Sleep duration and activity levels in Estonian and Swedish children ...

Eur J Appl Physiol (2011) 111:2615–2623 DOI 10.1007/s00421-011-1883-6

ORIGINAL ARTICLE

Sleep duration and activity levels in Estonian and Swedish children and adolescents Francisco B. Ortega • Jonatan R. Ruiz • Idoia Labayen • Lydia Kwak • Jaanus Harro • Leila Oja • Toomas Veidebaum • Michael Sjo¨stro¨m

Received: 17 September 2010 / Accepted: 18 February 2011 / Published online: 5 March 2011 Ó Springer-Verlag 2011

Abstract We aimed to examine the associations of sleep duration with time spent on sedentary, moderate and vigorous activities in children and adolescents. The sample consisted of 2,241 (53.5% girls) Estonian and Swedish children (9–10 years) and adolescents (15–16 years), from the European Youth Heart Study, in 1998–1999. Sleep duration was calculated by the difference between selfCommunicated by Susan A. Ward.

Electronic supplementary material The online version of this article (doi:10.1007/s00421-011-1883-6) contains supplementary material, which is available to authorized users. F. B. Ortega (&) J. R. Ruiz I. Labayen L. Kwak M. Sjo¨stro¨m Unit for Preventive Nutrition, Department of Biosciences and Nutrition at NOVUM, Karolinska Institutet, 14183 Huddinge, Sweden e-mail: [email protected] F. B. Ortega Department of Medical Physiology, School of Medicine, University of Granada, Granada, Spain J. R. Ruiz Department of Physical Education and Sport, School of Physical Activity and Sport Sciences, University of Granada, Granada, Spain I. Labayen Department of Nutrition and Food Science, University of the Basque Country, Vitoria, Spain J. Harro Department of Psychology, Estonian Centre of Behavioral and Health Sciences, University of Tartu, Tartu, Estonia L. Oja T. Veidebaum Estonian Centre of Behavioral and Health Sciences, National Institute for Health Development, Tallinn, Estonia

reported bedtime and time for getting up on a normal weekday. Sedentary time/physical activity was measured by accelerometry (valid data on 1,462 participants). Adolescents had lower odds than children, and Swedish higher odds than Estonian, of meeting the sleep recommendations ([9 h) (OR = 0.22, 95% CI 0.17–0.27; and 1.32, 1.07–1.61, respectively). Participants sleeping longer than 10 h spent more time on physical activities (all intensities) and less time on sedentary activities than those sleeping shorter durations (all P \ 0.001). The associations with physical activity became non-significant after additional adjustment for age or sexual maturation (Tanner stages), whereas the associations with sedentary time became borderline significant (P = 0.09/0.03, for age and Tanner, respectively). In conclusion, these results do not suggest a link between sleep durations and activity in a relatively large sample of children and adolescents from two European countries. Consequently, the common assumption that physical activity is a mediator in the relationship between short sleep durations and obesity is not supported by our findings. Keywords

Sleep Exercise Childhood Adolescence

Introduction Physically active people have higher levels of healthrelated fitness, a lower risk of developing a number of chronic diseases, and lower rates of various chronic diseases than people who are inactive (U.S. Department of Health and Human Services; Physical Activity Guidelines Advisory Committee 2008). In youth, strong evidence demonstrates that both physical fitness and health status are substantially enhanced by frequent physical activity (PA)

