; WD ⫽weekdays; WE ⫽weekend days; RG ⫽running group; CG ⫽
control group. For further statistical information, see Tables 1 and 2. Points are
means, and bars are standard deviations.
Table 3
Descriptive and statistical overview of psychological functioning, separately by group (RG vs. CG) and assessment time (before vs. after assessment)
Psychological functioning RG CG Statistics
Before
assessment
n⫽27
M (SD)
After
assessment
n⫽27
M (SD)
Before
assessment
n⫽24
M (SD)
After
assessment
n⫽24
M (SD)
Group Time Group ⫻time
interaction
F
2
F
2
F
2
Perceived stress scale 2.94 (.68) 2.95 (.68) 2.81 (.61) 2.63 (.63) 1.71 .034 2.25 .044 2.97 .057
Coping strategies
Positive coping 3.29 (.64) 3.14 (.78) 3.32 (.68) 3.41 (1.22) .45 .009 .12 .002 1.48 .029
Negative coping 2.93 (1.08) 2.65 (.68) 2.71 (.59) 2.63 (.67) .40 .008 2.26 .044 .67 .014
Somatosensory amplification 14.74 (3.35) 12.14 (4.83) 15.25 (6.72) 15.58 (7.96) 1.55 .031 5.78* .106 9.69** .165
Curiosity and exploratory behavior 5.28 (.58) 5.26 (.49) 5.46 (.69) 5.28 (.73) .37 .008 2.66 .051 1.75 .034
Insomnia severity 13.89 (3.83) 11.22 (3.30) 13.17 (3.03) 13.88 (3.45) 1.14 .023 8.94** .154 26.57*** .352
Degrees of freedom: Always ⫽(1, 49).
*p⬍.05.
** p⬍.01.
*** p⬍.001.
N. Kalak et al. / Journal of Adolescent Health 51 (2012) 615–622620
sure to daylight, and eating at school were equal for both condi-
tions; therefore, these possible confounders could be ruled out.
Second, it is unclear to what extent the intervention was success-
ful because it took place in the mornings as compared with
interventions during the day or in the evening, and research on
this is limited and results are inconclusive [36,37]. Specifically,
there is no research with respect to adolescents. Third, results
may potentially be biased because they are based on data from
particularly healthy and motivated adolescents willing to com-
plete questionnaires and to undergo sleep EEG registration. Thus,
participants might not be representative for adolescents as a
whole [6,7]. In this view, we also note that participants in the
present sample were far from sleeping for the recommended 9
hours per night (see Table 4). Fourth, we did not assess individual
fitness, nor did we assess running objectively. However, it was
our firm intent to arrange a study design close to a practicable
and easy-to-implement intervention. Finally, the relevant di-
mensions may be linked by means of as-yet unidentified further
variables, such as the secretion of adenosine, cortisol, or melato-
nin or issues related to motivation and volition. Based on these
limitations, future research should (a) compare exercise inter-
ventions at different times of day, (b) assess the impact of exer-
cise intervention based on objectively assessed physiological
parameters, or (c) assess neurobiological variables, such as cor-
tisol, melatonin, or BDNF (brain derived neurotrophic factor).
Conclusion
Moderate running in the morning for 3 consecutive weeks
impacted positively on objective and subjective sleep and psy-
chological functioning among healthy adolescents. Moderate,
but regular, exercise such as running should be promoted as both
a remedy and a preventative measure for poor sleep and poor
psychological functioning.
Acknowledgments
The authors thank Marielle Koenig and Vladimir Djurdjevic
for sleep EEG scoring. Moreover, they are grateful to Ladina
Schlatter for data collection and data entry. Finally, they thank
Nick Emler (Surrey, UK) for proofreading the manuscript.
The entire study was conducted without external funding,
and the authors have declared that no competing interests exist.
References
[1] Paus T, Keshavan M, Giedd JN. Why do many psychiatric disorders emerge
during adolescence? Nat Rev Neurosci 2008;9:947–57.
[2] Colrain IM, Baker FC. Changes in sleep as a function of adolescent develop-
ment. Neuropsychol Rev 2011;21:5–21.
[3] Brand S, Kirov R. Sleep and its importance in adolescence and in common
adolescent somatic and psychiatric conditions. Int J Gen Med 2001;4:
425–42.
[4] Mercer PW, Merritt SL, Cowell JM. Differences in reported sleep need among
adolescents. J Adolesc Health 1998;23:59–63.
[5] Moore M, Meltzer LJ. The sleepy adolescent: Causes and consequences of
sleepiness in teens. Pediatr Respir Rev 2008;9:114–20.
[6] Kaneita Y, Ohida T, Osaki Y, et al. Association between mental health status
and sleep status among adolescents in Japan: A nationwide cross-sectional
survey. J Clin Psychiatry 2007;68:1426–35.
[7] Roberts RE, Roberts CR, Duong HT. Chronic insomnia and its negative con-
sequences for health and functioning of adolescents: A 12-month prospec-
tive study. J Adolesc Health 2008;42:294–302.
