Sleep duration estimates of Canadian children and adolescents

Transcription

1 J Sleep Res. (2016) 25, Sleep duration in young Canadians Sleep duration estimates of Canadian children and adolescents JEAN-PHILIPPE CHAPUT 1 and IAN JANSSEN 2 1 Healthy Active Living and Obesity Research Group, Children s Hospital of Eastern Ontario Research Institute, Ottawa, ON, Canada and 2 School of Kinesiology and Health Studies, Department of Public Health Sciences, Queen s University, Kingston, ON, Canada Keywords bedtime, population health, prevalence, deprivation, surveillance, youth Correspondence: Jean-Philippe Chaput PhD, Healthy Active Living and Obesity Research Group, Children s Hospital of Eastern Ontario Research Institute, 401 Smyth Road, Ottawa, K1H 8L1, Ontario, Canada. Tel.: ext. 3683; fax: ; jpchaput@cheo.on.ca Accepted in revised form 1 March 2016; received 10 December 2015 DOI: /jsr SUMMARY The objective of this study was to provide contemporary duration estimates of Canadian school-aged children and adolescents and to determine the proportion adhering to the duration recommendations. This study included participants aged from the 2013/2014 Canadian Health Behaviour in School-aged Children study (HBSC), a nationally representative cross-sectional study. Bedtime and wake-up times were reported by participants and their duration was calculated. Participants were then classified as having a duration that met the recommended range (9 11 h per night for year-olds or 8 10 h per night for year-olds), a duration that was shorter than the recommended range or a duration that was longer than the recommended range. An estimated 68% of children aged and 72% of adolescents aged for the recommended amount per night when averaged across all days of the week. Short ers represent 31% of school-aged children and 26% of adolescents. Long ers are rare (<2% overall). Children and adolescents ~1 h more at weekends compared to weekdays. Approximately 5% of the participants typically went to bed after midnight on weekdays and 31% did so at weekends; these proportions reached 11 and 45%, respectively, within year-olds. In general, differences in times between boys and girls are small and not clinically significant. In conclusion, almost one-third of Canadian children and adolescents less than the recommended amount. Public health efforts should continue to monitor the of Canadian children and adolescents and identify subgroups of the population more likely to be affected by insufficient. INTRODUCTION Sleep is an essential component of healthy development and is required for physical and mental health (Gruber et al., 2014). However, deprivation has become pervasive in contemporary societies with 24/7 availability of commodities and technologies (Ohayon, 2012). A growing body of evidence shows that chronic deprivation poses a serious threat to the academic success, health and safety of children and adolescents (Owens, 2014). Insufficient is associated with negative outcomes in several areas of health and functioning, including obesity, depression, school performance and risk-taking behaviour (Shochat et al., 2014). The National Sleep Foundation in the United States (Hirshkowitz et al., 2015) recommends that school-aged children aged 6 13 for 9 11 h per night and that adolescents aged for 8 10 h per night. These duration recommendations have also been supported by a recent systematic review, including 141 papers that examined the relationships between duration and various health outcomes in a large number of children and adolescents from 40 different countries (Chaput et al., in press). However, ing for shorter or longer than the recommended times may not necessarily mean that it will affect health adversely, and the ideal amount of may vary from one person to another (Gruber et al., 2014). Sleep duration recommendations are nevertheless relevant from a population health standpoint and play an important role in informing public policies, interventions, parents and children/ youth of healthy behaviours (Matricciani et al., 2012a, 2013). Surveillance of duration at the population level is thus important, and can help to achieve these objectives. 541

