Sleep Medicine 14 (2013) 1105–1111

Contents lists available at ScienceDirect

Sleep Medicine
journal homepage: www.elsevier.com/locate/sleep

Original Article

Late bedtimes weaken school performance and predispose adolescents

to health hazards
Ilona Merikanto a,b,⇑, Tuuli Lahti a,c, Riikka Puusniekka d, Timo Partonen a
a
Department of Mental Health and Substance Abuse Services, National Institute for Health and Welfare, Helsinki, Finland
b
Department of Biosciences, University of Helsinki, Helsinki, Finland
c
Department of Behavioural Sciences and Philosophy, University of Turku, Turku, Finland
d
Department of Children, Young People and Families, National Institute for Health and Welfare, Helsinki, Finland

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

Article history: Study objectives: Our study explored if bedtimes influenced school performance and motivation, as well
Received 28 January 2013 as the odds ratio (OR) for health-related concerns in adolescents.
Received in revised form 11 June 2013 Methods: The School Health Promotion Study was based on an anonymous self-report questionnaire con-
Accepted 15 June 2013
ducted in 90% of the municipalities in Finland. The study was conducted during 2008 and 2010 in South-
Available online 11 August 2013
ern Finland, Eastern Finland, and Lapland, and during 2009 and 2011 in Western Finland, Northern
Finland, and Åland. Several indicators were used to measure school performance and motivation. Acci-
Keywords:
dents and health-related complaints, such as depressive symptoms, sleep quality, neck or shoulder pains,
Accident
Adolescent
lower back pains, stomachaches, anxiety or nervousness, irritation or tantrums, headaches, and tiredness
Depressive or dizziness were analyzed in relation to the usual bedtime. Our study had a relatively large sample size
Eveningness (N = 384,076), consisting of students in the eighth and the ninth grades of secondary schools and the first
Motivation and the second grades of upper secondary and vocational schools (ages 14–20 years) in Finland.
Sleep Results: All of the various indicators used to assess school performance and motivation suggest that the
later the bedtime of adolescents, the lower their school performance and their motivation. Similarly later
bedtimes increase the OR for depressive symptoms and other negative health consequences in adoles-
cents as well as a tendency towards accidents. All of these problems were emphasized in students with
bedtimes of 11:30 PM and later.
Conclusions: Late bedtimes, especially those after 11:30 PM, indicate poor sleep which deteriorates
school performance and motivation and increases the OR for depressive symptoms and other health-
related issues in adolescents.
Ó 2013 Elsevier B.V. All rights reserved.

1. Introduction
On average adolescents need 8–9 h of sleep per night to feel rested
[1–3]. There is a large interindividual variation in sleep duration; for
example, the mean sleep duration among 14-year-old adolescents
is 8.6 h (the range 7.2–10.1 h covers all but the top and bottom 2%)
[4]. Sufficient sleep is important for learning and memory functions
and for proper attention and performance, especially among
adolescents [5–10]. During sleep the synaptic connections which
have been active while awake, especially those related to learning
experiences, are strengthened while those that are not as frequently
used are weakened [11–14]. Thus insufficient sleep does not just
⇑ Corresponding author. Address: Department of Mental Health and Substance
Abuse Services, National Institute for Health and Welfare, FI-00271 Helsinki,
Finland. Tel.: +358 295248213.
E-mail addresses: ilona.merikanto@helsinki.fi, ilona.merikanto@thl.fi
(I. Merikanto).
increase daytime sleepiness and a risk for accidents, but it also
may deteriorate cognitive test performance and academic perfor-
mance. Sleep also has a profound influence on motivation and mood.
For example, depressed individuals have significantly more sleep
problems compared to nondepressed individuals [15,16]. Previous
studies with small sample sizes have reported that poor sleep is
associated with weaker academic performance and working mem-
ory as well as compromised emotional information processing and
impaired mood in adolescents [7–10,17–22]. In previous studies,
late bedtimes have been associated with poorer diet and larger body
mass index regardless of sleep duration and with a higher risk for
childhood depression [23–25].
In our study, we explored if bedtimes influenced adolescents’
school performance and motivation, as well as their tendency to
have accidents, depressive symptoms, poor sleep quality, neck or
shoulder pains, lower back pains, stomachaches, anxiety or ner-
vousness, irritation or tantrums, headaches, or tiredness or dizzi-
ness. Our hypothesis is that later bedtimes increase health