[8] Lund HG, Reider BD, Whiting AB, Prichard JR. Sleep patterns and predictors
of disturbed sleep in a large population of college students. J Adolesc Health
2010;46:124–32.
[9] Dewald JF, Meijer AM, Oort FJ, et al. The influence of sleep quality, sleep
duration and sleepiness on school performance in children and adoles-
cents: A meta-analytic review. Sleep Med Rev 2010;14:179–89.
[10] Loprinzi PD, Cardinal BJ. Associations between objectively-measured physical
activity and sleep, NHANES 2005-2006. Ment Health Phys Act 2011;4:65–9.
Table 4
Descriptive and statistical overview of sleep EEG variables, separately by group (running vs. control) and assessment time (before vs. after assessment)
RG CG Statistics
Before
assessment
n⫽27
M (SD)
After
assessment
n⫽27
M (SD)
Before
assessment
n⫽24
M (SD)
After
assessment
n⫽24
M (SD)
Group Time Group ⫻time
interaction
F
2
F
2
F
2
TST (minutes) 409.32 (24.46) 434.89 (45.10) 400.17 (67.10) 406.00 (56.00) 2.71 .052 3.95 .075 1.56 .031
Sleep onset latency (minutes) 10.17 (5.33) 7.42 (4.73) 9.38 (9.16) 10.43 (4.41) 4.31* .081 3.59 .082 6.12** .120
Sleep efficiency 93.38 (4.41) 94.03 (2.22) 91.83 (4.72) 92.51 (3.65) 4.78* .089 .66 .013 .00 .000
Awakenings after SO (number) 7.52 (7.64) 6.41 (3.25) 5.50 (3.79) 6.67 (2.75) .70 .014 .01 .000 1.75 .034
Awakenings after SO (minutes) 22.76 (21.78) 22.18 (18.07) 29.86 (22.04) 20.67 (9.88) .60 .012 1.64 .032 1.27 .025
Stage 1 (minutes) 13.06 (8.40) 7.57 (3.75) 13.13 (7.98) 10.04 (5.34) .59 .012 24.23*** .331 1.90 .037
Stage 1 (%) 3.23 (2.32) 1.56 (1.14) 3.18 (1.88) 2.43 (1.26) .91 .018 28.24*** .366 3.96 .075
Stage 2 (minutes) 186.28 (15.28) 192.96 (28.95) 190.75 (43.40) 199.21 (45.55) .41 .008 2.26 .044 .03 .001
Stage 2 (%) 45.63 (4.48) 43.58 (6.68) 47.41 (6.92) 48.51 (6.86) 4.90* .091 .28 .006 3.11 .060
Stage 3 (minutes) 28.07 (9.18) 29.65 (12.42) 26.08 (7.34) 30.29 (9.58) .08 .002 3.65 .069 .76 .015
Stage 3 (%) 6.83 (1.99) 8.62 (3.89) 6.68 (2.09) 7.68 (3.05) .61 .012 12.42*** .202 .99 .020
Stage 4 (minutes) 99.20 (21.40) 105.69 (15.05) 87.38 (24.13) 81.88 (11.38) 16.94*** .257 .03 .001 4.13* .078
Stage 4 (%) 24.24 (5.33) 27.17 (7.38) 22.26 (7.02) 20.93 (6.45) 7.18** .128 .61 .012 4.32* .081
Light sleep (minutes) 199.33 (21.05) 200.17 (29.36) 203.04 (45.63) 209.25 (47.54) .51 .010 .46 .009 .27 .005
Light sleep (%) 48.89 (6.33) 45.13 (7.32) 48.26 (10.97) 48.54 (9.85) .51 .010 1.39 .028 1.89 .037
Deep sleep (minutes) 127.22 (24.23) 135.33 (18.04) 114.29 (21.22) 112.17 (15.72) 13.34*** .214 1.19 .024 4.49* .099
Deep sleep (%) 31.05 (5.81) 35.78 (10.39) 29.48 (6.79) 28.06 (8.98) 5.72* .105 1.79 .035 6.17** .112
REM-S (minutes) 82.78 (18.11) 84.63 (27.65) 82.83 (24.77) 99.39 (30.30) 1.42 .028 6.36** .115 4.12 .078
REM-S (%) 20.08 (3.37) 19.09 (5.76) 22.27 (9.09) 23.40 (10.94) 3.60 .068 .003 .000 .65 .013
REM latency (minutes) 83.88 (18.39) 98.53 (36.90) 80.20 (24.43) 75.75 (32.67) 4.50* .091 1.15 .023 4.05* .091
Degrees of freedom: Always ⫽(1, 49).
TST ⫽total sleep time; SO ⫽sleep onset; REM-S ⫽rapid eye movement sleep; M ⫽mean; SD ⫽standard deviation.
*p⬍.05.
** p⬍.01.
*** p⬍.001.
N. Kalak et al. / Journal of Adolescent Health 51 (2012) 615–622 621
[11] TynjÅlÅ J, Kannas L, LevÅlahti E, VÅlimaa R. Perceived sleep quality and its
precursors in adolescents. Health Promot Int 1999;14:155–66.