2 542 J. -P. Chaput and I. Janssen Although some studies have reported that school-aged children and adolescents now less and report more problems and tiredness compared to decades ago (Keyes et al., 2015; Kronholm et al., 2015; Matricciani et al., 2012b), we are not aware of any published studies reporting on duration on a representative sample of Canadian school-aged children and adolescents. This information is needed, and will help to (i) determine the prevalence of children and adolescents meeting the duration recommendations in Canada; and (ii) tailor and target more effectively future intervention strategies aimed at improving the habits of young Canadians. Therefore, the objective of this study was to provide contemporary duration estimates of Canadian children and adolescents and to determine what proportion of children and adolescents adhere to current duration recommendations. METHODS Study base and sampling This study is based on Canadian records from the 2013/2014 Health Behaviour in School-aged Children study (HBSC). The HBSC is a World Health Organization collaborative cross-national study (Currie et al., 2009). The Canadian HBSC followed the international sampling protocol. Classes within 349 schools were selected for participation using a weighted probability technique to ensure proportional representation based on geographical location, language of instruction, religion and community size. Students in grades 6 10 were the target population. These grades correspond to approximately in Canada, although several 10-, 16- and 17-year-olds are in these grades. Students enrolled in private, special needs, on-reserve or faith schools (other than publicly funded Roman Catholic schools) were not included. Collectively, they represent <7% of the Canadian population in this age range (Freeman et al., 2012). Seventyseven per cent of students who were selected participated in the survey. Participation involved completing a confidential questionnaire in the classroom setting. Consent was obtained from students, their parents/guardians, individual schools and school boards. The study received ethics approval from the General Research Ethics Board of Queen s University (file no ). A small proportion (n =540, 1.8%) of the original sample completed a condensed questionnaire that did not include the time questions. An additional 371 (1.2%) were outside the target age range (e.g. grade 12 students taking a grade 10 class). Thus, the eligible sample for the present study consisted of children and youth aged Of these, 72 (0.1%) were excluded because of missing data on their age and/or gender. An additional 3531 (12.1%) were excluded because they were missing data on one or more of the time questions. Finally, 707 (2.4%) participants whose weekday and/or weekend duration was greater than 3 standard deviations from the mean were excluded, as we assumed that their abnormal data reflected recording errors that occur during survey completion. This left a final sample size of participants for the present study. Participants who were excluded due to missing data did not differ significantly in their descriptive characteristics (except for gender, 41 versus 52% female) compared with those who were included in the analysis. Sleep assessment Within the questionnaire participants were asked to report in h:min the typical time they turned out the lights to go to and the typical time they woke up in the morning during the past week, separately for weekdays and weekends. From this information we determined the proportion of participants who went to bed at 21:00 hours or earlier, between 21:01 and 22:00 hours, between 22:01 and 23:00 hours, between 23:01 and 24:00 hours, between 24:01 and 01:00 hours and after 01:00 hours. Information on timing (bedtime and wake-up time) is relevant, because duration on weekdays is dependent upon bedtime in a context where school start-times do not change. We also calculated each