How Are Adolescents Sleeping? Adolescent Sleep Patterns and Sociodemographic Differences in 24 European and North American Countries

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY-NC-ND 4.0 https://creativecommons.org/licenses/by-nc-nd/4.0/ How Are Adolescents Sleeping? Adolescent Sleep Patterns and Sociodemographic Differences in 24 European and North American Countries © 2020 Elsevier Published version Gariepy, Genevieve; Danna, Sofia; Gobiņa, Inese; Rasmussen, Mette; Gaspar de Matos, Margarida; Tynjälä, Jorma; Janssen, Ian; Kalman, Michal; Villeruša, Anita; Husarova, Daniela; Brooks, Fiona; Elgar, Frank J.; Klavina-Makrecka, Solvita; Šmigelskas, Kastytis; Gaspar, Tania; Schnohr, Christina Gariepy, G., Danna, S., Gobiņa, I., Rasmussen, M., Gaspar de Matos, M., Tynjälä, J., Janssen, I., Kalman, M., Villeruša, A., Husarova, D., Brooks, F., Elgar, F. J., Klavina-Makrecka, S., Šmigelskas, K., Gaspar, T., & Schnohr, C. (2020). How Are Adolescents Sleeping? Adolescent Sleep Patterns and Sociodemographic Differences in 24 European and North American Countries. Journal of Adolescent Health, 66(6, Supplement), S81-S88. https://doi.org/10.1016/j.jadohealth.2020.03.013 2020 Original article How Are Adolescents Sleeping? Adolescent Sleep Patterns and Sociodemographic Differences in 24 European and North American Countries Genevieve Gariepy, Ph.D. a , * ,Sofia Danna, M.Sc. b , Inese Gobin¸ a, Ph.D. c , Mette Rasmussen, Ph.D. d , Margarida Gaspar de Matos, Ph.D. e , Jorma Tynjälä, Ph.D. f , Ian Janssen Ph.D. g , Michal Kalman Ph.D. h , Anita Villeru sa, M.D. i , Daniela Husarova, Ph.D. j , Fiona Brooks, Ph.D. k , Frank J. Elgar, Ph.D. l , Solvita Klavina-Makrecka M.Sc. m , Kastytis  Smigelskas, Ph.D. n , Tania Gaspar, Ph.D. o , and Christina Schnohr, Ph.D. p a Department of Social and Preventive Medicine, University of Montreal, Montreal, Canada b Douglas Mental Health University Institute, Montreal, Canada c Department of Public Health and Epidemiology, Institute of Public Health, Riga Stradins University, Riga, Latvia d National Institute of Public Health, University of Southern Denmark, Copenhagen, Denmark e Institute of Environmental Health, School of Human Kinetics, University of Lisbon, Lisbon, Portugal f Faculty of Sport and Health Sciences, University of Jyvaskyla, Jyvaskyla, Finland g Department of Public Health Sciences, School of Kinesiology and Health Studies, Queen's University, Kingston, Canada h Department of Recreation and Leisure Studies, Faculty of Physical Culture, Palacky University Olomouc, Olomouc, Czech Republic i Department of Public health and Epidemiology, Institute of Public Health, Riga Stradins University, Riga, Latvia j Department of Health Psychology, Faculty of Medicine, PJ Safarik University, Kosice, Slovakia k Office of the Deputy Vice-Chancellor Research, University of Technology Sydney, Ultimo, Sydney, Australia l Institute for Health and Social Policy, McGill University, Montreal, Canada m Department of Public Health and Epidemiology, Riga Stradins University, Riga, Latvia n Health Research Institute, Lithuanian University of Health Sciences, Kaunas, Lithuania o Psychology and Educational Sciences Institute, Lusiada University/CLISSIS, ISAMB/Lisbon University, Lisbon, Portugal p Department of Public Health, University of Copenhagen, Copenhagen, Denmark Article history: Received October 1, 2019; Accepted March 10, 2020 Keywords: Sleep patterns; Sleep; Adolescents; International surveys; Epidemiology; Public Health; Age disparities; Socioeconomic differences ABSTRACT Purpose: Insufficient and poor sleep patterns are common among adolescents worldwide. Up to now, the evidence on adolescent sleep has been mostly informed by country-specific studies that used different measures and age groups, making direct comparisons difficult. Cross-national data on adolescent sleep that could inform nations and international discussions are lacking. We examined the sleep patterns of adolescents across 24 countries and by gender, age, and affluence groups. Methods: We obtained sleep data on 165,793 adolescents (mean age 13.5 years; 50.5% girls) in 24 European and North American countries from the recent cross-sectional Health Behaviour in School-aged Children surveys (2013e2014 and 2017e2018). For each country, we calculated the age-standardized mean in sleep duration, timing, and consistency and the proportions meeting IMPLICATIONS AND CONTRIBUTION The sleep patterns of adolescents vary across countries and by sociodemographic groups. Insufficient sleep on school days is prevalent cross-nationally. National policies and strategies that Conflicts of interest: The authors have no conflicts of interest to disclose. Disclosure: This supplement was supported by the World Health Organization European Office and the University of Glasgow. The articles have been peerreviewed and edited by the editorial staff of the Journal of Adolescent Health. The opinions or views expressed in this supplement