123

2616

(Ruiz and Ortega 2009). Physically active children and adolescents, compared to their inactive peers, have higher levels of cardiorespiratory endurance and muscular strength, and well-documented health benefits including reduced body fatness, more favourable cardiovascular and metabolic disease risk profiles, enhanced bone health, and reduced symptoms of anxiety and depression (U.S. Department of Health and Human Services; Physical Activity Guidelines Advisory Committee 2008). In addition to the health benefits of being highly active, health risks are related to sedentary behaviours (Dunstan et al. 2010; Hamilton et al. 2007; Healy et al. 2008). Sedentary behaviour or inactivity refers to activities that do not increase energy expenditure substantially above the resting level and includes activities such as sleeping, sitting, lying down, and watching television, and other forms of screen-based entertainment (Pate et al. 2008). Operationally, sedentary behaviour includes activities that involve energy expenditure equivalent to 1 metabolic equivalent units (METs) to 1.5 METs (Pate et al. 2008). Sedentariness or physical inactivity has been considered as one of the biggest public health problems of the twentyfirst century (Blair 2009) and a leading cause of death (Mokdad et al. 2004). In young people, high doses of sedentary behaviours, such as TV viewing, are related to a higher metabolic risk, independently of PA (Ekelund et al. 2006). Moreover, results from longitudinal studies suggested that sedentary time (TV viewing) during childhood and adolescence is related to adult health and mortality (Hancox et al. 2004). Sleep is vital to cognitive performance, productivity, health and well-being; even mild sleep restriction degrades performance over a few days (Krueger et al. 2008). Findings from cross-sectional and longitudinal studies related short sleep duration to overweight/obesity status in children, adolescents and adults (Chaput et al. 2006, 2007; Eisenmann et al. 2006; Must and Parisi 2009; Sekine et al. 2002; Taheri 2006; Taheri et al. 2004; von Kries et al. 2002). Sleep duration is important in the regulation of body weight homeostasis by modulating key hormones, such as leptin and ghrelin (Spiegel et al. 2004; Taheri et al. 2004). Another potential mechanism when short sleep duration could result in obesity could be morning tiredness, which may hamper physical activity and promote sedentary behaviours. Although this mechanism has been consistently suggested in the literature, there is little empirical data supporting a link between sleep duration and inactivity/activity. The available information between sleep duration and sedentarism in young people is mainly based on self-reported behaviours such as TV viewing and time spent in front of the computer, which generally suggests a higher sedentary time in short sleepers (Ozturk et al. 2009; Van den Bulck 2004; Wells et al. 2008). Objective data on

123

Eur J Appl Physiol (2011) 111:2615–2623

sedentary time in youth is lacking. Regarding PA and its association with sleep duration, the currently available information is inconsistent and contradictory in youth (Gupta et al. 2002; Van den Bulck 2004; Wells et al. 2008). We found one study that measured PA objectively by means of a wrist accelerometer (Gupta et al. 2002). The authors did not report any association between sleep duration and PA. The use of accelerometers in large scale population-based studies is increasing, and provides an opportunity to further explore the relationships between sleep duration and objectively measured sedentary and physical activity at different intensities. In this context, the present study aimed to examine the associations of sleep duration with time spent on sedentary, moderate and vigorous activities, as measured by accelerometry, in Estonian and Swedish children and adolescents.

Methods Study sample and design Data collection took place from 1998 to 1999 in healthy children (9–10 years) and adolescents (15–16 years) from Estonia and Sweden. In Estonia, the city of Tartu and its surrounding rural area was the geographical sampling area. In Sweden, seven municipalities in the Stockholm area and ¨ rebro were chosen for data collection. The pooling one in O of data was assumed to be possible because of the use of identical protocols, designed and carried out simultaneously, in both countries (Poortvliet et al. 2003; Riddoch et al. 2005; Wennlo¨f et al. 2003). A total of 2,241 children and adolescents (1,172 Estonian and 1,069 Swedish; 1,042 boys and 1,199 girls) had valid and complete data on gender, age, bedtime and when they awoke and were therefore included in the study. A subsample of 1,462 participants (784 children and 678 adolescents) had valid data on objectively assessed PA. Participants with valid data and those with missing data on PA were similar in the key study variable, i.e. % participants meeting the sleep recommendations ([9 h) was 74% and 77%, respectively (Chi-square tests, P [ 0.1). The study was approved by local ethical committees ¨ rebro City Council no. 690/98, Huddinge University (O Hospital no. 474/98, and University of Tartu no. 49/301997). One parent or legal guardian provided written informed consent, and all children gave verbal consent. Sleep data The participants answered two sleep-related questions in a self-administered questionnaire: (1) ‘What time do you usually get up on a School day?’ There were four possible