participant s duration for weekdays and weekends and averaged across all 7 days of the week. Participants were then classified as having a duration that met the recommended range (9 11 h per night for 6 13-year-olds or 8 10 h per night for year-olds), a duration that was shorter than the recommended range or a duration that was longer than the recommended range (Hirshkowitz et al., 2015). Any durations falling even 1 min outside these ranges were classified as not meeting the recommendations (e.g. for an 11-year-old, duration of 8 h 59 min was classified as short and duration of 11 h 1 min was classified as long ). Age and gender categories For descriptive and comparative purposes, participants were grouped into the following age categories: (n = 1090 boys, 1239 girls), (n = 4406 boys, 4776 girls), (n = 5360 boys, 5989 girls) and (n = 988 boys, 1048 girls). We also grouped participants into the following categories: school-aged children (10 13 ; n = 5496 boys, 6015 girls) and adolescents (14 17 ; n = 6348 boys, 7037 girls). Comparisons were also made across genders (n = boys, girls). Statistical analysis Analyses were conducted using SAS version 9.4 (SAS Institute, Cary, NC, USA), and accounted for the sample weights and clustered nature of the survey. Estimates of means and the prevalence of participants in different ranges were produced. Differences in group means were determined using the PROC SURVEYREG command and

3 Sleep duration of Canadian children and adolescents 543 CONTRAST option. Differences in prevalence estimates were determined using the PROC SURVEYFREQ command, which calculates the Rao Scott chi-square test. Where appropriate, Bonferonni corrections were made to account for multiple group comparisons. RESULTS The duration distribution curves for children and adolescents on weekdays and at weekends are illustrated in Fig. 1. The duration data follow a normal distribution and there are no issues with skewness or kurtosis. Average duration decreased with increasing age, with later bedtimes reported in the older age groups (Table 1). In contrast, wake-up time was relatively constant (within 15 min) across the age groups. Sleep duration was longer at weekends compared to weekdays across all age groups, a phenomenon generally referred to as catch-up. Although respondents went to bed ~1.5 h later at weekends, their wake-up time was ~2.5 h later (9:32 versus 6:59 h for wake-up time at weekends and on weekdays, respectively). Differences between boys and girls were minimal, particularly for bedtimes. In comparison to boys, girls woke up 9 min earlier on weekdays and woke up 8 min later at weekends. Thus, when averaged across all 7 days of the week, average duration in boys and girls, while statistically different (P < 0.001), was quite similar (9:01 versus 8:56 h, respectively). Table 2 presents the distribution of bedtimes across age groups and gender. Later bedtimes was a common observation in both genders as participants grew older. In adolescents aged 16 17, 11% went to bed after midnight on weekdays and 44% did so at weekends. In general, differences between boys and girls were minimal and few were statistically significant. Of note, 31% of boys and 24% of girls aged went to bed after 01:00 hours at weekends (P = 0.11 between genders). On a related note, 46% of boys and 37% of girls aged went to bed at 21:00 hours or earlier on weekdays (P = between genders). Overall, 70% of the participants met the duration recommendations, 28% were short ers and 2% were long ers, based on average daily duration (Table 3). In this cohort, 68% of school-aged children aged and 72% of adolescents aged slept the recommended amount per night, on average, Figure 1. Sleep duration distribution curves for children (n =11 511) and adolescents (n =13 385) on weekdays and at weekends.