1389-9457/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.sleep.2013.06.009
1106 I. Merikanto et al. / Sleep Medicine 14 (2013) 1105–1111
hazards and poorer performance in school, as sleep time is more
restricted in school days compared to those with earlier bedtimes.
In contrast to previous studies, our sample size was relatively large
and we used several indicators for the outcome.
2. Methods
2.1. Participants
The School Health Promotion Study consisted of participants in
eighth and ninth grades of secondary schools and first and second
grades of upper secondary and vocational schools (ages 14–
20 years). The data collection was based on an anonymous self-re-
port paper questionnaire conducted in 90% of the municipalities in
Finland, and it took place locally in schools under teacher supervi-
sion during 2008 and 2010 in Southern Finland, Eastern Finland,
and Lapland and during 2009 and 2011 in Western Finland, North-
ern Finland, and Åland. Data from 384,076 participants (195,458
girls and 188,618 boys; total participation rate, 97.9%) were used
for the analyses in our study.
2.2. Assessment
Bedtimes were categorized into four classes based on an answer
to the question, ‘‘What time do you generally go to sleep on school
days?’’(10 response options: 1 = approximately 9:00 PM; and
10 = approximately 1:30 AM or later): those who went to sleep
around 10:00 PM or earlier (18.0% of boys and 25.2% of girls); those
who went to sleep around 10:30 PM (20.2% of boys and 25.9% of
girls); those who went to sleep around 11:00 PM (23.9% of boys
and 23.1% of girls); and those who went to sleep around
11:30 PM to 1:30 AM or later (37.9% of boys and 25.8% of girls).
School performance was assessed with 11 different items,
including if the student had difficulties following class learning,
with teamwork, with homework, in preparing for examinations,
finding a suitable study method, initiating or finishing self-directed
tasks, completing writing tasks, completing reading tasks, com-
pleting practical work tasks, and getting along with teachers or
friends (scale, 1 = not at all; 2 = quite little; 3 = quite much;
4 = very much). School motivation was assessed with the following
six items: days of school absence during last 30 days due to skip-
ping, days of school absence during last 30 days due to illness, days
of school absence during last 30 days due to other reasons (scale,
1 = not at all; 2 = one day; 3 = 2–3 days; 4 = over 3 days), feeling
of being overwhelmed with school work, feelings that school had
lost meaning, and feelings of inadequacy in school performance
(scale, 1 = hardly ever; 2 = a few times a month; 3 = a few days a
week; 4 = almost daily).
Health-related issues were assessed with 16 different items,
including the students’ own opinions of their health status (scale,
1 = very good; 2 = quite good; 3 = mediocre; 4 = very bad) and
whether or not a student had had neck or shoulder pains, lower
back pains, stomachaches, anxiety or nervousness, irritation or tan-
trums, trouble falling asleep or nightly awakenings, headaches, or
tiredness or dizziness during the last 6 months (scale, 1 = rarely
or never; 2 = approximately once a month; 3 = approximately once
a week; 4 = almost daily). Quality of sleep was assessed with stu-
dents’ own opinions of their sleep (scale, 1 = no problems with
sleeping; 2 = sleeping as well as before; 3 = feeling more tired once
having woken up than before; 4 = insomnia is bothersome;
5 = insomnia or having difficulties with falling asleep or waking
up too early or in the middle of sleep). Tendency for accidents
was assessed with five items, including accidents during the past
term in recess, in gym class, in other classes, in work training, or
on the way to school requiring medical attention (scale, 1 = none;
2 = once; 3 = two or several times).
Depressive symptoms were assessed using a modified 11-item
version of the Beck Depression Inventory [26,27] (response alter-
natives are given in parentheses): (1) ‘‘How is your mood’’ (quite
good, not gloomy or sad, feeling gloomy or sad, constant feeling
of gloominess, or so depressed that cannot take it anymore); (2)
‘‘How do you feel about the future?’’ (optimistic, neutral, future
seems quite depressing, I feel the future has nothing to offer, or
the future seems hopeless and I do not think it will get better);
(3) ‘‘How do you consider your life has gone thus far?’’ (I have suc-
ceeded quite often, I do not think I have failed, I have failed often,
my life has been a series of failings, or I have failed completely as a
person); (4) ‘‘How satisfied or unsatisfied do you feel about your-
self?’’ (I am quite pleased with my life, I am not unsatisfied, I do
not enjoy things like before, I do not get satisfaction from any-
where, or I am completely unsatisfied in everything); (5) ‘‘How
do you consider yourself?’’ (I am quite good and not worthless, I
feel myself worthless quite often, I feel myself worthless almost
all the time, or I am altogether worthless); (6) ‘‘Do you have feel-
ings of disappointment?’’ (I am pleased with myself, I am not dis-
appointed with myself, I am disappointed with myself, I am
disgusted with myself, or I hate myself); (7) ‘‘Have you had
thoughts of hurting yourself?’’ (I have never had suicidal thoughts,
no such thoughts or no desire to hurt myself, I feel it would be bet-
ter if I were dead, I have clear plans for suicide, or I would commit
suicide if I had the opportunity); (8) ‘‘How do you feel about meet-
ing strangers?’’ (I like talking to people, I have not lost interest in
other people, others do not interest me as much as they used to,
I have totally lost interest in other people, or I have lost interest
in other people and do not care about them); (9) ‘‘How do you feel
when making decisions?’’ (making different decisions is easy for
me, I can make decisions the same as I used to, my self-confidence
has decreased and I try to avoid making decisions, I have great dif-
ficulties in making decisions, or I cannot make decisions at all);
(10) ‘‘What do you think about your appearance?’’ (I am quite
pleased with my appearance, my appearance does not bother me,
it concerns me that I look unpleasant, I feel I am ugly, or I am sure
I am ugly and repulsive); and (11) ‘‘How is your appetite?’’ (I have
no problems with it, it is the same, it is worse than before, it is
much worse than before, or I have no appetite at all). With this
questionnaire, the item on sleep quality was separately assessed.
2.3. Statistics
Using v2 tests we judged the statistical significance of the dif-
ferences in the distribution of sociodemographic, socioeconomic,
and health characteristics between those whose general bedtime
was at 10:00 PM or earlier and those whose general bedtime was
at 11:30 PM or later (Table 1) and across genders (Supplementary
Table 1). The v2 tests also were used to judge the statistical signif-
icance of the differences in the distribution of school performance
and motivation characteristics (Supplementary Table 2) and health
issues (Supplementary Table 3) across genders.
Complete data for the indicators of school performance and
motivation and health issues for each general bedtime were avail-
able for 104,208 participants (51,982 girls and 52,226 boys; 27% of
the total participation rate). Binary logistic regression analyses
were used to estimate the odds ratio (OR) with 95% confidence
interval for the indicators of school performance and motivation
(Table 2) and for health issues (Table 4) for each general bedtime.
Poisson regression analyses were used to estimate the b values for
the quality of sleep, health status, skipping school, and depressive
symptoms (Table 3). All of the analyses were controlled for gender,
class year, working status, circadian problems due to Internet use,
 I. Merikanto et al. / Sleep Medicine 14 (2013) 1105–1111 1107

Table 1
Background sociodemographic, socioeconomic, and health-related characteristics across bedtimes.