answers ranging from \06:30 to [07:30, at 30 min intervals; (2) ‘What time do you usually go to bed on a School day?’ with four possible answers ranging from \20:00 to [22:00, at 1 h intervals. The sleep duration was calculated by the difference between bedtime and time for getting up. For the extreme categories, e.g. going to bed later than 10 pm, the time given, i.e. 10, was used for the calculations. For the middle categories, e.g. going to bed between 8 and 9 pm, an average value, i.e. 8.5, was used for the analyses. The National Sleep Foundation defines optimal sleep in children and adolescents as sleeping more than 9 h (National Sleep Foundation 2006). In the present study, a three-category variable (B9 h,[9 h and B10 h,[10 h) and a two-category variable (B9 h,[9 h) were used in the analyses. Sedentary time and physical activity Sedentary time and physical activity were measured over four consecutive days (including at least one weekend day) with an activity monitor (MTI model WAM 7164, Manufacturing Technology Inc., Shalimar, Florida, formerly known as Computer Science and Applications Inc.) worn on the right hip. Participants were asked to wear the accelerometer during the daytime (from waking-up to bedtime), and taking it off only for water-based activities (e.g. swimming, showering, etc.). At least 3 days of recording (including a weekend day), with a minimum of 10 h registration per day was set as an inclusion criterion. The epoch time was set to 1 min. Activity indicators include sedentary, moderate, vigorous and moderate to vigorous PA (MVPA); they were calculated and presented as the average time per day during the complete registration (min/day). Further, the average PA was calculated and expressed as the total counts recorded, divided by the total daily registered time (counts/ min). Sedentary behaviour was considered as the time spent under 100 counts/min. Moderate PA (3–6 METs) and vigorous PA ([6 METs) intensities were based on cut-off limits published elsewhere (Trost et al. 2002). Moderate to vigorous PA was the time spent in at least moderate intensity PA ([3 METs) calculated as the sum of time spent in moderate and vigorous PA. Confounders Country, sex, age, sexual maturation, daylight length and maternal education were used as confounders in the analyses. Sexual maturation status was assessed by a trained physician of the same gender as the child, using brief observation, according to Tanner and Whitehouse (Tanner and Whitehouse 1976). Breast development in girls and genital development in boys were used for pubertal classification. Daylight length: Light time in a day greatly

2617

differ from December 22nd (the ‘shortest day’) to June 22nd (the ‘longest day’). This difference is greater in Nordic countries, such as Estonia and Sweden, than in countries closer to the equator line. Since this fact can potentially influence some behaviours (such as bedtime and/or PA patterns), we computed a daylight length score using the information from the test date. We set December 22nd as day 0 and performed an increasing day count both forward and backward until June 22, so that the higher the score the higher the daylight length. Theoretically, this score could range from 0 to 178; however, no examinations were carried out during Christmas or summer holidays so the actual range observed in the study sample was from 4 to 156 (mean ± standard deviation, 73.9 ± 39.8). There were no differences in daylight length between Estonian and Swedish participants (73.8 ± 42.4 vs. 74.1 ± 36.8, respectively, P [ 0.8). Maternal educational level was used as a proxy of socio-economic status (Kaplan and Keil 1993) and also based on what parents reported; it was categorized as at both below university level and at university level. Statistical analysis Sleep patterns are described as percentage of participants within each sleep duration category in the whole sample, and by country, sex and age. The odds of meeting current sleep recommendations ([9 h) according to country, sex and age were calculated by binary logistic regression. All the factors mentioned were simultaneously entered into the model, so that they were all adjusted for each other. Bivariate correlations (Spearman coefficients) among the main study variables and confounders were performed in order to preliminary examine the association between sleep duration and activity indicators. This analysis also allowed to determine whether the potential confounders were correlated to the main study variables (both exposures and outcomes) and should therefore be accounted for in further analyses. Differences in sedentary time and physical activity according to sleep duration categories (B9 h, [9 h and B10 h, [10 h) were examined by analysis of variance (ANOVA) and covariance (ANCOVA), without and with adjustment for the confounders (country, sex, age and daylight length), respectively. After squared root transformation of vigorous PA, all the residuals showed a satisfactory distribution. Bonferroni adjustments were used for the post-hoc comparisons. Country, sex and age did not interact with sleep duration in its association with activity indicators (all P [ 0.2). Therefore, the analyses were performed in the whole sample. Statistical power was above 90% in all the models. All calculations were performed using SPSS v.17.0 software for Windows. For all analyses, the significance level was 5%.

123

2618

Results The percentage of participants according to sleep duration categories is depicted in Fig. 1. Overall, 25% of the participants slept B9 h and another 25% slept [10 h. The percentage of participants sleeping more than 10 h was higher in Swedish than in Estonian (32% vs. 19%, respectively) and in children (9 years) than in adolescents (15 years) (48% vs. 2%, respectively). Overall, 76% of the participants met the sleep recommendations (optimal sleep [9 h). Median (percentile 25th and 75th) sleep duration in the 9-year participants was 10:00 h (9:15 h and 10:15 h) in boys and 10:15 h (9:30 h and 10:15 h) in girls (Independent-Samples Median test, P = 0.001). The corresponding figures in the 15-year participants were 9.15 h (8:45 h and 9:15 h) in boys and 9:15 h (8:45 h and 9:15) in girls (P = 0.1). The odds ratio (OR) and 95% confidence intervals (CI) for meeting current sleep recommendations according to country, sex and age is shown in Table 1. Swedish participants showed higher odds of meeting the sleep recommendations than their Estonian peers (OR = 1.30, 95% CI 1.06–1.60). Adolescents (15 years) had lower odds of meeting the sleep recommendations compared with children (9 years) (OR = 0.22, 95% CI 0.17–0.27), whereas no significant differences were found between boys and girls. Sleep duration was negatively correlated with sedentary time and positively correlated with all the PA indicators (all P \ 0.001) (Table 2). We also found significant correlations between the confounders (country, sex, age, sexual maturation and daylight length) and most of the main study variables (sleep duration and activity indicators). Daylight length was positively associated with vigorous PA