are those of the authors and do not necessarily represent the official position of the funder. *Address correspondence to: Genevieve Gariepy, Ph.D., Department of Social and Preventive Medicine, University of Montreal, 7101 du Parc, Montreal, QC H3N 1X9, Canada. E-mail address: geneviev[email protected] (G. Gariepy). www.jahonline.org 1054-139X/Ó2020 Published by Elsevier Inc. on behalf of Society for Adolescent Health and Medicine. This is an open access article under the CC BY-NC-ND license (http:// creativecommons.org/licenses/by-nc-nd/4.0/). https://doi.org/10.1016/j.jadohealth.2020.03.013 Journal of Adolescent Health 66 (2020) S81eS88 sleep recommendations on school and nonschool days from self-reported bedtimes and wake times. We conducted stratified analyses by gender, age, and family affluence group. Results: Adolescent sleep patterns varied cross-nationally. The average sleep duration ranged between 7:47 and 9:07 hours on school days and between 9:31 and 10:22 hours on nonschool days, and the proportion of adolescents meeting sleep recommendations ranged between 32% and 86% on school days and between 79% and 92% on nonschool days. Sleep patterns by gender and affluence groups were largely similar, but older adolescents slept less and went to bed later on school days than younger adolescents in all countries. Conclusions: The sleep patterns of adolescents vary across countries and sociodemographic groups. Insufficient sleep on school days is common in many countries. Public health and policy efforts to promote healthy adolescent sleep are encouraged. Ó2020PublishedbyElsevierInc.onbehalfofSocietyforAdolescentHealthandMedicine.Thisisanopen access article under the CC BY-NC-NDlicense (http://creativecommons.org/licenses/by-nc-nd/4.0/). support adolescent sleep are encouraged. Adolescent sleep is a public health concern in many countries. Poor sleep patternsdincluding sleep that is insufficient, inconsistent, nonrestorative, disrupted, or poorly timed (e.g., late bedtime)daffect some 30%e70% of adolescents in Europe and North America [1,2], where at least one fourth of adolescents have difficulty falling asleep at night and feel tired during the day [1]. Sleep patterns of adolescents differ on school and nonschool days, and their circadian sleepewake cycle is delayed by 2e 3 hours relative to adults [1,3]. Early school start times conflict with this circadian shift [3,4], leading to truncated sleep during the week [5e7]. On weekends and holidays, adolescents tend to sleep longer and later [1]. This discrepancy in sleep times is called social jetlag [8] and has been associated with health, behavioral, and cognitive issues [9e11]. Overall, poor sleep is linked to physical health problems such as colds, headaches, and obesity [12,13], emotional problems such as anxiety, depression, and suicidal thoughts [10,12,14], risky behaviors [15], poor academic performance [9,12], and worse quality of life [13]. Poor sleep also relates to physical inactivity, sedentary behaviors including excessive screen time, and poor diet [12], underscoring the need to include healthy sleep into youth health promotion efforts. The evidence on adolescent sleep is mostly informed by country-specific studies that involve different measures and age groups over the last few decades, making direct comparisons difficult [1,2,16e18]. Nonetheless, there are indications that adolescent sleep varies across regions and cultures. A 2011 systematic review of 41 studies found later bedtimes in Europe than in North America (22:46 vs. 22:06) and longer sleep on school days (8:26 vs. 7:28 hours) and weekends (10:01 vs. 8:48 hours) [1]. The reviewalso found that going to bed about 2 hours later on weekends than on school days was common across regions. Moreover, contemporary information on cross-national sleep patterns is scarce and mostly describes sleep duration, whereas new evidence shows that other sleep patterns also affect health [11]. Social patterns in sleep are found between age, gender, and socioeconomic groups. Older adolescents go to bed later, sleep less on school days, and sleep more on weekendsda common trend in Europe, North America, and Asia [1,18,19]dattributable to biological [3,4], behavioral [20], and social changes [21] during adolescence. Some studies found that boys sleep less and go to bed later than girls [16,19,22], particularly on weekends [19], whereas others found that boys sleep more [2,16,17]. Furthermore, studies from Canada, the U.S., Australia, and Norway found that shorter and poorer quality sleep is more common in lower socioeconomic groups [2,16,23], but comparable data from other countries are scarce. Using data on sleep collected in school-based surveys of adolescents in Europe and Canada, we examined cross-national differences in sleep duration, sleep times, and social jetlag and differences by gender, age, and socioeconomic groups. We hypothesized that sleep patterns would vary across countries