General bedtime on school nights

10:00 PM or earlier Approximately 10:30 PM Approximately 11:00 PM Approximately 11:30 PM or later
(n = 83,089) (n = 88,858) (n = 90,337) (n = 121,792)
Gender (%)
Boys 40.8 42.9 50.0 58.6⁄⁄⁄⁄
Girls 59.2 57.1 50.0 41.4⁄⁄⁄⁄
Year in class (%)
Eighth grade 37.0 29.8 23.9 19.2⁄⁄⁄⁄
Ninth grade 26.3 27.9 27.4 26.4
First year upper secondary 10.5 14.3 14.9 13.5⁄⁄
school
Second year upper secondary 7.5 11.7 13.6 14.5⁄⁄⁄⁄
school
First year vocational school 11.2 9.5 11.6 14.5⁄⁄
Second year vocational school 7.5 6.8 8.6 11.9⁄⁄⁄⁄
Working status during the past school term (%)
No 78.7 78.6 78.0 76.8⁄
Yes 0.4 0.4 0.4 0.5
1–5 h 12.1 12.8 12.2 11.0⁄⁄⁄⁄
6–10 h 4.9 4.9 5.2 5.7
>10 h 3.8 3.4 4.2 6.0⁄⁄⁄⁄
Circadian rhythm problems due to time spent using the Internet (%)
Yes 6.4 11.4 18.0 33.9⁄⁄⁄⁄
No 93.6 88.6 82.0 66.1⁄⁄⁄⁄
Friends to open up to (%)
No friends 9.5 7.7 7.5 8.7
One friend 25.1 22.3 21.2 22.3⁄⁄
Several friends 65.4 70.0 71.3 69.0
Current smoking status (%)
Daily 10.8 11.8 17.8 30.5⁄⁄⁄⁄
Once/wk or more 3.2 4.1 5.4 6.2⁄⁄
Less than once/ wk 6.7 9.7 11.4 11.2⁄⁄⁄⁄
Quit 13.2 14.8 16.3 15.6⁄⁄
Nonsmoker 66.0 59.7 49.2 36.6⁄⁄⁄⁄
Alcohol consumption (%)
Once/wk or more 5.7 6.7 10.7 21.1⁄⁄⁄⁄
Once or twice/mo 24.4 33.7 40.9 42.3⁄⁄⁄⁄
Less than once or twice/mo 20.3 22.7 21.6 17.6⁄⁄
Never 49.6 36.8 26.8 19.0⁄⁄⁄⁄
Exercising for at least a half an hour (%)
Several times/d 14.0 12.5 11.6 10.7⁄⁄⁄⁄
4–7 times/wk 45.0 46.1 44.4 38.6
1–3 times/wk 33.6 34.9 35.8 36.5⁄⁄⁄⁄
Less 5.9 5.6 6.9 10.4⁄⁄⁄⁄
Not at all 1.4 0.9 1.3 3.8⁄
Eating proper breakfast during the school week (%)
5 mornings 70.6 54.6 56.2 39.2⁄⁄
3–4 mornings 12.4 14.7 16.8 16.7⁄⁄⁄⁄
1–2 mornings 7.3 8.9 11.6 15.1⁄⁄⁄⁄
Less 9.7 10.8 15.4 29.0⁄⁄⁄⁄

Abbreviations: h, hour; wk, week; mo, month; d, day.

⁄
P < .05; ⁄⁄P < .01; ⁄⁄⁄P < .001; ⁄⁄⁄⁄P < .0001 as tested between those who went to sleep the earliest and those who went to sleep the latest.

number of friends to open up to, smoking status, alcohol consump-
tion, exercising habits, and breakfast eating habits.
2.4. Ethics
The School Health Promotion Study was approved by the ethical
committee of Tampere University Hospital. It was conducted
according to accepted international ethical standards in accor-
dance with the Declaration of Helsinki and its amendments.
It is of note that students who went to sleep at 11:30 PM or later
more frequently smoke daily, drank alcohol once a week or more,
and did not eat a proper breakfast during the school week, com-
pared with those who went to sleep at 10:00 PM or earlier (Ta-
ble 1). The distribution of the key sociodemographic,
socioeconomic, and health characteristics across genders are pre-
sented in Supplementary Table 1.
3.1. Association of bedtimes with school performance

3. Results Generally, students who went to sleep later than at 10:00 PM

had higher OR for the 12 indicators of negative school performance
The distribution of bedtimes by the key sociodemographic, when compared to those who went to sleep at 10:00 PM or earlier;
socioeconomic, and health characteristics are presented in Table 1. only two indicators yielded no difference (Table 2). In other words,
1108 I. Merikanto et al. / Sleep Medicine 14 (2013) 1105–1111

Table 2
Binary logistic regression predicting school performance by bedtime (odds ratio [95% confidence interval]).a