4 544 J. -P. Chaput and I. Janssen across all days of the week. Among those not meeting the duration recommendations, 31% of school-aged children and 26% of adolescents were considered short ers. Very few Canadian children and adolescents were long ers based on average daily duration. Adolescents aged were the age group with the highest proportion of short ers on weekdays (43%). Across all age groups, short was significantly more prevalent on weekdays (36%) compared to weekends (17%). Long at weekends was also significantly more common than during weekdays (17 versus 2% for all ages between weekends and weekdays, respectively). The proportion of boys and girls having an average daily duration in the recommended range was not different (69.8 versus 70.2%, P > 0.2). However, on weekdays more boys than girls slept for the recommended amount, while at weekends more girls than boys slept for the recommended amount (P < 0.05). DISCUSSION The present study is the first to report, using a representative sample of Canadian children and adolescents, the proportion meeting contemporary duration recommendations. We observed that 70% of year-old Canadians meet the duration recommendations, 28% are short ers and 2% are long ers when based on average daily duration. Different patterns were observed when duration was examined separately on weekdays and at weekends where only 62 and 66%, respectively, had a duration in the recommended range. We also observed that Canadian children and adolescents have shorter durations as they grow older, which primarily reflects later bed times. In fact, almost half of adolescents aged go to bed after midnight at weekends. Finally, children and adolescents approximately 1 h longer at weekends compared to weekdays, despite going to bed approximately 1.5 h later. Collectively, these findings have important implications and can inform the development of policies and interventions aimed at promoting the benefits of having a good night s for overall health and wellbeing. First, this study highlights that approximately one-third of Canadian children and youth less than current duration recommendations. This proportion may be concerning, considering the well-known adverse effects of insufficient on physical and mental health (Owens, 2014; Shochat et al., 2014). It has been argued that should be taken more seriously by the public health community, i.e. given as much attention and resources as nutrition and physical activity (Chaput, 2014; Chaput et al., 2014). The descriptive data presented in this study support this argument, as a high prevalence of children and adolescents are not getting the recommended amount of. However, future studies will need to determine the proportion of these short ers reporting problems, tiredness and adverse health outcomes before concluding that ing less than the recommended amount is necessarily detrimental. Secondly, the findings of this study point to population subgroups and time-periods at Table 1 Mean bedtimes, wake-up times and durations within year-old children and youth Averaged across all days of the week Weekdays Weekends Age and gender group Bedtime Wake-up time Sleep duration Bedtime Wake-up time Sleep duration Bedtime Wake-up time Sleep duration Both genders :05 (3) 07:35 (2) 9:43 (2) 21:29 (2) 07:05 (2) 9:35 (2) 22:49 (4) 08:52 (3) 10:03 (3) :22 (2)* 07:38 (2) 9:18 (2)* 21:58 (2)* 06:58 (2)* 9:00 (2)* 23:23 (2)* 09:19 (2)* 9:55 (2) :05 (2)* 07:46 (2)* 8:41 (2)* 22:38 (2)* 06:57 (3) 8:18 (2)* 00:11 (2)* 09:48 (2)* 9:37 (2)* :19 (3)* 07:50 (3) 8:30 (3)* 22:55 (3)* 07:04 (4) 8:09 (4)* 00:22 (5)* 09:46 (5) 9:24 (4)* All ages 22:44 (2) 07:43 (2) 8:58 (2) 22:19 (2) 06:59 (2) 8:40 (2) 23:47 (3) 09:32 (2) 9:45 (1) Boys :48 (3) 07:31 (3) 9:43 (3) 21:25 (3) 07:05 (2) 9:40 (3) 22:46 (5) 08:36 (5) 9:51 (5) :20 (2)*, 07:38 (2) 9:18 (2)* 21:55 (2)*, 07:06 (3) 9:06 (2)*, 23:23 (2)*, 09:11 (3)*, 9:48 (2) :05 (2)* 07:50 (2)*, 8:45 (2)*, 22:39 (2)* 07:04 (2) 8:25 (3)*, 00:12 (3)* 09:46 (3)*, 9:34 (2)*, :23 (5)* 07:55 (4) 8:31 (4)* 22:57 (4)* 07:08 (5) 8:12 (5)* 00:30 (7)* 09:52 (7) 9:21 (5)* All ages 22:43 (3) 07:44 (2) 9:01 (2) 22:17 (2) 07:04 (2) 8:46 (2) 23:48 (3) 09:28 (3) 9:39 (2) Girls :05 (3) 07:39 (2) 9:43 (2) 21:33 (3) 07:04 (2) 9:31 (2) 22:53 (4) 09:07 (5) 10:14 (4) :25 (2)*, 07:39 (2) 9:15 (2)* 22:00 (2)*, 