and between sociodemographic subgroups. Specifically, we anticipated that older adolescents would go to bed later, sleep less on school days, and sleep more on weekends than younger adolescents, and that lower socioeconomic groups would sleep less than higher socioeconomic groups, but we were agnostic about gender differences, given the mixed findings in the literature. Methods Sample and procedure The Health Behaviour in School-aged Children (HBSC) study is a World Health Organization collaborative cross-sectional study currently conducted in 47 countries across Europe and North America (http://www.hbsc.org)[24]. Data collection procedures in all countries were conducted in accordance with a standardized international protocol [24]. Each member country obtained ethics clearance to conduct the survey from a university-based review board or equivalent regulatory body. Data are collected in school settings every 4 years from a nationally representative random cluster sample of 11-,13-, and 15-year-old adolescents in each participating country. The HBSC survey uses a two-stage sampling to recruit students within stratified samples of schools that represented the regional, economic, and publicprivate distribution in each country. The primary sampling units are schools. More detailed information about methodology of the HBSC study is reported elsewhere [24]. For the first time in 2013e2014 and again in 2017e2018, the HBSC survey included sleep items in one of the survey's optional questionnaire packages. HBSC countries could choose to add this optional package or survey items on sleep to the HBSC mandatory questionnaire, depending on their needs and priorities. Twenty-four countries elected to include sleep items in their 2013e2014 or 2017e2018 national survey and were included in this study (n ¼185,278). For the nine countries that collected sleep data in their 2013e 2014 and 2017e2018 surveys, we combined the samples of both surveys to maximize the use of the available data and the statistical power to estimate subgroup differences within each country. Sleep patterns across the 2013e2014 and 2017e2008 G. Gariepy et al. / Journal of Adolescent Health 66 (2020) S81eS88S82 surveys were comparable. Participants who did not respond to any sleep item (n ¼18,489) or did not report their age (n ¼996) were excluded, leaving a final sample of 165,793 participants from 24 countries (Table 1 and Supplementary Table 1). Measures Sleep patterns were determined using four items that asked, “When do you usually go to bed?”and “When do you usually wake up?”on school days and weekends or holidays (nonschool days). Bedtimes ranged in half-hour intervals from “No later than 21:00”to “2:00 or later”for school days and to “4:00 or later”for nonschool days. Wake times ranged in half-hour intervals from “No later than 5:00”to “8:00 or later”for school days and from “No later than 7:00”to “14:00 or later”for nonschool days. We calculated sleep duration on school and nonschool days using bed and wake times. For each half-hour interval, we used the time at midpoint in our calculations (e.g., 22:15 if the bedtime response was “between 22:00 and 22:30”). For responses at the ends of the scale, we used the minimum/maximum stated time (e.g., 14:00 if the waketime response was “14:00 or later”and 5:00 if the waketime response was “No later than 5:00”). We calculated social jetlag using the difference in bedtimes on nonschool and school days [25]. We also assessed whether students met the minimum sleep recommendation for their age using current international guidelines [26,27], which recommend at least 9 hours for those aged 5e13 years (optimal range 9e11 hours) and 8 hours for those aged 14e17 years (optimal range 8e10 hours). Socioeconomic status (SES) was assessed using the Family Affluence Scale, a checklist of six family assets: number of cars, having own bedroom, number of computers, number of bathrooms, family holidays in the past year, and having a dishwasher. The Family Affluence Scale is a validated scale for comparative measures of family affluence for European and North American adolescents [28]. We categorized students into tertile groups of low, medium, or high family affluence relative to the distribution of total scores from the same age, gender, country, and survey year. We also used self-reported age and gender (boy or girl). Stratified analyses used the three sampled age groups in HBSC [24], 11-year-olds (10.5e12.4 years), 13-year-olds (12.5e 14.4 years), and 15-year-olds (14.5e16.5 years). Statistical analysis All analyses were run separately for each country. We first conducted a descriptive analysis of the sample characteristics. We then estimated the age-standardized average sleep durations, bedtimes, wake times, and social jetlag. We standardized the results for age using the age distribution of all countries combined as a reference to account for the different age distributions across countries. Finally, we