Fig. 1 Sleep duration by country, sex and age groups

123

Eur J Appl Physiol (2011) 111:2615–2623 Table 1 Odds ratio (OR) and 95% confidence intervals for meeting current sleep recommendations ([9 h) according to country, sex and age ORa

95% Confidence interval

Estonian

1

(Reference)

Swedish

1.32

1.07–1.61

Country

Sex Boys

1

(Reference)

Girls

0.94

0.77–1.16

9 years

1

(Reference)

15 years

0.22

0.17–0.28

Age group

a

All the variables were simultaneously entered into the model so that they are adjusted for each other

and average PA (both P \ 0.001) and negatively with sleep duration (P \ 0.05). Participants sleeping longer than 10 h spent less time in sedentary activities and more time in physical activities (all the indicators) than those sleeping 10 h or less (all P \ 0.001) (Table 3). The associations persisted after adjustment for country, sex and daylight length (all P \ 0.001), but became non-significant after additional adjustment for age (P = 0.09 for sedentary time and P [ 0.1 for all PA indicators) (Fig. 2). The figure shows how the slopes that connect the mean values become flat after adjustment for age, i.e. no difference among means, which indicate that the association between sleep duration and activity was explained by age. For exploratory purposes, we conducted the analyses stratifying by age group and the same pattern was observed in both the 9-year-old


2619

Table 2 Bivariate correlations (Spearman coefficients) among the main study variables and confounders Sex

Age

Sexual Daylight Sleep maturation length duration

Country (1 = Est, 2 = Swe) -0.014 -0.008 0.013

Sex (1 = Boys, 2 = Girls)

0.038 0.079

-0.058

Age (years)

0.131

0.016

0.902

0.117

Sexual maturation (stages)

0.079

Daylight length (days)

0.012

-0.510

-0.457

-0.043*

Sedentary Moderate Vigorous MVPA PA PAà 0.050

0.018

0.150

-0.175

0.651

-0.772

0.614

-0.708

0.038

-0.018

Sleep duration (h/day)

-0.358

Sedentary (min/day) Moderate PA (min/day)

0.172

-0.230

-0.442

Average PA (counts/min)

0.047

0.035

-0.247

-0.402

-0.198 -0.753

-0.690

-0.382

0.009

0.118

-0.403 0.116

0.398

0.219

-0.835

-0.551 0.667

Vigorous PA (min/day)à

0.386

0.208

-0.825 -0.767 0.987 0.816 0.768

0.840 0.867

MVPA (min/day) PA physical activity, MVPA moderate to vigorous physical activity * P \ 0.05;

P \ 0.001;

à

Squared root-transformed

Table 3 Differences in sedentary time and physical activity levels according to sleep duration N

B9 h (a) Mean

SE

[9 h and B10 h (b)

[10 h (c)

Mean

SE

Mean

P SE

Pairwise comparisons* a–b

a–c

b–c

Sedentary (min/day)

1,462

384.5

5.2

359.1

3.6

285.7

4.8

\0.001

[

[

[

Moderate PA (min/day)

1,462

91.9

3.6

110.9

2.4

166.4

3.3

\0.001

\

\

\

Vigorous PA (min/day)

1,462

3.8

0.1

3.9

0.1

4.8

0.1

\0.001

ns

\

\

MVPA (min/day)

1,462

108.6

4.2

129.2

2.9

191.5

3.9

\0.001

\

\

\

Average PA (counts/min)

1,462

580.1

13.3

612.2

9.0

702.4

12.3

\0.001

ns

\

\

Analysis of variance was performed with sleep duration as fixed factor and with activity indicators as dependent variables. SE standard error, PA physical activity, MVPA moderate to vigorous physical activity * Pairwise comparisons: the symbol [ in the column a–b, for instance, indicates a significant difference (P \ 0.001) in the direction a [ b; ns non significant

Squared root-transformed

and the 15-year-old groups, confirming the absence of interaction by age mentioned above. Sleep duration did not relate to activity/inactivity in either 9 or 15-year participants (Supplementary material: Table S1 and Table S2, for 9 and 15-year participants, respectively). Adjusting for sexual maturation instead of age nullified all the associations between sleep duration and PA indicators (P [ 0.1) and substantially attenuated the association between sleep duration and sedentary time, yet remained significant (P = 0.03). Additional adjustment for maternal education did not alter the results.