General bedtime during school nights

Approximately Approximately Approximately 11:30 PM or
10:30 PM 11:00 PM later
Trouble following class education (N = 104,208; no = 33,866, yes = 70,342) 1.1 (1.1–1.2)** 1.2 (1.2–1.3)** 1.4 (1.3–1.5)**
Trouble with teamwork performance (N = 104,150; no = 56,461, yes = 47,689) 0.9 (0.9–1.0) 1.0 (0.9–1.0) 1.0 (1.0–1.1)
Trouble completing homework assignments (N = 104,214; no = 33,790, 1.1 (1.1–1.2)** 1.3 (1.3–1.4)** 1.5 (1.4–1.6)**
yes = 70,424)
Trouble preparing for examinations (N = 104 153; no = 24,075, yes = 80,078) 1.1 (1.1–1.2)** 1.3 (1.2–1.3)** 1.4 (1.4–1.5)**
Trouble finding a suitable study method (N = 104,164; no = 30,835, yes = 73,329) 1.1 (1.0–1.1)** 1.2 (1.1–1.2)** 1.2 (1.2–1.3)**
Trouble initiating or finishing self-directed tasks (N = 104,161; no = 28,529, 1.1 (1.1–1.2)** 1.3 (1.2–1.3)** 1.4 (1.3–1.4)**
yes = 75,632)
Trouble with writing tasks (N = 104,106; no = 35,276, yes = 68,830) 1.1 (1.0–1.1)* 1.2 (1.1–1.2)** 1.2 (1.2–1.3)**
Trouble with reading tasks (N = 104,165; no = 37,169, yes = 66,996) 1.1 (1.0–1.1) 1.2 (1.1–1.2)** 1.2 (1.2–1.3)**
Trouble with practical work tasks (N = 39,033; no = 20,879, yes = 18,154)b 1.0 (1.0–1.1) 1.1 (1.0–1.1) 1.2 (1.1–1.2)**
Trouble getting along with teachers (N = 104 166; no = 49,978, yes = 54,188) 1.1 (1.1–1.1)** 1.2 (1.1–1.2)** 1.3 (1.2–1.3)**
Trouble getting along with school friends (N = 104,162; no = 70,521, yes = 33,641) 0.9 (0.9–0.9)** 0.9 (0.9–1.0) 1.0 (0.9–1.0)
Feelings of being overwhelmed with schoolwork (N = 104,213; no = 36,803, 1.1 (1.0–1.1)* 1.2 (1.1–1.2)** 1.2 (1.2–1.3)**
yes = 67,410)
Feelings that school has lost its meaning (N = 104,156; no = 557,901, yes = 46,255) 1.1 (1.0–1.1) 1.1 (1.1–1.2)** 1.4 (1.4–1.5)**
Feelings of inadequacy in school performance (N = 103,804; no = 57,886, 1.0 (1.0–1.1) 1.1 (1.0–1.1) 1.3 (1.2–1.3)**
yes = 45,918)
a
Adjusted for gender, year in class, working status, circadian problems due to Internet use, number of friends to open up to, smoking status, alcohol consumption,
exercising habits, and breakfast eating habits. Those subjects whose bedtime generally was 10:00 PM or earlier were used as reference category.
b
Assessed only for vocational school students.
*
P < .001.
**
P < .0001.

Table 3
Poisson regression analyses predicting school performance by bedtime (b values [95% confidence interval]).a

General bedtime during school nights

Approximately 10:30 PM Approximately Approximately 11:30 PM or
11:00 PM later
Absence from school during the last 30 days due to illness (n = 102,226) 0.006 (–0.007 to 0.02) 0.03 (0.02–0.05)** 0.1 (0.1–0.1)**
Absence from school during last 30 days due to skipping class (n = 99,936) 0.03 (0.01–0.04) 0.1 (0.09–0.1)** 0.3 (0.3–0.3)**
Absence from school during last 30 days due to other reasons (n = 100,476) 0.02 (0.003–0.03) 0.04 (0.02–0.05)** 0.1 (0.1–0.1)**
Opinion of one’s own health status (n = 104,092) 0.1  104 ( 0.01 to 0.01) 0.03 (0.01–0.04)* 0.1 (0.09–0.1)**
Quality of sleep (n = 104,201) 0.04 (0.02–0.05)** 0.08 (0.07–0.09)** 0.2 (0.2–0.2)**
Modified version of the Beck Depression Inventory (n = 103,346) 0.1 ( 0.2 to 0.1)** –0.06 ( 0.1 to 0.02) 0.5 (0.5–0.6)**
b
Assessed only for vocational school students.
a
Adjusted for gender, year in class, working status, circadian problems due to Internet use, number of friends to open up to, smoking status, alcohol consumption,
exercising habits, and breakfast eating habits. Those whose bedtimes generally was 10:00 PM or earlier were used as reference category.
*
P < .001.
**
P < .0001.

those with later bedtimes were performing worse. The later the
bedtime, the more difficulties the students had completing home-
work assignments, preparing for examinations, following class
learning, initiating or finishing self-directed tasks, and getting
along with teachers. In regard to practical work tasks and feelings
of inadequacy in school performance, only those who went to bed
at 11:30 PM or later had more difficulties than those who went to
sleep at 10:00 PM or earlier (Table 2).
Participants who went to sleep at 10:30 PM or later did not sig-
nificantly differ in teamwork-related tasks from those who went to
sleep at 10:00 PM or earlier. However, those who went to sleep at
10:30 PM or later got along better with friends than those who
went to sleep at 10:00 PM or earlier. There was no significant dif-
ference with getting along with school friends when comparing the
earliest bedtime and bedtimes later than 10:30 PM (Table 2).
3.2. Association of bedtimes with motivation
Similarly school motivation was weaker in those who went to
sleep later than 10:30 PM. Those who went to sleep at 10:30 PM
and later also had more absences from school due to skipping or
other reasons than those who went to sleep at 10:00 PM or earlier
(Table 3). Students with bedtimes later than at 10:00 PM also had
more feelings of being overwhelmed with schoolwork, and stu-
dents who went to sleep 11:00 PM or later also more often felt that
school had lost its meaning (Table 2). Overall, problems with
school performance and motivation were emphasized in students
who went to sleep at 11:30 PM or later.
3.3. Association of bedtimes with health-related issues
Concerning health-related issues, there was a higher OR for
tiredness or dizziness in those with later bedtimes (Table 4). In
addition, the OR for difficulties falling asleep or nightly awaken-
ings, headaches, neck or shoulder pains, irritation or tantrums,
stomachaches, lower back pains, or anxiety or nervousness were
all emphasized in those who went to sleep at 11:30 PM or later.
Students who went to sleep at 11:00 PM or later more
frequently were absent from school due to illnesses compared to
those who went to sleep at 10:00 PM or earlier. These students also
 I. Merikanto et al. / Sleep Medicine 14 (2013) 1105–1111 1109