conducted stratified analyses by gender, age, and family affluence groups in each country. We statistically compared differences in subgroups using the adjusted Wald test for continuous sleep outcomes and the corrected weighted Pearson chi-square statistic for categorical sleep outcomes to account for the complex survey design. Because of the varying sample sizes across countries (nrange from 2,667 to 15,432), we interpreted the statistical significance with caution and with consideration for the public health relevance of the estimates. Analyses were done using Stata/SE 15 (StataCorp LLC, College Station, TX). We accounted for clustering at the school level and incorporated sampling weights where relevant to ensure results were representative of the country using the svy option in Stata. Results Average sleep patterns on school and nonschool days by country are presented in Figures 1 and 2, respectively, and in Supplementary Table 2. School day sleep duration ranged from 7:47 hours in Poland to 9:07 hours in Flemish Belgium, whereas on nonschool days, it ranged from 9:31 hours in Greece to 10:22 hours in Flemish Belgium. School day bedtimes ranged from 21:42 hours in Flemish Belgium to 23:18 hours in Greece, whereas nonschool day bedtimes ranged from 23:14 hours in Flemish Belgium to 0:54 hours in Spain. School day wake times ranged from 06:18 hours in Hungary to 07:23 hours in Spain. Nonschool day wake times ranged from 09:11 hours in Hungary to 10:36 hours in Spain. The percentage of adolescents meeting sleep recommendations on school days ranged from 32.0% in Poland to 86.3% in Flemish Belgium and on nonschool days from 79.2% in Greece to 92.4% in Flemish Belgium (Figure 3). Social jetlag ranged from 1:11 hours in Moldova to 2:12 hours in Sweden. Boys reported longer sleep duration on school days than girls in two thirds of countries (15/24 countries), although differences were small (10 minutes), even if statistically significant. On nonschool days, girls slept longer in all countries, from 4 minutes more in Switzerland to 34 minutes more in Slovenia. Boys had slightly later bedtimes on school days (2e10 minutes) and nonschool days (1e27 minutes). On school days, girls woke earlier than boys in all countries, but on nonschool days, boys woke earlier than girls in about two thirds of countries. Although boys were more likely to meet sleep recommendations on school days in most countries (17/24 countries), girls were more likely to meet sleep recommendations on nonschool days in all countries. Boys had about the same or slightly more social jetlag than Table 1 Survey years and sample size of Health Behaviour in School-aged Children countries with sleep data Country Survey years Sample size Flemish Belgium 2017e2018 4,305 French Belgium 2013e2014, 2017e2018 9,791 Canada 2013e2014 11,295 Czech Republic 2013e2014, 2017e2018 15,432 Denmark 2013e2014, 2017e2018 6,019 Estonia 2013e2014, 2017e2018 8,633 Finland 2013e2014, 2017e2018 7,975 Greece 2017e2018 3,817 Hungary 2013e2014, 2017e2018 7,397 Iceland 2017e2018 6,967 Latvia 2013e2014, 2017e2018 9,887 The Netherlands 2017e2018 4,634 Norway 2017e2018 3,085 Poland 2017e2018 5,205 Portugal 2017e2018 2,971 Republic of Moldova 2017e2018 4,682 Scotland 2013e2014, 2017e2018 8,794 Slovakia 2013e2014, 2017e2018 10,736 Slovenia 2013e2014 4,982 Spain 2013e2014 7,576 Sweden 2013e2014 7,511 Switzerland 2013e2014 2,667 Ukraine 2017e2018 6,600 Wales 2013e2014 4,832 G. Gariepy et al. / Journal of Adolescent Health 66 (2020) S81eS88 S83 girls, ranging from no difference in Ukraine to 22 minutes in Sweden. Older adolescents slept less in all countries (Supplementary Table 3). On school days, differences between the youngest and oldest groups ranged from 54 minutes in Slovakia to 1:44 hours in Slovenia. On nonschool days, the smallest difference was 11 minutes in Flemish Belgium, and the greatest difference was 52 minutes in Ukraine. Most countries also saw longer delays in bedtimes in older adolescents, both on school days (from 52 minutes in Flemish Belgium to 1:35 hours in Iceland) and nonschool days (from 1:15 minutes in Flemish Belgium and Ukraine to 2:00 hours in Sweden). Although age differences in wake times on school days varied considerably between countries, wake times on nonschool days were consistently later in older adolescents, with 15-year-olds waking 23 minutes later than 11-year-olds in Ukraine and up to 1:22 hours later in Norway. Age differences were also found in the percentage of adolescents who met sleep recommendations. On school days, 11year-olds were most likely to get the recommended amount of sleep, ranging from 42.8% in Poland to 92.9% in Flemish Belgium, whereas on nonschool days, 15-year-olds were most likely to meet recommendations, ranging from 83.7% in Greece to 96.3% in Iceland. Finally, more social jetlag was observed in the oldest group compared with the youngest in all countries. Supplementary Table 4 shows cross-national sleep patterns