06:56 (2)*, 8:56 (3)*, 23:24 (3)*, 09:26 (3)*, 10:02 (2) *, :05 (3)* 07:41 (2) 8:37 (2)*, 22:38 (3)* 06:50 (3) 8:12 (3)*, 00:10 (3)* 09:50 (3)*, 9:39 (2)*, :16 (4)* 07:45 (3) 8:29 (4) 22:52 (4)* 06:59 (5) 8:07 (6) 00:14 (5) 09:41 (5) 9:27 (5) All ages 22:45 (2) 07:41 (2) 8:56 (2) 22:20 (2) 06:55 (2) 8:34 (3) 23:47 (3) 09:36 (2) 9:49 (2) Data presented as mean in h:min (standard error in minutes). Bedtimes and wake-up times are based on a 24-h time format. *Significantly different from estimate for next younger age group within same gender (adjusted for multiple comparisons). Significantly different from estimate for other gender within same age group.

5 Sleep duration of Canadian children and adolescents 545 Table 2 Proportion of year-old children and youth going to bed at different times Age and gender group Weekday bedtimes 21:00 or 21:01 earlier 22:00 22:01 23:00 23:01 24:00 00:01 01:00 After 01:00 Weekend bedtimes 21:00 or 21:01 earlier 22:00 22:01 23:00 23:01 24:00 00:01 01:00 After 01:00 Both genders (2.3) 43.2 (1.7) 12.0 (1.1) 2.9 (0.7) 0.5 (0.2) 0.2 (0.1) 10.6 (1.1) 31.6 (1.6) 27.0 (1.4) 19.3 (1.3) 6.3 (0.8) 5.2 (0.7) (1.8)* 45.2 (0.9) 22.6 (0.9)* 7.0 (0.4)* 1.8 (0.2)* 0.7 (0.1) 5.1 (0.4)* 19.8 (0.8)* 29.1 (0.9) 24.4 (0.9)* 10.9 (0.7)* 10.7 (0.5)* 7.8 (0.6)* 31.5 (1.2)* 37.6 (0.8)* 16.3 (0.8)* 4.8 (0.4)* 2.0 (0.2)* 2.0 (0.2)* 8.7 (0.7)* 21.7 (0.9)* 27.5 (0.5)* 17.4 (0.8)* 22.6 (0.9)* 6.9 (1.1) 21.8 (1.8)* 39.6 (1.9) 20.7 (1.6)* 8.1 (0.9)* 2.9 (0.5) 2.5 (0.5) 7.8 (1.1) 19.0 (1.6) 26.2 (1.3) 17.1 (1.3) 27.4 (1.7)* All ages 16.3 (1.0) 36.6 (1.0) 29.9 (1.0) 12.1 (0.6) 3.6 (0.3) 1.5 (0.1) 4.0 (0.3) 14.8 (0.7) 24.6 (0.7) 25.5 (0.5) 14.0 (0.5) 17.1 (0.7) Boys (3.1) 38.2 (2.3) 12.9 (1.6) 2.3 (0.6) 0.5 (0.3) 0.0 (0.0) 11.5 (1.8) 33.9 (2.5) 25.8 (1.1) 17.8 (1.7) 5.9 (1.2) 5.1 (1.1) (1.3)* 46.6 (1.1)*, 21.9 (1.6)* 5.9 (0.5)*, 1.1 (0.2) 0.9 (0.2)* 5.4 (0.5)* 19.5 (0.9)* 29.5 (1.1) 24.1 (1.1)* 10.8 (0.7)* 10.9 (0.7)* 7.5 (0.7)* 32.7 (1.3)* 36.9 (1.0)* 15.9 (0.9)* 4.8 (0.4)* 2.2 (0.3)* 2.5 (0.4)* 9.2 (0.9)* 21.5 (1.2)* 26.2 (1.0) 16.8 (0.9)* 23.8 (1.1)* 7.2 (1.3) 20.2 (2.5)* 38.6 (2.5) 22.0 (1.8)* 8.3 (1.3)* 3.7 (1.0) 3.3 (0.9) 6.5 (1.2) 17.4 (2.2) 24.3 (2.1) 17.6 (2.4) 30.9 (3.4)* All ages 17.0 (1.1) 37.0 (1.1) 29.3 (1.0) 11.5 (0.7) 3.4 (0.3) 1.7 (0.2) 4.5 (0.4) 15.0 (0.8) 24.4 (0.9) 24.4 (0.8) 13.7 (0.6) 18.0 (0.9) Girls (2.4) 47.7 (2.2) 11.1 (1.3) 3.5 (1.0) 0.6 (0.3) 0.4 (0.3) 9.7 (1.3) 29.4 (2.3) 28.0 (1.7) 20.8 (1.8) 6.6 (1.1) 5.4 (0.9) 21.8 (1.3)* 43.9 (1.0)*, 23.2 (1.2)* 8.3 (0.6)*, 2.3 (0.3) 0.5 (0.2)* 4.8 (0.6)* 20.2 (1.0)* 28.7 (0.9) 24.7 (1.2)* 11.1 (0.8)* 10.5 (0.7)* 8.1 (0.7)* 30.3 (1.5)* 38.3 (1.1)* 16.6 (1.1)* 4.8 (0.5)* 1.8 (0.3)* 1.6 (0.3)* 8.3 (0.7)* 21.9 (1.1)* 28.6 (0.8) 17.9 (1.0)* 21.6 (1.0)* 6.6 (1.4) 23.2 (2.3)* 40.5 (2.7) 19.5 (2.3)* 7.9 (1.1)* 2.2 (0.7) 1.8 (0.6) 8.9 (1.4) 20.4 (1.4) 27.9 (2.2) 16.7 (2.0) 24.2 (2.0) All ages 15.6 (1.0) 36.2 (1.1) 30.5 (1.1) 12.6 (0.7) 3.8 (0.4) 1.3 (0.2) 3.6 (0.3) 14.7 (0.9) 24.8 (0.7) 26.5 (0.7) 14.2 (0.7) 16.3 (0.8) Data presented as % (standard error). *Significantly different from estimate for next younger age group within same gender (adjusted for multiple comparisons). Significantly different from estimate for other gender within same age group. particularly high risk. For example, targeting the hygiene of year-olds on weekdays appears to be particularly relevant, as this is the age group and time of the week when is most limited. The evidence strongly implicates early school start-times as a key modifiable contributor to insufficient (Adolescent Sleep Working Group, Committee on Adolescence and Council on School Health, 2014; Minges and Redeker, 2015). An accumulating body of evidence has demonstrated that delaying school start-times is an effective countermeasure to chronic deprivation and has a wide range of potential benefits to students with regard to physical and mental health, safety and academic achievement (Adolescent Sleep Working Group, Committee on Adolescence and Council on School Health, 2014; Minges and Redeker, 2015). Other contributing