Table 4
Binary logistic regression analyses predicting health issues by bedtime (odds ratios [95% confidence interval]).a

General bedtime during school nights

Approximately Approximately Approximately 11:30 PM
10:30 PM 11:00 PM or later
Neck or shoulder pains during the last 6 mo (N = 103,942; no = 35,114, yes = 68,828) 1.1 (1.1–1.2)** 1.1 (1.1–1.2)** 1.2 (1.2–1.3)**
Lower back pains during the last 6 mo (N = 103,817; no = 49,795, yes = 54,022) 1.1 (1.0–1.1)* 1.1 (1.1–1.1)** 1.2 (1.2–1.2)**
Stomachaches during the last 6 mo (N = 103,244; no = 45,585, yes = 57,659) 1.1 (1.0–1.1)* 1.1 (1.1–1.1)** 1.2 (1.2–1.3)*
Anxiety or nervousness during the last 6 mo (N = 103,960; no = 38,273, yes = 65,687) 1.1 (1.0–1.1) 1.1 (1.0–1.1) 1.1 (1.1–1.1)**
Irritation or tantrums during the last 6 mo (N = 103,977; no = 38,784, yes = 65,193) 1.1 (1.1–1,1)** 1.1 (1.1–1.2)** 1.2 (1.1–1.2)**
Trouble falling asleep or with nightly awakenings during the last 6 mo (N = 103,956; 1.1 (1.1–1.2)** 1.1 (1.0–1.1)** 1.3 (1.2–1.3)**
no = 49,505, yes = 54,451)
Headaches during the last 6 months (N = 103,988; no = 34,414, yes = 69,574) 1.1 (1.1–1.2)** 1.1 (1.1–1.2)** 1.2 (1.2–1.3)**
Tiredness or dizziness during the last 6 mo (N = 104,043; no = 28,417, yes = 75,626) 1.2 (1.1–1.2)** 1.4 (1.3–1.4)** 1.6 (1.5–1.7)**
Accident in recess during the past school term requiring medical attention (N = 103,859; 0.8 (0.8–0.9) 0.7 (0.7–0.8)** 1.0 (0.9–1.1)
no = 99,977, yes = 3882)
Accident in gym class during the past school term requiring medical attention (N = 103,824; 0.9 (0.9–1.0) 0.9 (0.9–1.0) 1.0 (0.9–1.0)
no = 92,700, yes = 11,124)
Accidents in other classes during the past school term requiring medical attention (N = 103,721; 0.9 (0.8–1.0) 0.8 (0.7–0.8)** 0.9 (0.8–1.0)
no = 99,151, yes = 4570)
**
Accidents in work training during the past school term requiring medical attention (N = 38,924; 0.8 (0.7–1.0) 0.7 (0.6–0.8) 0.9 (0.8–1.0)
no = 37,293, yes = 1631b
Accidents on a way to school during the past school term requiring medical attention 0.8 (0.7–0.9)** 0.7 (0.7–0.8)** 0.9 (0.8–1.0)
(N = 103,727; no = 99,162, yes = 4565)

Abbreviation: mo, month.

a
Adjusted for gender, year in school, working status, circadian problems due to Internet use, number of friends to open up to, smoking status, alcohol consumption,
exercising habits, and breakfast eating habits. Those whose bedtimes generally was 10:00 PM or earlier were used as reference category.
b
Assessed only for vocational school students.
*
P < .001.
**
P < .0001.