by family affluence groups. An equal mix of positive and negative affluence gradients was observed for sleep duration. However, most differences between affluence groups were small. On school days, the difference in sleep duration between the least and most affluent adolescents ranged from 31 minutes in Moldova to þ10 minutes in Scotland and Switzerland. On nonschool days, sleep duration between the least and most affluent adolescents ranged from 22 minutes less sleep in Moldova to þ13 minutes more sleep in Scotland. On school days, bedtimes were later in the high affluence group compared with the low affluence group in most countries. The greatest difference was in Moldova (þ31 minutes). In contrast, the high affluence group went to bed earlier in French Belgium, Denmark, Scotland, Wales, and Switzerland, up to 19 minutes earlier in Switzerland. On nonschool days, bedtimes were later in the high affluence group in half the countries, with the largest difference in Moldova (þ54 minutes) and earlier in other countries, the largest difference in Scotland (30 minutes). There was little variability in wake times among affluence groups. On school days, the difference ranged from þ5 minutes later in Canada and Hungary to 10 minutes earlier in Switzerland. There was also no dominating trend on nonschool days. The difference between the high and low affluence groups ranged from þ32 minutes in Moldova to 18 minutes earlier in Scotland. On school days, more low affluence adolescents met sleep recommendations in about half the countries. The greatest difference was in Moldova (þ19.1 percentage points). The high affluence group better met sleep recommendations in other countries, with the highest difference found in Scotland (þ7.2 percentage points). Similar trends were observed on nonschool days. The proportion meeting sleep recommendations on nonschool days was higher in the high affluence group in some countries, with the largest difference in French Belgium (þ5.2 percentage points), and higher in the low affluence group in other countries, with the largest difference reported in Moldova (þ6.5 percentage points). More affluent adolescents reported more social jetlag in most countries. The greatest increase was in Switzerland, where high affluence adolescents had 35 minutes more social jetlag than low affluence adolescents. Conversely, low affluence adolescents had Figure 1. Average sleep times and sleep duration on school days by country. G. Gariepy et al. / Journal of Adolescent Health 66 (2020) S81eS88S84 greater social jetlag in other countries, up to 25 minutes in French Belgium. Discussion This study examined sleep duration, timing, and consistency in over 165,000 adolescents from 24 primarily European countries. Sleep duration ranged from 7:47 to 9:07 hours on school days and from 9:31 to 10:22 hours on nonschool days. A considerably smaller proportion of students met sleep duration recommendations on school days (range 32%e86%) than on nonschool days (range 79%e92%). Social jetlag was observed in all countries, as bedtimes were consistently later and sleep durations consistently longer on nonschool than school days. These differences between school and nonschool days are consistent with adolescent sleep patterns previously observed in the literature [1]. In line with previous studies, sleep timing and duration changed with age [1,18,19] but were only slightly different between boys and girls [2,16,17,19,22]. Older adolescents slept less, went to bed later, and experienced greater social jetlag than younger adolescents. These changes reflect the normative developmental course of the circadian clock, which shifts toward eveningness around puberty, peaks at around age 16 years, and shifts back in early adulthood [3,4]. The increased demands of schoolwork and the use of social media and electronic devices [20] further delay sleep times and add to the biological tendency of older adolescents to fall asleep later, at a time when parental control over sleep times diminishes [21]. However, we examined differences in sleep patterns across age and not pubertal development. Pubertal stage was not assessed in most HBSC countries. Previous studies note that changes in sleep during adolescence are attributable to various physiological and hormonal changes related to puberty and less so with age per se [29,30]. Future studies could add nuance to these analyses by addressing pubertal development in addition to age. Although we found little overall gender differences in sleep patterns, there were notable country-specific differences. For example, boys slept 10 minutes longer on school days than girls in Poland, and girls slept 34 minutes longer on nonschool days in Slovenia, suggesting that gender differences in behavioral patterns, roles, or responsibilities may influence adolescent sleep in some countries. We observed some socioeconomic gradients in sleep measures, but they tended to be small and fluctuate between