Discussion The main finding of this study was that individuals sleeping longer spent more time in PA at all intensities, but these associations were fully explained by age or sexual maturation. An explanation for the apparent association between

sleep duration and activity levels might be the fact that adolescents sleep shorter and are less active than children, so a model not adjusting for age can provide misleading results on this topic. Similarly, the association between sleep duration and sedentary time was greatly attenuated after controlling for age or sexual maturation, yet a borderline significant association was still present. We came to a similar conclusion in a previous study conducted in Spanish adolescents using self-report methods to assess physical activity and sedentary behaviours in boys and girls (Ortega et al. 2010). Spanish adolescents sleeping shorter (\8 h) spent more time watching TV than those sleeping longer, whereas the association was not sex-consistent for leisure-time PA. Wells et al. (2008) observed an inverse association between short sleep duration and TV viewing in 10–12year-old Brazilian children. Ozturk et al. (2009) found that 6–17-year-old Turkish children sleeping very little (B4 h/day) spent more time on sedentary behaviour such as computer use; yet, they also observed a lower TV viewing time in

123

2620


Fig. 2 Differences in sedentary time and physical activity levels according to sleep duration after adjustment for potential confounders. Analysis of variance was performed with sleep duration as fixed factor and with activity indicators as dependent variables. PA indicates physical activity; MVPA, moderate to vigorous physical activity. * Squared root-transformed variable

those children. Data from 13 and 16-year-old Belgian adolescents showed an inverse relationship between sleep duration and TV viewing (Van den Bulck 2004). Our findings suggest the need to control the confounding effect of age when examining the association between sleep and activity levels. The study from Wells et al. (2008) comprised a rather age-homogenous study sample so there was no need to control for age. The studies on Turkish (Ozturk et al. 2009) and Belgian (Van den Bulck 2004) children had a wide age range and should account for it. There is no adjustment for age reported in the study conducted in 6–17-year-old Turkish children, while the Belgian study conducted on 13 and 16-year-old adolescents did adjust for age. Age-adjustment is a key confounder, and it complicates comparison in-between studies if it is not performed in all the studies of interest. In addition, the studies mentioned above were focused on specific sedentary self-reported behaviours (i.e. TV viewing or

123

computer use), while we measured (objectively) overall sedentary time. Physiological interpretation of our results is difficult due to the lack of key sleep-related variables (e.g. sleep quality, sleep latency, day sleepiness) and due to the design of the study. The cross-sectional nature of the available studies (including ours) does not allow us to determine causality or direction of the study associations. In fact, we believe that the associations in this case might be reciprocal. It is reasonable to think that sleep deprivation leads to tiredness the following morning and day, increasing the likelihood of being more sedentary (Taheri 2006), but it is equally possible that watching TV or the use of other technologies during the night may lead to a delayed bedtime, shortening in turn sleep duration, as shown in several studies (Calamaro et al. 2009; Li et al. 2007). The literature on sleep duration and PA is limited and contradictory. The study conducted on 10–12-year-old


Brazilian (children) showed that the children who slept less were more physically active (Wells et al. 2008). Leisuretime PA was assessed by self-reported methods. Selfreported leisure-time PA data on 13 and 16-year-old Belgian adolescents showed no association with sleep duration after adjustment for age (Van den Bulck 2004). A study using wrist accelerometers failed to find an association between sleep duration and total PA in US adolescents, measuring both sleep and activity patterns by actigraphy over a 24 h period (Gupta et al. 2002). Unfortunately, age was not considered, so it is difficult to accurately compare its output with our findings. In a study conducted on Spanish adolescents, we observed a positive relationship between sleep duration and physical activity in boys, but not in girls, after adjusting for age (Ortega et al. 2010). Although further research is still needed, our data together with the previously reported studies do not support that short sleep duration is associated with lower levels of PA in children and adolescents. On the other hand, the studies presented above together with our data on sedentary time seems to support that sleep duration might be more related to sedentary time than to PA at different intensities. More studies using objective methods and adjusting for age/ maturation are needed to confirm or contrast our findings. In addition, future studies should analyse a more complete set of sleep-related variables. It could be that other factors such as sleep quality, sleep latency or day sleepiness, rather than sleep duration, are explaining PA levels in youth. Objective data on key sleep-related variables are needed to better understand the physiological mechanisms underlying the relationship between sleep and activity. In contrary to popular belief, the sleep recommendations for adolescents are the same as for children. Adolescents need at least as much sleep as pre-adolescents (in general, more than 9 h nightly) (Carskadon et al. 1980). The reduced sleep time reported by adolescents may be related more to environmental factors (social, academic and peer pressures) than to a declining ‘need’ for sleep (Carskadon et al. 1980). Research supports the existence of biological mechanisms leading to a later bed time at adolescence, as well as a later time waking up (Carskadon et al. 1998). This phenomenon is known as adolescent phase delay. An optimal modification of the school schedule in order to adjust to the special sleeping pattern of adolescents would be to allow adolescents to start school later in the day than children; exactly the opposite of what is the practice in most countries today. The current system when adolescents are required to start the school very early in the day results in a high percentage of adolescents with chronically altered sleeping patterns, realignment or misalignment of circadian phase relationships affecting sleep and wakefulness (Carskadon et al. 1998; Wolfson and Carskadon 1998).