[12] Liu X, Uchiyama M, Okawa M, Kurita H. Prevalence and correlates of self-
reported sleep problems among Chinese adolescents. Sleep 2000;23:27–34.
[13] Brand S, Beck J, Gerber M, et al. Evidence of favorable sleep-EEG patterns in
adolescent male vigorous football players compared to controls. World
J Biol Psychiatry 2010;11:465–75.
[14] Brand S, Gerber M, Hatzinger M, et al. High exercise levels are related to
favorable sleep patterns and psychological functioning in adolescents: A
comparison of athletes and controls. J Adolesc Health 2010;46:133–41.
[15] Brand S, Gerber M, Beck J, et al. Exercising, sleep-EEG patterns, and psycho-
logical functioning are related among adolescents. World J Biol Psychiatry
2010;11:129–40.
[16] Dworak M, Wiater A, Alfer D, et al. Increased slow wave sleep and reduced
stage 2 sleep in children depending on exercise intensity. Sleep Med 2008;
9:266–72.
[17] Pfeiffer KA, Dowda M, Dishman RK, et al. Sport participation and physical
activity in adolescent females across a four-year period. J Adolesc Health
2006;39:523–9.
[18] Stein C, Fisher L, Berkey C, Colditz G. Adolescent physical activity and
perceived competence: Does change in activity level impact self-percep-
tion? J Adolesc Health 2007;40:462.e1–8.
[19] Sigfusdottir ID, Asgersdottir BB, Sgurdsson JF, Gudjonsson GH. Physical
activity buffers the effects of family conflict on depressed mood: A study on
adolescent girls and boys. J Adolesc 2011;34:895–902.
[20] Babiss LA, Gangwisch JE. Sports participation as a protective factor against
depression and suicidal ideation in adolescents as mediated by self-esteem
and social support. J Dev Behav Pediatr 2009;30:376–84.
[21] Ekeland E, Heian F, Hagen KB. Can exercise improve self esteem in children
and young people? A systematic review of randomised controlled trials. Br J
Sports Med 2005;39:792–8.
[22] Mota-Pereira J, Silverio J, Carvalho S, et al. Moderate exercise improves
depression parameters in treatment-resistant patients with major depres-
sive disorder. J Psychiatr Res 2011;45:1005–11.
[23] Ackenheil M, Stotz G, Dietz-Bauer R, Vossen A. Mini International Neu-
ropsychiatric Interview. Mu¨ nchen, Germany: Psychiatrische Univer-
sitÅtsklinik, 1999.
[24] Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress.
J Health Soc Behav 1983;24:385–96.
[25] Erdmann G, Janke W. Stressverarbeitungsfragebogen [coping with stress-
questionnaire]. G×ttingen, Germany: Hogrefe, 2008.
[26] Barsky AJ, Wyshak G, Klerman GL. The somatosensory amplification
scale and its relationship to hypochondriasis. J Psychiatr Res 1990;24:
323–34.
[27] Kashdan TB, Rose P, Fincham FD. Curiosity and exploration: Facilitating
positive subjective experiences and personal growth opportunities. J Pers
Assess 2004;82:291–305.
[28] Rechtschaffen A, Kales A. A manual of standardized terminology techniques
and scoring system for sleep stages in human subjects. US Department of
Health, Education and Welfare, Public Health Service. Washington DC: US
Government Printing Office, 1968.
[29] Bastien CH, ValliÉres A, Morin CM. Validation of the Insomnia Severity Index
(ISI) as an outcome measure for insomnia research. Sleep Med 2001;2:297–
307.
[30] Cohen J. Statistical power analysis for the behavioural sciences. Hillsdale,
NJ: Lawrence Erlbaum Associates, 1988.
[31] Youngstedt SD, Freelove-Charton JD. Exercise and sleep. In: Faulkner GE,
Taylor AH, eds. London, England: Routledge, 2005:159–89.
[32] Gerber M, Puehse U. Do exercise and fitness protect against stress-induced
health complaints? A review of the literature. Scand J Public Health 2009;
37:801–19.
[33] Gerber M, Pu¨ hse U. “Don’t crack under pressure!” Do leisure time phys-
ical activity and self-esteem moderate the relationship between school-
based stress and psychosomatic complaints? J Psychosom Res 2008;65:
363–9.
[34] Puterman E, Lin J, Blackburn E, et al. The power of exercise: Buffering the
effect of chronic stress on telomere length. PLoS One 2010;5:e10837.
[35] Tajet-Foxell B, Rose FD. Pain and pain tolerance in professional ballet danc-
ers. Br J Sports Med 1995;29:31–4.
[36] Buxton OM, Lee CW, L’Hermite-Baleriaux M, et al. Exercise elicits phase
shifts and acute alterations of melatonin that vary with circadian phase.
Am J Physiol Regul Integr Comp Physiol 2003;284:R714–24.
[37] Cain SW, Rimmer DW, Duffy JF, Czeisler CA. Exercise distributed across day
and night does not alter circadian period in humans. J Biol Rhythms 2007;
22:534–41.
N. Kalak et al. / Journal of Adolescent Health 51 (2012) 615–622622