and potentially remediable factors to consider that could lead potentially to earlier bedtimes include excessive demands on students time because of homework, extracurricular activities, after-school employment, social activities, electronic media use and lack of parental monitoring or rules about bedtimes. The findings observed in the present study are in line with previous investigations. For example, population-based estimates indicate that approximately one-fourth to one-third of American adolescents have insufficient (Smaldone et al., 2007), which is similar to the 28% observed in our study. A study of adolescents from Austria, Belgium, France, Germany, Greece, Hungary, Italy, Spain and Sweden reported an average duration of approximately 8 h per night (Garaulet et al., 2011); in our study the average duration was 8:30 in year-olds. Previous studies also reported that children and youth less with age, with attempts to address the accumulated weekday debt during the weekend (Keyes et al., 2015; Leger et al., 2012; Matricciani et al., 2012b; Owens, 2014). Also, the fact that children and adolescents for approximately 1 h longer at weekends compared to weekdays in the present study is inconsistent with recommendations suggesting that having

6 546 J. -P. Chaput and I. Janssen Table 3 Proportion of year-old children and youth in different duration categories Age and gender group Sleep duration averaged across all days of the week Weekday duration Weekend duration Recommended Short Long Recommended Short Long Recommended Short Long Both genders (1.2) 17.8 (1.2) 1.2 (0.3) 79.8 (1.1) 17.9 (1.1) 2.3 (0.4) 68.0 (1.5) 17.8 (1.2) 14.1 (1.1) (1.1)* 34.9 (1.1)* 0.9 (0.1) 58.7 (1.6)* 39.8 (1.6)* 1.6 (0.2) 67.8 (0.8) 20.2 (0.7) 12.1 (0.5) (1.0)* 24.6 (1.0)* 2.1 (0.2) 62.2 (1.4) 36.0 (1.5) 1.8 (0.2) 63.8 (0.8)* 14.6 (0.6)* 21.5 (0.6)* (2.4)* 31.6 (2.3)* 2.6 (0.5)* 53.8 (2.4)* 43.1 (2.5)* 3.1 (0.7)* 63.6 (2.4) 17.6 (1.8) 18.8 (1.5) All ages 70.0 (0.8) 28.4 (0.9) 1.6 (0.2) 61.8 (1.1) 36.3 (1.2) 1.9 (0.2) 65.6 (0.5) 17.2 (0.4) 17.1 (0.5) Boys (2.0) 19.8 (1.9) 2.3 (0.6) 77.7 (1.5) 18.5 (1.4) 3.8 (0.8) 63.5 (2.4) 22.2 (2.0) 14.2 (1.4) (1.2)* 34.0 (1.2)* 1.0 (0.2)*, 61.6 (1.8)*, 36.5 (1.8) 2.0 (0.3)*, 64.1 (1.1) 24.0 (1.1) 12.0 (0.7) (1.1)* 23.3 (1.0)* 3.1 (0.4) 64.7 (1.3) 32.4 (1.4) 2.8 (0.4) 63.7 (1.2) 15.5 (0.8)* 20.8 (0.8)* (3.3)* 33.2 (2.9)* 4.1 (1.0) 53.0 (3.4)* 42.5 (3.2)* 4.5 (1.0) 64.0 (2.4) 18.6 (1.8) 17.4 (1.7) All ages 69.8 (0.9) 27.8 (0.9) 2.4 (0.3) 63.6 (1.2) 33.6 (1.2) 2.8 (0.3) 63.8 (0.7) 19.5 (0.6) 16.6 (0.5) Girls (1.4) 16.0 (1.4) 0.3 (0.2) 81.7 (1.6) 17.4 (1.4) 0.9 (0.4) 72.1 (1.9) 13.8 (1.3) 14.0 (1.5) (1.4)* 35.7 (1.4)* 0.8 (0.2)*, 56.1 (1.8) 42.7 (1.8)*, 1.2 (0.2)*, 71.1 (1.0) 16.8 (0.8) 12.1 (0.7) (1.2)* 25.8 (1.3)* 1.1 (0.3) 60.0 (1.9) 39.1 (1.9) 0.9 (0.2) 63.9 (1.0) 13.9 (0.7)* 22.2 (0.9)* (2.6)* 30.1 (2.7)* 1.2 (0.5) 54.5 (2.7)* 43.6 (3.1)* 1.9 (0.8) 63.3 (3.3) 16.6 (2.3) 20.1 (2.1) All ages 70.2 (1.0) 28.9 (1.1) 0.9 (0.2) 60.1 (1.4) 38.8 (1.4) 1.1 (0.1) 67.2 (0.7) 15.2 (0.5) 17.6 (0.6) Data presented as % (standard error). *Significantly different from estimate for next younger age group within same gender (adjusted for multiple comparisons). Significantly different from estimate for other gender within same age group. consistent bedtimes and wake-up times are preferable for overall health (Gruber et al., 2014). An important observation of population-based research in this field of research is the lack of use of objective measures for duration. Although polysomnography is considered the gold standard technique in laboratory experiments, actigraphy is gaining popularity and recognition for the assessment of in epidemiological research (Meltzer et al., 2012). While actigraphy generally provides good estimates of duration (Sadeh, 2011), questions used typically in epidemiological studies overestimate duration compared with objective measures and can introduce inaccuracies (Girschik et al., 2012). However, current duration recommendations are based largely on selfreported data (Hirshkowitz et al., 2015). It