2621

It is of social and political interest to observe the difference in sleep duration between countries. Estonian children and adolescent were less likely to meet the sleep recommendation than their Swedish peers, a fact that could potentially have health implications. Nevertheless, we want to stress that objective and comparable data on sleep duration in Estonian and Swedish children and adolescents are needed before this novel finding can be further considered. Limitations and strengths Some relevant factors related to sleep patterns are lacking in this study and should be examined in the future, e.g. information about sleep disturbance measured by the number of times waking up during nights, morning tiredness, or sleep duration at the weekends. We did not collect either information on daytime naps or siesta. However, it has been suggested that siesta does not affect the duration of nocturnal sleep (Jean-Louis et al. 2000; Paraskakis et al. 2008; Valencia-Flores et al. 1998). We do not have information on sleep latency, which is also an important sleeprelated factor that has recently showed to be associated with activity levels (Nixon et al. 2009). The assessment of sleep duration was based on participants’ self-reported bed and waking up times. In addition, the questionnaire included restricted categories for bed and waking up times, such as waking up earlier than 6:30, which is a source of error and reduces the sensitivity of the measure. Although, more sophisticated and objective methods to study sleep patterns, such as polysomnography (the gold standard in sleep measurement) or accelerometry, can provide more accurate information, recent research on youth has shown that sleep habits can be studied reliably by looking at children’s selfreports (Loessl et al. 2008; Wolfson et al. 2003). Parental reports on sleep duration via a questionnaire or sleep diary are other common methods used in observational studies, but they also have limitations. Parents may know when their children went to their rooms and theoretically to bed, but they cannot possibly identify exactly at what time they went to bed, particularly when computers, game consoles and TVs are available in the adolescents’ bedroom (Must and Parisi 2009). The present study ought to be acknowledged as it is based on a relatively large sample of children and adolescents from two different countries evaluated by using objective measures of sedentary time and physical activity at different intensities. The lack of significant interaction with country indicates that the results presented are consistent for Estonian and Swedish children. The use of the daylight length as confounder is an original contribution of this study to the previous literature. Our data suggests that longer days, which can also be an indirect indicator of

123

2622


better weather and temperature, is associated with higher levels of PA (average and vigorous), and shorter sleep duration. Future studies examining either PA or sleep patterns should therefore account for this variable, particularly in those geographical areas were a substantial change in the daylight time occurs between summer and winter.

Conclusion This is the first study examining the relationship between sleep duration and objectively measured sedentary behaviour (accelerometry) and that investigates the association between sleep duration and PA at different intensities. Overall, our results do not suggest a link between sleep duration and physical activity. Consequently, the common assumption that physical activity is a mediator in the relationship between short sleep duration and obesity is not supported by our findings. Acknowledgments We are grateful to the participants of the ECPBHS and their parents and the whole ECPHSC study team. We also thank the Swedish participants and families, as well as the EYHS field-work team. We also want to thank Charlotte Goodrose-Flores for the English revision. The study was supported by grants from the Stockholm County Council. This study was also supported by grants from the Estonian Ministry of Education and Science (No 0180027 and 0942706) and the Estonian Science Foundation (No 6932 and 6788). This study is also being supported by grants from the Spanish Ministry of Education (EX-2008-0641), Swedish Council for Working Life and Social Research (FAS), the Swedish Heart–Lung Foundation (20090635); and the Spanish Ministry of Science and Innovation (RYC-2010-05957). Conflict of interest

None declared.