countries (i.e., positive in some countries and negative in others). Our results differ from previous studies reporting an association between lower SES and shorter sleep among adolescents [2,16,23]. However, previous evidence stemmed from smaller, national samples and used different SES measures. Furthermore, a retrospective cohort study from Norway found that sleep problems were more common in adolescents who experienced downward socioeconomic mobility compared with stable or upward mobility [31]. The discrepancy with our findings may therefore point to methodological differences but also suggest that the relationship between SES and sleep is complex and varies across national and cultural settings. This study found significant cross-national variations in sleep patterns, suggesting that country-level factors may play a role in protecting and promoting adolescent sleep. One determinant may be differences in school start times. A growing body of evidence finds that later school start times are associated with longer sleep durations, primarily by delaying wake times [5e7]. Figure 2. Average sleep times and sleep duration on nonschool days by country. G. Gariepy et al. / Journal of Adolescent Health 66 (2020) S81eS88 S85 Intervention studies have replicated these findings [32], supporting the implementation of this policy to protect adolescent sleep. The HBSC surveys did not collect information on school start times, and national data on school start times are not available. However, a study from Canada found a national average school start time of 08:43 hours in 2013e2014 [6], whereas a study from the U.S. found an average national start time of 08:03 hours in 2011e2012 [7], indicating variability in school start times across countries. Commuting time to school may also affect the sleep patterns for adolescents and vary between countries [33]. The results from a time use study on American high school students suggest that each additional minute spent commuting to school was associated with 1.3 fewer minutes sleeping [34]. The mode of transportation to commute to school (e.g., walking vs. driving) that is normative or accessible in different countries may also affect sleep. In addition, time spent on homework is associated with shorter sleep in adolescents [14] and varies across countries [35]. Moreover, cultural differences in bed times and meal times [36] and in health and lifestyle norms, such as physical activity and healthy eating [20], may contribute to differences in adolescent sleep patterns across countries. Finally, geographic differences could also add to crossnational sleep variations. For instance, in countries situated at higher latitudes, changes in daylight timing and duration throughout the year tend to correlate with sleep times, such that the later and longer daylight of the summer relates to later bedtimes and less sleep compared with the winter [37]. A notable strength of our study is the use of an international standardized protocol for data collection, thereby ensuring high validity of cross-national comparisons. By using large, nationally representative samples from a total of 24 countries, the present study adds significantly to the literature on sociodemographic differences in adolescent sleeping patterns by documenting a number of robust differences across national settings and by highlighting the relevance of national influences for sleep. Our study used self-reported sleep data. Although some studies found that self-reported sleep correlates highly with actigraphy data [38], others have found discrepancies that suggest caution with interpretation [39,40]. However, sleep data were measured using standardized sleep questions making relative cross-national comparisons more likely to be valid despite imprecisions. Furthermore, self-reported sleep data generally overestimate sleep durations because sleep onset tends to be later than bedtime, therefore resulting in more conservative estimates in our study. The cutpoints used to define minimal sleep recommendations [26,27] were based on a systematic review of the best available evidence [13]. However, much of this evidence was not of high quality [13], and it is possible that the cutpoints could change in the future as the quantity and quality of evidence improve. Nonetheless, sleep duration recommendations, even when based on weaker evidence, are important from a population health perspective, as they play an important role in surveillance and public health policy formation. Finally, our data are limited to students attending school and may not be generalizable to adolescents outside the school setting. This study provides the largest cross-national account of adolescent sleep and sociodemographic patterning to date. The results show that, although the sleep patterns of adolescents vary across countries and sociodemographic subgroups, insufficient sleep and poor sleep patterns are prevalent among adolescents in most countries and