had a worse opinion of their health status compared to those who
went to sleep at 10:00 PM or earlier. Furthermore, students with
bedtimes at 10:30 PM and later felt that their sleep quality was
worse than those who went to sleep at 10:00 PM or earlier
(Table 3).
Moreover, those who went to sleep at 10:30 PM or at 11:00 PM
experienced less depressive symptoms than those with bedtimes
of 10:00 PM or earlier; yet, depressive symptoms more frequently
occurred among students who went to sleep at 11:30 PM or later
compared to those who went to sleep at 10:00 PM or earlier (Table 3).
In addition, the OR for accidents requiring medical attention
that occurred during recess, other classes, or work training were
lower among those who went to sleep at 11:00 PM than among
those who went to bed at 10:00 PM or earlier (Table 4); regarding
these accidents, individuals who went to sleep at 10:30 or
11:30 PM or later did not significantly differ from those who went
to sleep at 10:00 PM or earlier. For accidents on the way to school,
the OR was lower among students with bedtimes of 10:30 or
11:30 PM than those with bedtimes of 10:00 PM or earlier.
3.4. Gender differences in school performance and motivation
Girls had more difficulties in preparing for examinations and in
finding a suitable study method than boys, whereas boys had more
difficulties with writing tasks than girls (Supplementary Table 2).
Girls also got along better with teachers than boys. In general girls
had lower school motivation than boys. First they had had more
absences from school due to skipping or other reasons than boys.
Second girls had more feelings of being overwhelmed with school-
work than boys and feelings that school had its lost meaning. Fur-
thermore, girls more often felt that their school performance was
inadequate (Supplementary Table 2).
3.5. Gender differences in health-related issues
Girls had more health-related issues than boys, more absences
from school due to illnesses than boys (Supplementary Table 2),
and also had a worse opinion of their health status and felt that their
sleep quality was worse. They also more frequently reported
depressive symptoms than boys. Furthermore, girls more frequently
complained of neck or shoulder pains, lower back pains, stomach
ache, anxiety or nervousness, irritation or tantrums, headaches, dif-
ficulties falling asleep or nightly awakenings, and tiredness or dizzi-
ness than boys. In contrast, boys reported more accidents in recess
and in other classes than girls (Supplementary Table 3).
4. Discussion
In our study, weaker school performance and lack of school moti-
vation were associated with later bedtimes. These issues were
emphasized in students who went to sleep at 11:30 PM or later. Later
bedtimes also were associated with poor sleep quality, sleep prob-
lems, and tiredness. Similarly the OR for other health-related issues
was increased in those who went to sleep later than at 10:00 PM.
Thus our results are in line with previous studies indicating that poor
sleep in adolescents might deteriorate academic performance and
predispose individuals to health issues and that the OR for these out-
comes was increased. However, our study was the first to present re-
sults using multiple items for school performance and motivation
(e.g., absences) and for different health-related issues and hazards
(e.g., accidents) from a large nationwide sample covering the whole
age group of the eighth and ninth grade population.
Based on our results on self-reported sleep, late bedtimes indi-
cated poor sleep. In our study, students with later bedtimes were
not satisfied with their sleep quality and reported more sleep prob-
lems (e.g., tiredness, insomnia symptoms, difficulties falling asleep,
or awakening too early) compared to students with earlier bed-
times. Late bedtimes also might indicate a tendency towards eve-
ningness. In previous studies, eveningness also was related to
sleep problems [28,29]. Furthermore, it is possible that later bed-
times may lead to the accumulation of sleep debt, as students have
to wake up early for school on school days. Compensatory sleep
during weekends may delay circadian rhythms even further, as
students with generally later bedtimes are unlikely to advance
their schedules [30].
In almost all the indicators of school performance and motiva-
tion, the trend was toward more difficulties in school and weaker
1110 I. Merikanto et al. / Sleep Medicine 14 (2013) 1105–1111
motivation as bedtimes were delayed. However, those who went to
sleep at 10:30 PM got along better with friends than those who
went to sleep at earlier bedtimes. These students also performed
as well as those who went to sleep earlier in practical tasks, while
those who went to sleep at 11:30 PM or later had more difficulties.
Similarly depressive symptoms were reported less in students who
went to sleep at 10:30 or 11:00 PM compared to those who went to
bed earlier or later. These students also had more friends to open
up to compared to those who went to bed earlier or later, which
might in turn lower depressive symptoms and also improve team-
work performance [31,32]. Interestingly these students also had
less accidents requiring medical attention; regarding accidents
on the way to school, those who went to sleep at 10:00 PM or ear-
lier had a higher risk for these accidents than those who went to
sleep later.
It is not surprising that those with later bedtimes had lower
motivation for school, as poor sleep lowers memory and learning
ability, and thus overall performance [5,6,18,29]. For instance, later
bedtimes were associated with feeling overwhelmed with school-
work. Absences from school due to skipping also increased, as bed-
times were delayed. This finding may have been due to either a
lack of interest in school, of which these students also reported
more than those who went to sleep earlier, or due to students
being too tired to go to their morning classes. To our knowledge,
our study is the first to report absences from school due to skipping
or other reasons in relation to general bedtimes.
Our data on the health-related issues, such as neck or shoulder
pains, lower back pains, stomachaches, anxiety or nervousness,
irritation or tantrums, headaches, and tiredness or dizziness, con-