References Blair SN (2009) Physical inactivity: the biggest public health problem of the 21st century. Br J Sports Med 43:1–2 Calamaro CJ, Mason TB, Ratcliffe SJ (2009) Adolescents living the 24/7 lifestyle: effects of caffeine and technology on sleep duration and daytime functioning. Pediatrics 123:e1005–e1010 Carskadon MA, Harvey K, Duke P, Anders TF, Litt IF, Dement WC (1980) Pubertal changes in daytime sleepiness. Sleep 2:453–460 Carskadon MA, Wolfson AR, Acebo C, Tzischinsky O, Seifer R (1998) Adolescent sleep patterns, circadian timing, and sleepiness at a transition to early school days. Sleep 21:871–881 Chaput JP, Brunet M, Tremblay A (2006) Relationship between short sleeping hours and childhood overweight/obesity: results from the ‘Quebec en Forme’ Project. Int J Obes (Lond) 30:1080–1085 Chaput JP, Despres JP, Bouchard C, Tremblay A (2007) Short sleep duration is associated with reduced leptin levels and increased adiposity: results from the Quebec Family Study. Obesity (Silver Spring) 15:253–261

123

Dunstan DW, Barr EL, Healy GN, Salmon J, Shaw JE, Balkau B, Magliano DJ, Cameron AJ, Zimmet PZ, Owen N (2010) Television viewing time and mortality. The Australian Diabetes, Obesity and Lifestyle Study (AusDiab). Circulation 121(3): 384–391 Eisenmann JC, Ekkekakis P, Holmes M (2006) Sleep duration and overweight among Australian children and adolescents. Acta Paediatr 95:956–963 Ekelund U, Brage S, Froberg K, Harro M, Anderssen SA, Sardinha LB, Riddoch C, Andersen LB (2006) TV viewing and physical activity are independently associated with metabolic risk in children: the European Youth Heart Study. PLoS Med 3:e488 Gupta NK, Mueller WH, Chan W, Meininger JC (2002) Is obesity associated with poor sleep quality in adolescents? Am J Hum Biol 14:762–768 Hamilton MT, Hamilton DG, Zderic TW (2007) Role of low energy expenditure and sitting in obesity, metabolic syndrome, type 2 diabetes, and cardiovascular disease. Diabetes 56:2655–2667 Hancox RJ, Milne BJ, Poulton R (2004) Association between child and adolescent television viewing and adult health: a longitudinal birth cohort study. Lancet 364:257–262 Healy GN, Wijndaele K, Dunstan DW, Shaw JE, Salmon J, Zimmet PZ, Owen N (2008) Objectively measured sedentary time, physical activity, and metabolic risk: the Australian Diabetes, Obesity and Lifestyle Study (AusDiab). Diabetes Care 31: 369–371 Jean-Louis G, Kripke DF, Ancoli-Israel S, Klauber MR, Sepulveda RS (2000) Sleep duration, illumination, and activity patterns in a population sample: effects of gender and ethnicity. Biol Psychiatry 47:921–927 Kaplan GA, Keil JE (1993) Socioeconomic factors and cardiovascular disease: a review of the literature. Circulation 88:1973–1998 Krueger JM, Rector DM, Roy S, Van Dongen HP, Belenky G, Panksepp J (2008) Sleep as a fundamental property of neuronal assemblies. Nat Rev Neurosci 9:910–919 Li S, Jin X, Wu S, Jiang F, Yan C, Shen X (2007) The impact of media use on sleep patterns and sleep disorders among schoolaged children in China. Sleep 30:361–367 Loessl B, Valerius G, Kopasz M, Hornyak M, Riemann D, Voderholzer U (2008) Are adolescents chronically sleepdeprived? An investigation of sleep habits of adolescents in the Southwest of Germany. Child Care Health Dev 34:549–556 Mokdad AH, Marks JS, Stroup DF, Gerberding JL (2004) Actual causes of death in the United States, 2000. JAMA 291: 1238–1245 Must A, Parisi SM (2009) Sedentary behavior and sleep: paradoxical effects in association with childhood obesity. Int J Obes (Lond) 33:S82–S86 National Sleep Foundation (2006) Sleep in America Poll. National Sleep Foundation, Washington, DC Nixon GM, Thompson JM, Han DY, Becroft DM, Clark PM, Robinson E, Waldie KE, Wild CJ, Black PN, Mitchell EA (2009) Falling asleep: the determinants of sleep latency. Arch Dis Child 94:686–689 Ortega FB, Chillon P, Ruiz JR, Delgado M, Albers U, AlvarezGranda JL, Marcos A, Moreno LA, Castillo MJ (2010) Sleep patterns in Spanish adolescents: associations with TV watching and leisure-time physical activity. Eur J Appl Physiol 110(3): 563–573 Ozturk A, Mazicioglu M, Poyrazoglu S, Cicek B, Gunay O, Kurtoglu S (2009) The relationship between sleep duration and obesity in Turkish children and adolescents. Acta Paediatr 98:699–702 Paraskakis E, Ntouros T, Ntokos M, Siavana O, Bitsori M, Galanakis E (2008) Siesta and sleep patterns in a sample of adolescents in Greece. Pediatr Int 50:690–693