subgroups. These results build on previous evidence from country-specific studies and point to national settings and sociodemographic groups as strong influences to consider in the promotion of healthy adolescent sleep. Because good sleep is essential to several aspects of a healthy adolescent development, public health policies should Figure 3. Percentage of adolescents that meet sleep recommendations on school days and nonschool days by country. G. Gariepy et al. / Journal of Adolescent Health 66 (2020) S81eS88S86 include cross-sectoral interventions that target adolescent sleep. These policies should consist of modifiable, policy-level measures that support healthy adolescent sleep, such as delaying school start times and educating adolescents, parents, and teachers about the importance of sleep hygiene. Further research on cross-national differences of other aspects of sleep, such as sleeping difficulties and subjective sleep quality, is warranted, as they are important components of overall sleep. Future research is also needed to understand the cross-country behavioral factors and the social, structural, and cultural determinants that influence sleep patterns, such as school and work schedules. Acknowledgments The Health Behaviour in School-aged Children is an international study carried out in collaboration with WHO/EURO. The International Coordinator was Candace Currie (University of St Andrews) for the 2013/2014 survey and Jo Inchley (University of Glasgow) for the 2017/2018 survey. The Data Bank Manager was Professor Oddrun Samdal (University of Bergen). The 2013/2014 and 2017/2018 surveys included in this study were conducted by the following principal investigators in the 24 countries: Flemish Belgium (Bart De Clercq and Anne Hublet), French Belgium (Katia Castetbon and Danielle Piette), Canada (William Pickett and Wendy Craig), Czech Republic (Michal Kalman), Denmark (Mette Rasmussen), Estonia (Leila Oja and Katrin Aasvee), Finland (Jorma Tynjälä), Greece (Anna Kokkevi), Hungary (Ágnes Németh), Iceland (Arsaell M. Arnarsson), Latvia (Iveta Pudule), the Netherlands (Gonneke Stevens, Saskia van Dorsselaer, Wilma Vollebergh, and Tom der Bogt), Norway (Oddrun Samdal), Poland (Joanna Mazur), Portugal (Margarida Gaspar de Matos), Republic of Moldova (Galina Lesco), Scotland (Jo Inchley, Candace Currie), Slovakia (Andrea Madarasova Geckova), Slovenia (Helena Jericek), Spain (Carmen Moreno), Sweden (Petra Lofstedt and Lilly Augustine), Switzerland (Marina Delgrande-Jordan, Hervé Kuendig, and Emmanuelle Kuntsche), Ukraine (Olga Balakireva), and Wales (Chris Roberts). Funding Sources The work was supported by the European Regional Development Fund-Project "Effective Use of Social Research Studies for Practice" (No. CZ.02.1.01/0.0/0.0/16_025/0007294) and by funding from the Technology Agency of the Czech Republic (ÉTA TL01000335) and the Ministry of Education, Youth and Sports, Inter-Excellence, LTT18020 (HBSC Czech Republic); the Public Health Agency of Canada (HBSC Canada); the Juho Vainio Foundation and the University of Jyvaskyla (HBSC Finland); and the PortugalNational Foundation for Science and Technology (HBSC Portugal). Supplementary Data Supplementary data related to this article can be found at https://doi.org/10.1016/j.jadohealth.2020.03.013. References [1] Gradisar M, Gardner G, Dohnt H. Recent worldwide sleep patterns and problems during adolescence: A review and meta-analysis of age, region, and sleep. Sleep Med 2011;12:110e8. [2] Patte KA, Qian W, Leatherdale ST. Sleep duration trends and trajectories among youth in the COMPASS study. Sleep Health 2017;3:309e16. [3] Randler C, Faßl C, Kalb N. From lark to owl: Developmental changes in morningness-eveningness from new-borns to early adulthood. Scientific Rep 2017;7:45874. [4] Crowley SJ, Acebo C, Carskadon MA. Sleep, circadian rhythms, and delayed phase in adolescence. Sleep Med 2007;8:602e12. [5] Marx R, Tanner-Smith EE, Davison CM, et al. Later school start times for supporting the education, health, and well-being of high school students. Cochrane Database Syst Rev 2017;7:CD009467. [6] Gariepy G, Janssen I, Sentenac M, et al. School start time and sleep in Canadian adolescents. J Sleep Res 2017;26:195e201. [7] Wheaton AG, Chapman DP, Croft JB. School start times, sleep, behavioral, health, and academic outcomes: A review of the literature. J Sch Health 2016;86:363e81. [8] Wittmann M, Dinich J, Merrow M, et al. Social jetlag: Misalignment of biological and social time. Chronobiol Int 2006;23:497e509. [9] Diaz-Morales JF, Escribano C. Social jetlag, academic achievement and cognitive performance: Understanding gender/sex differences. Chronobiology Int 2015;32:822e31. [10] Henderson SEM, Brady EM, Robertson N. Associations between social jetlag and mental health in young people: A systematic review. Chronobiology Int 2019;36:1316e33. [11] Touitou Y. Adolescent sleep misalignment: A chronic jet lag and a matter of public health. J