stitute a new addition to the research literature. The later the bed-
time, the more problems students were experiencing in regard to
all of the named health-related issues. Later bedtimes also in-
creased absences from school due to illness and were associated
with a poorer opinion of their own health status. Because depres-
sive symptoms and sleep problems usually are intertwined, it is
not surprising that those who went to sleep later had more depres-
sive symptoms [15,29]. For example, depressed individuals often
show deficiencies in melatonin secretion, having either lower mel-
atonin levels or an earlier timing in their melatonin secretion
rhythm compared to healthy individuals [15,33–35]. This observa-
tion also could partly explain why those who went to sleep early
also had more depressive symptoms compared to those who went
to sleep at 10:30 PM.
Those with later bedtimes had some lifestyle habits that might
delay bedtimes, increase depressive symptoms, and lead to health
hazards (e.g., more alcohol consumption, smoking status, excessive
exercising or hardly at all, sleeping problems from Internet usage,
eating less breakfast than those with earlier bedtimes). Likewise,
previous studies have found an association between insufficient
sleep and an increased risk for smoking and alcohol consumption,
lower physical activity, depressive symptoms, excessive computer
usage, and physical fighting [36–38]. In 2011 Foti et al. [36] sug-
gested that excessive physical activity and computer usage led to
insufficient sleep times among adolescents. Similarly working dur-
ing a school term leaves less time for schoolwork and free time, un-
less one is spared from going to sleep later. Thus it is not surprising
that those who worked more during a school term had later bed-
times. On the other hand, it has been reported that adolescents
had a higher tendency towards eveningness than older individuals
[39]. Further, it is possible that later bedtimes might increase the
likelihood of unhealthy lifestyle habits, predispose individuals to
depression, and deteriorate school performance. For instance,
evening-type ninth graders are more likely to experiment with
smoking and to smoke on a daily basis, to consume more alcohol,
and to engage in less physical activity than intermediate-type or
morning-type ninth graders [40].
Furthermore, the tendency towards eveningness predisposes
individuals to ‘‘social jetlag,’’ as school and other social schedules
interfere with sleep preferences during weekdays, thus resulting
in misalignment of the internal circadian timing system [41,42].
It also has been proposed that the association of smoking with eve-
ningness might be a consequence of social jetlag [41]. The ten-
dency towards eveningness is heightened in puberty, especially
in girls [43].
Girls had lower school motivation than boys, even though their
bedtimes generally were earlier. On the other hand, girls were hav-
ing more sleeping problems, such as more insomnia symptoms,
trouble falling asleep or staying asleep, and more feelings of tired-
ness than boys. Thus it is not surprising that girls had more school
absences due to skipping or illness, more feelings that schoolwork
was overwhelming, and more feelings of health-related issues than
boys.
It has been suggested that class starting times should be de-
layed to better cater to the adolescent circadian rhythm. Although
these actions have reportedly improved attention and lessened
daytime sleepiness, academic performance has even been reported
to deteriorate the OR for an increase in alcohol consumption
[38,44,45]. If classes were to start later, they also must end later.
Precaution should be taken to prevent this change from excessively
delaying other daily activities, as there more frequently would be
late-opening hours for a broad range of recreational activities to
meet the increasing demand, which would not help in increasing
sleep time. Instead favoring schedules that promote sufficiently
long nighttime sleep and behaviors that promote early enough
bedtimes should be emphasized to avoid social jetlag.
4.1. Limitations and strengths
A limitation to our study is that our data are based on self-re-
port only. We also lacked exact information regarding sleep dura-
tion and waking times and separate data regarding weekend
changes. Furthermore, we did not use structured sleep rating scales
and had no information regarding the stage of puberty or medical
history of the respondents. However, we were able to use diverse
data indicative of school performance and motivation, depressive
symptoms, and other health characteristics. A strength of our study
was the large sample size derived from students aged 14 to
20 years on a nationwide basis.
5. Conclusions
Later bedtimes, especially those after 11:30 PM, are related to
poorer school performance and less motivation. They also are asso-
ciated with depressive symptoms and an increased OR for a nega-
tive health status in adolescents. Schedules that encourage earlier
bedtimes would be welcome to promote health and well-being
among adolescents.
Conflict of interest
The ICMJE Uniform Disclosure Form for Potential Conflicts of
Interest associated with this article can be viewed by clicking on
the following link: http://dx.doi.org/10.1016/j.sleep.2013.06.009.
References
[1] Carskadon MA, Harvey K, Duke P, Anders TF, Litt IF, Dement WC. Pubertal
changes in daytime sleepiness. Sleep 1980;2:453–60.
[2] Carskadon MA. The second decade. In: Guilleminault C, editor. Sleeping and
waking disorders: indications and techniques. Menlo Park: Addison-Wesley;
1982.
 I. Merikanto et al. / Sleep Medicine 14 (2013) 1105–1111
[3] Mercer PW, Merritt SL, Cowell JM. Differences in reported sleep need among
adolescents. J Adolesc Health 1998;23:259–63.
[4] Iglowstein I, Jenni OG, Molinari L, Largo RH. Sleep duration from infancy to
adolescence: reference values and generational trends. Pediatrics
2003;111:302–7.
[5] Gais S, Lucas B, Born J. Sleep after learning aids memory recall. Learn Mem
2006;13:259–63.
[6] Potkin KT, Bunney Jr WE. Sleep improves memory: the effect of sleep on long
term memory in early adolescence. PLoS One 2012;7:e42191.
[7] Mak KK, Lee SL, Ho SY, Lo WS, Lam TH. Sleep and academic performance in
Hong Kong adolescents. J Sch Health 2012;82:522–7.