Eur J Appl Physiol (2011) 111:2615–2623 Pate RR, O’Neill JR, Lobelo F (2008) The evolving definition of ‘‘sedentary’’. Exerc Sport Sci Rev 36:173–178 Poortvliet E, Yngve A, Ekelund U, Hurtig-Wennlof A, Nilsson A, Hagstromer M, Sjostrom M (2003) The European Youth Heart Survey (EYHS): an international study that addresses the multidimensional issues of CVD risk factors. Forum Nutr 56:254–256 Riddoch C, Edwards D, Page A, Froberg K, Anderssen S, Wedderkopp N, Brage S, Cooper AR, Sardinha LB, Harro M, Klasson-Heggebo L, van Mechelen W, Boreham C, Ekelund U, Andersen LB, the European Youth Heart Study team (2005) European Youth Heart Study-Cardiovascular disease risk factors in children: rationale, aims, study design, and validation of methods. J Phys Act Health 2:115–129 Ruiz JR, Ortega FB (2009) Physical activity and cardiovascular disease risk factors in children and adolescents. Curr Cardiovasc Risk Rep 3:281–287 Sekine M, Yamagami T, Handa K, Saito T, Nanri S, Kawaminami K, Tokui N, Yoshida K, Kagamimori S (2002) A dose-response relationship between short sleeping hours and childhood obesity: results of the Toyama Birth Cohort Study. Child Care Health Dev 28:163–170 Spiegel K, Tasali E, Penev P, Van Cauter E (2004) Brief communication: sleep curtailment in healthy young men is associated with decreased leptin levels, elevated ghrelin levels, and increased hunger and appetite. Ann Intern Med 141:846–850 Taheri S (2006) The link between short sleep duration and obesity: we should recommend more sleep to prevent obesity. Arch Dis Child 91:881–884 Taheri S, Lin L, Austin D, Young T, Mignot E (2004) Short sleep duration is associated with reduced leptin, elevated ghrelin, and increased body mass index. PLoS Med 1:e62 Tanner JM, Whitehouse RH (1976) Clinical longitudinal standards for height, weight, height velocity, weight velocity, and stages of puberty. Arch Dis Child 51:170–179

2623 Trost SG, Pate RR, Sallis JF, Freedson PS, Taylor WC, Dowda M, Sirard J (2002) Age and gender differences in objectively measured physical activity in youth. Med Sci Sports Exerc 34:350–355 U.S. Department of Health and Human Services; Physical Activity Guidelines Advisory Committee (2008) Physical activity guidelines for Americans Valencia-Flores M, Castano VA, Campos RM, Rosenthal L, Resendiz M, Vergara P, Aguilar-Roblero R, Garcia Ramos G, Bliwise DL (1998) The siesta culture concept is not supported by the sleep habits of urban Mexican students. J Sleep Res 7:21–29 Van den Bulck J (2004) Television viewing, computer game playing, and Internet use and self-reported time to bed and time out of bed in secondary-school children. Sleep 27:101–104 von Kries R, Toschke AM, Wurmser H, Sauerwald T, Koletzko B (2002) Reduced risk for overweight and obesity in 5- and 6-y-old children by duration of sleep—a cross-sectional study. Int J Obes Relat Metab Disord 26:710–716 Wells JC, Hallal PC, Reichert FF, Menezes AM, Araujo CL, Victora CG (2008) Sleep patterns and television viewing in relation to obesity and blood pressure: evidence from an adolescent Brazilian birth cohort. Int J Obes (Lond) 32:1042–1049 Wennlo¨f A, Yngve A, Sjo¨stro¨m M (2003) Sampling procedure, participation rates and representativeness in the Swedish part of the European Youth Heart Study (EYHS). Public Heal Nutr 6:291–299 Wolfson AR, Carskadon MA (1998) Sleep schedules and daytime functioning in adolescents. Child Dev 69:875–887 Wolfson AR, Carskadon MA, Acebo C, Seifer R, Fallone G, Labyak SE, Martin JL (2003) Evidence for the validity of a sleep habits survey for adolescents. Sleep 26:213–216

123