Physiology-Paris 2013;107:323e6. [12] Shochat T, Cohen-Zion M, Tzischinsky O. Functional consequences of inadequate sleep in adolescents: A systematic review. Sleep Med Rev 2014; 18:75e87. [13] Chaput J-P, Gray CE, Poitras VJ, et al. Systematic review of the relationships between sleep duration and health indicators in school-aged children and youth. Appl Physiol Nutr Metab 2016;41:S266e82. [14] Becker SP, Langberg JM, Byars KC. Advancing a biopsychosocial and contextual model of sleep in adolescence: A review and introduction to the special issue. J Youth Adolescence 2015;44:239e70. [15] Short MA, Weber N. Sleep duration and risk-taking in adolescents: A systematic review and meta-analysis. Sleep Med Rev 2018;41:185e96. [16] Keyes KM, Maslowsky J, Hamilton A, et al. The great sleep recession: Changes in sleep duration among US adolescents, 1991-2012. Pediatrics 2015;135:460e8. [17] Hysing M, Pallesen S, Stormark KM, et al. Sleep patterns and insomnia among adolescents: A population-based study. J Sleep Res 2013;22:549e56. [18] Gustafsson ML, Laaksonen C, Salantera S, et al. Changes in the amount of sleep and daytime sleepiness: A follow-up study of schoolchildren from ages 10 to 15 years. Int J Nurs Pract 2019;25:e12689. [19] Olds T, Blunden S, Petkov J, et al. The relationships between sex, age, geography and time in bed in adolescents: A meta-analysis of data from 23 countries. Sleep Med Rev 2010;14:371e8. [20] Bartel KA, Gradisar M, Williamson P. Protective and risk factors for adolescent sleep: A meta-analytic review. Sleep Med Rev 2015;21:72e85. [21] Pyper E, Harrington D, Manson H. Do parents’support behaviours predict whether or not their children get sufficient sleep? A cross-sectional study. BMC Public Health 2017;17:432. [22] Norell-Clarke A, Hagquist C. Changes in sleep habits between 1985 and 2013 among children and adolescents in Sweden. Scand J Public Health 2017;45:869e77. [23] Felden ÉPG, Leite CR, Rebelatto CF, et al. Sleep in adolescents of different socioeconomic status: A systematic review. Rev Paul Pediatr 2015;33:467e73. [24] Inchley J, Currie D, Cosma A, et al. Health Behaviour in School-aged Children (HBSC) study protocol: Background, methodology and mandatory items for the 2017/18 survey. St Andrews: CAHRU; 2018. [25] Jankowski KS. Social jet lag: Sleep-corrected formula. Chronobiology Int 2017;34:531e5. [26] Hirshkowitz M, Whiton K, Albert SM, et al. National Sleep Foundation’s updated sleep duration recommendations: Final report. Sleep Health 2015; 1:233e43. [27] Tremblay MS, Carson V, Chaput J-P, et al. Canadian 24-hour movement guidelines for children and youth: An integration of physical activity, sedentary behaviour, and sleep. Appl Physiol Nutr Metab 2016;41: S311e27. [28] Torsheim T, Cavallo F, Levin KA, et al. Psychometric validation of the revised family affluence scale: A latent variable approach. Child Indicators Res 2016;9:771e84. [29] Carskadon MA, Acebo C. Regulation of sleepiness in adolescents: Update, insights, and speculation. Sleep 2002;25:606e14. [30] Sadeh A, Dahl RE, Shahar G, et al. Sleep and the transition to adolescence: A longitudinal study. Sleep 2009;32:1602e9. [31] Sivertsen B, Bøe T, Skogen JC, et al. Moving into poverty during childhood is associated with later sleep problems. Sleep Med 2017;37:54e9. [32] Minges KE, Redeker NS. Delayed school start times and adolescent sleep: A systematic review of the experimental evidence. Sleep Med Rev 2016;28: 86e95. [33] Pereira EF, Moreno C, Louzada FM. Increased commuting to school time reduces sleep duration in adolescents. Chronobiol Int 2014;31:87e94. G. Gariepy et al. / Journal of Adolescent Health 66 (2020) S81eS88 S87 [34] Voulgaris CT, Smart MJ, Taylor BD. Tired of commuting? Relationships among journeys to school, sleep, and exercise among American teenagers. J Plann Education Res 2019;39:142e54. [35] Organisation for Economic Co-operation and Development. Does homework perpetuate inequities in education? Paris, France: OECD Publishing; 2014. [36] Smith CS, Folkard S, Schmieder RA, et al. Investigation of morningeevening orientation in six countries using the preferences scale. Personal Individual Differences 2002;32:949e68. [37] Bartel K, van Maanen A, Cassoff J, et al. The short and long of adolescent sleep: The unique impact of day length. Sleep Med 2017;38:31e6. [38] Nascimento-Ferreira MV, Collese TS, de Moraes ACF, et al. Validity and reliability of sleep time questionnaires in children and adolescents: A systematic review and meta-analysis. Sleep Med Rev 2016; 30:85e96. [39] Short MA, Gradisar M, Lack LC, et al. The discrepancy between actigraphic and sleep diary measures of sleep in adolescents. Sleep Med 2012;13:378e 84. [40] Guedes LG, Abreu GdA, Rodrigues DF, et al. Comparison between self-reported sleep duration and actigraphy among adolescents: Gender differences. Revista Brasileira de Epidemiologia 2016;19: 339e47. G. Gariepy et al. / Journal of Adolescent Health 66 (2020) S81eS88S88