[8] Boschloo A, Krabbendam L, Dekker S, Lee N, de Groot R, Jolles J. Subjective
sleepiness and sleep quality in adolescents are related to objective and
subjective measures of school performance. Front Psychol 2013;4:38.
[9] Stroebele N, McNally J, Plog A, Siegfried S, Hill JO. The association of self-
reported sleep, weight status, and academic performance in fifth-grade
students. J Sch Health 2013;83:77–84.
[10] Rhie S, Lee S, Chae KY. Sleep patterns and school performance of Korean
adolescents assessed using a Korean version of the pediatric daytime
sleepiness scale. Korean J Pediatr 2011;54:29–35.
[11] Cirelli C, Tononi G. Is sleep essential? PLoS Biol 2008;6:e216.
[12] Donlea JM, Shaw PJ. Sleeping together using social interactions to understand
the role of sleep in plasticity. Adv Genet 2009;68:57–81.
[13] Gilestro GF, Tononi G, Cirelli C. Widespread changes in synaptic markers as a
function of sleep and wakefulness in Drosophila. Science 2009;324:109–12.
[14] Vyazovskiy VV, Olcese U, Lazimy YM, Faraguna U, Esser SK, Williams JC, et al.
Cortical firing and sleep homeostasis. Neuron 2009;63:865–78.
[15] Srinivasan V, Pandi-Perumal SR, Trakht I, Spence DW, Hardeland R, Poeggeler
B, et al. Pathophysiology of depression: role of sleep and the melatonergic
system. Psychiatry Res 2009;165:201–14.
[16] Merikanto I, Lahti T, Kronholm E, Peltonen M, Laatikainen T, Vartiainen E, et al.
Evening-types are prone to depression [published online ahead of print May
20, 2013]. Chronobiol Int 2013;30:719–25.
[17] Gomes AA, Tavares J, de Azevedo MH. Sleep and academic performance in
undergraduates: a multi-measure, multi-predictor approach. Chronobiol Int
2011;28:786–801.
[18] Jiang F, Van Dyke RD, Zhang J, Li F, Gozal D, Shen X. Effect of chronic sleep
restriction on sleepiness and working memory in adolescents and young
adults. J Clin Exp Neuropsychol 2011;33:892–900.
[19] Kim SJ, Lee YJ, Cho SJ, Cho IH, Lim W, Lim W. Relationship between weekend
catch-up sleep and poor performance on attention tasks in Korean adolescents.
Arch Pediatr Adolesc Med 2011;165:806–12.
[20] Soffer-Dudek N, Sadeh A, Dahl RE, Rosenblat-Stein S. Poor sleep quality
predicts deficient emotion information processing over time in early
adolescence. Sleep 2011;34:1499–508.
[21] Dagys N, McGlinchey EL, Talbot LS, Kaplan KA, Dahl RE, Harvey AG. Double
trouble? the effects of sleep deprivation and chronotype on adolescent affect. J
Child Psychol Psychiatry 2012;53:660–7.
[22] Danielsson NS, Harvey AG, Macdonald S, Jansson-Fröjmark M, Linton SJ. Sleep
disturbance and depressive symptoms in adolescence: the role of catastrophic
worry. J Youth Adolesc 2013;42:1223–33.
[23] Olds TS, Maher CA, Matricciani L. Sleep duration or bedtime? Exploring the
relationship between sleep habits and weight status and activity patterns.
Sleep 2011;34:1299–307.
[24] Golley RK, Maher CA, Matricciani L, Olds TS. Sleep duration or bedtime?
Exploring the association between sleep timing behaviour, diet and BMI in
children and adolescents. Int J Obes (Lond) 2013;37:546–51.
1111
[25] Lin JD, Tung HJ, Hsieh YH, Lin FG. Interactive effects of delayed bedtime and
family-associated factors on depression in elementary school children. Res
Dev Disabil 2011;32:2036–44.
[26] Beck AT, Beck RW. Screening depressed patients in family practice: a rapid
technic. Postgrad Med 1972;52:81–5.
[27] Beck AT, Rial WY, Rickels K. Short form of depression inventory: cross-
validation. Psychol Rep 1974;34:1184–6.
[28] Merikanto I, Kronholm E, Peltonen M, Laatikainen T, Lahti T, Partonen T.
Relation of chronotype to sleep complaints in the general Finnish population.
Chronobiol Int 2012;29:311–7.
[29] Tzischinsky O, Shochat T. Eveningness, sleep patterns, daytime functioning,
and quality of life in Israeli adolescents. Chronobiol Int 2011;28:338–43.
[30] Åkerstedt T, Kecklund G, Selén J. Early morning work—prevalence and relation
to sleep/wake problems: a national representative survey. Chronobiol Int
2010;27:975–86.
[31] Clark MS, Jansen KL, Cloy JA. Treatment of childhood and adolescent
depression. Am Fam Physician 2012;86:442–8.
[32] Greydanus DE, Calles Jr J. Suicide in children and adolescents. Prim Care
2007;34:259–73.
[33] Beck-Friis J, Kjellman BF, Aperia B, Unden F, von Rosen D, Ljunggren JG, et al.
Serum melatonin in relation to clinical variables in patients with major
depressive disorder and a hypothesis of low melatonin syndrome. Acta
Psychiatr Scand 1985;71:319–30.
[34] Wetterberg L, Aperia B, Gorelick DA, Gwirtzman HE, McGuire MT, Serafetinides
EA, et al. Age, alcoholism and depression are associated with low levels of
urinary melatonin. J Psychiatry Neurosci 1992;17:215–24.
[35] Monteleone P, Maj M. The circadian basis of mood disorders: recent
developments and treatment implications. Eur Neuropsychopharmacol
2008;18:701–11.
[36] Foti KE, Eaton DK, Lowry R, McKnight-Ely LR. Sufficient sleep, physical activity,
and sedentary behaviors. Am J Prev Med 2011;41:596–602.
[37] McKnight-Eily LR, Eaton DK, Lowry R, Croft JB, Presley-Cantrell L, Perry GS.
Relationships between hours of sleep and health-risk behaviors in US
adolescent students. Prev Med 2011;53:271–3.
[38] Onyper SV, Thacher PV, Gilbert JW, Gradess SG. Class start times, sleep, and
academic performance in college: a path analysis. Chronobiol Int
2012;29:318–35.
[39] Roenneberg T, Kuehnle T, Juda M, Kantermann T, Allebrandt K, Gordijn M, et al.
Epidemiology of the human circadian clock. Sleep Med Rev 2007;11:429–38.
[40] Urbán R, Magyaródi T, Rigó A. Morningness-eveningness, chronotypes and
health-impairing behaviors in adolescents. Chronobiol Int 2011;28:238–47.
[41] Wittmann M, Dinich J, Merrow M, Roenneberg T. Social jetlag: misalignment of
biological and social time. Chronobiol Int 2006;23:497–509.
[42] Touitou Y. Adolescent sleep misalignment: a chronic jet lag and a matter of
public health. J Physiol Paris 2013. http://dx.doi.org/10.1016/
j.jphysparis.2013.03.008. [Epub ahead of print].
[43] Carskadon MA, Vieira C, Acebo C. Association between puberty and delayed
phase preference. Sleep 1993;16:258–62.
[44] Lufi D, Tzischinsky O, Hadar S. Delaying school starting time by one hour:
some effects on attention levels in adolescents. J Clin Sleep Med
2011;7:137–43.
[45] Wahlstrom K. School Start time and sleepy teens. Arch Pediatr Adolesc Med
2010;164:676–7.