would thus be misleading to use an objective assessment of duration and compare the estimates to the duration recommendations; it would result in an overestimation of short ers. This methodological aspect can have profound implications if future studies rely more heavily upon objective assessments of. However, if parents and their children/adolescents are key targets for the duration recommendations, it is more likely that they will use time in bed as their estimation of duration, and in this case self-reported should be used. It is also important to keep in mind that there is interindividual variability in needs (e.g. genetic differences or sociocultural contexts) and ing longer or shorter than the recommended times may not necessarily mean that it will affect health adversely (Gruber et al., 2014). While the duration recommendations inform health professionals and the general population of the healthy duration ranges, they may need to be adapted on a case-by-case basis. However, individuals with durations far outside the normal range may be engaging in behavioural restriction or may have other health problems. Intentionally restricting duration over a prolonged period of time may compromise overall health (Hirshkowitz et al., 2015).

7 Sleep duration of Canadian children and adolescents 547 Also, optimal reaches well beyond the notion of quantity and includes other dimensions such as quality, timing, architecture, consistency and continuity. The present study has several strengths and limitations that warrant discussion. An important strength is the large and representative sample. Furthermore, to our knowledge this is the first population-based study reporting on the duration estimates of Canadian children and adolescents, an important public health issue deserving greater attention and surveillance (Gruber et al., 2014; Owens, 2014). However, our data need to be interpreted in light of the following limitations. First, these data do not provide information about secular trends in the time of children and adolescents because the questionnaire items used to determine duration were only added to the last cycle of the HBSC. The duration items were only included in the Canadian survey, so we could not examine duration in the other 43 countries that are part of the HBSC network. Secondly, we need to keep in mind that self-reported duration estimates overestimate actual duration (Girschik et al., 2012) and are subject to recall and/or social desirability bias; however, they are well suited for surveillance, and recommendations rely heavily on self-reported data. Thirdly, duration represents only one aspect of hygiene, and other aspects (e.g. quality and timing) are important for overall health. Fourthly, although our analyses were split by age and gender to provide an overview of duration estimates by these factors, they did not characterize the sample further according to race/ethnicity, socioeconomic status, rural versus urban areas, etc. Finally, a selection bias may have occurred, as students who did not provide consent or who were absent from school on the day of the survey may have been systematically different from those who participated. In conclusion, 28% of Canadian children and adolescents are short ers, while 70% meet the duration recommendations of 9 11 h per night for school-aged children and 8 10 h per night for adolescents. Future research should examine longitudinal trajectories of duration in the paediatric Canadian population to look at patterns over time and identify subgroups of the population more likely to experience declines in habitual duration. Future studies should also determine the proportion of short ers who report daytime tiredness and adverse health problems. ACKNOWLEDGEMENTS IJ is supported by a Tier 2 Canada Research Chair Award. The Public Health Agency of Canada and Health Canada funded the Health Behaviour in School-aged Children study in Canada. The Canadian principal investigators of the HBSC study are Drs John Freeman and William Pickett and its national coordinator is Matthew King. The international coordinator of the HBSC is Dr Candace Currie at the University of St Andrews, Scotland. AUTHOR CONTRIBUTIONS JPC and IJ designed the study; IJ conducted the analyses; JPC and IJ interpreted the data; JPC wrote the manuscript. Both authors approved the final version of the manuscript. CONFLICT OF INTEREST The authors declared no conflicts of interest. REFERENCES Adolescent Sleep Working Group, Committee on Adolescence and Council on School Health. School start times for adolescents. Pediatrics, 2014, 134: Chaput, J. P. Sleep patterns, diet quality and energy balance. Physiol. Behav., 2014, 134: Chaput, J. P., Carson, V., Gray, C. E. and Tremblay, M. S. Importance of all movement behaviors in a 24 hour period for overall health. Int. J. Environ. Res. Public Health, 2014, 11: Chaput, J. P., Gray, C. E., Poitras, V. J. et al. Systematic review of the relationships between duration and health indicators in school-aged children and youth. Appl. Physiol. Nutr. Metab., in press. Currie, C., Nic Gabhainn, S. and Godeau, E. The Health Behaviour in School-aged Children: WHO Collaborative Cross-National (HBSC) Study: origins, concept, history and development Int. J. Public Health, 2009, 54(Suppl. 2): Freeman, J. G., King, M. and Pickett, W. (Eds) The Health of Canada s Young People: A Mental Health Focus. Public Health Agency of Canada, Ottawa, ON, Garaulet, M., Ortega, F. B., Ruiz, J. R. et al. Short duration is associated with increased obesity markers in European adolescents: effect of physical activity and dietary habits. The HELENA study. Int. J. Obes. (Lond.), 2011, 35: Girschik, J., Fritschi, L., Heyworth, J. and Waters, F. Validation of selfreported against actigraphy. J. Epidemiol., 2012, 22: Gruber, R., Carrey, N., Weiss, S. K. et al. Position statement on pediatric for psychiatrists. J. Can. Acad. Child Adolesc. Psychiatry, 2014, 23: Hirshkowitz, M., Whiton, K., Albert, S. M. et al. National Sleep Foundation s time duration recommendations: methodology and results summary. Sleep Health, 2015, 1: Keyes, K. M., Maslowsky, J., Hamilton, A. and Schulenberg, J. The Great Sleep Recession: changes in duration among US adolescents, Pediatrics, 2015, 135: Kronholm, E., Puusniekka, R., Jokela, J. et al. Trends in self-reported problems, tiredness and related school performance among Finnish adolescents from 1984 to J. Sleep Res., 2015, 24: Leger, D., Beck, F., Richard, J. B. and Godeau, E. Total time severely drops during adolescence. PLoS ONE, 2012, 7: e Matricciani, L., Olds, T. S., Blunden, S., Rigney, G. and Williams, M. T. Never enough : a brief history of recommendations for children. Pediatrics, 2012a, 129: Matricciani, L., Olds, T. and Petkov, J. In search of lost : secular trends in the time of school-aged children and adolescents. Sleep Med. Rev., 2012b, 16: Matricciani, L., Blunden, S., Rigney, G., Williams, M. T. and Olds, T. S. Children s needs: is there sufficient evidence to recommend optimal for children? Sleep, 2013, 36: Meltzer, L. J., Montgomery-Downs, H. E., Insana, S. P. and Walsh, C. M. Use of actigraphy for assessment in pediatric research. Sleep Med. Rev., 2012, 16:

8 548 J. -P. Chaput and I. Janssen Minges, K. E. and Redeker, N. S. Delayed school start times and adolescent : a systematic review of the experimental evidence. Sleep Med. Rev., 2015, 28: Ohayon, M. M. Determining the level of iness in the American population and its correlates. J. Psychiatr. Res., 2012, 46: Owens, J.; Adolescent Sleep Working Group and Committee on Adolescence. Insufficient in adolescents and young adults: an update on causes and consequences. Pediatrics, 2014, 134: e921 e932. Sadeh, A. The role and validity of actigraphy in medicine: an update. Sleep Med. Rev., 2011, 15: Shochat, T., Cohen-Zion, M. and Tzischinsky, O. Functional consequences of inadequate in adolescents: a systematic review. Sleep Med. Rev., 2014, 18: Smaldone, A., Honig, J. C. and Byrne, M. W. Sleepless in America: inadequate and relationships to health and well-being of our nation s children. Pediatrics, 2007, 119: S29 S37.