Background
Lack of sleep has become increasingly recognized as a contributor to the obesity epidemic in children and adolescents [
1]. Elevated rates of childhood obesity and overweight in western countries within the last decade [
2‐
4] have garnered substantial attention from public health, with a particular emphasis on promoting healthy active living, and limiting sedentary behaviour. While both sedentary behaviour and sleep involve low energy expenditure, they have opposite effects on children’s weight [
5]. Sleep is an essential component of healthy cognitive and physical development, and lack thereof can negatively impact children’s physical activity levels as well as hormones associated with increased risks of obesity, diabetes and hypertension [
1,
6‐
8]. Even after controlling for other known risk factors, inadequate sleep significantly increases the risk of pediatric obesity [
9,
10]. Shorter sleep duration has also been shown to be associated with poorer emotional regulation, academic achievement, and lower quality of life/well-being [
11].
Child sleep and family functioning (e.g., marital conflict; parenting stress) are inextricably linked [
12‐
14], and thus, it is imperative to view sleep from a family context [
15‐
17]. Indeed, behavioural interventions in toddlers and preschoolers have proven effective for improving both child sleep and family functioning [
18,
19]. It follows that parents’ behaviours play a key role in influencing children’s sleep patterns. Although the types of behaviours that support adequate sleep may be more limited than those for other health behaviours, type of support has been shown to be important for achieving child sleep outcomes. For example, periodically checking on the child has been associated with longer sleep duration in 11-year-olds [
20]; conversely, active physical comforting of one- and two-year-old children has been associated with children’s sleep problems (e.g., resisting bedtime, waking at night) [
15]. This highlights the need to clearly discern which parental support behaviours for which age groups are optimal for child sleep.
Previous research has found that parents can actively encourage their child’s development of self-regulation [
21], and this encouragement of autonomy and independence has been linked to fewer sleep problems and more consolidated child sleep [
22,
23]. Parents can also support their child’s sleep through enforcing rules, such as setting bedtimes and structuring the child’s evening time [
14]. Although sleep needs and challenges differ greatly across children’s ages, establishment of a bedtime routine and development of a structured environment (e.g., regular family mealtimes) may improve sleep for both young children and adolescents [
14,
24,
25]. Moreover, sleep quality and duration may be negatively impacted by media use, with a particular threat being screen time in the bedroom [
7,
26,
27]. Parents’ restriction or discouragement of nighttime media use has the potential to foster healthy child sleep behaviours [
26,
28], although determining the best means of doing so remains elusive.
Children’s sleep duration has decreased by about 30 to 60 min in recent decades [
29,
30], and in the Canadian context, recent estimates reveal that 31% of school-aged kids and 26% of adolescents are sleep-deprived [
31]. In acknowledgement of the emerging “sleepidemic,” the 2016 ParticipACTION Report Card on Physical Activity for Children and Youth brought sleep to the forefront of discussion of inactivity in Canadian children [
32]. This most recent report also promotes the new
Canadian 24-Hour Movement Guidelines for Children and Youth [
33], which outlines what a healthy 24-h period should look like for 5- to 17-year-olds. The guidelines introduce a new movement paradigm—of which sleep is an integral component—emphasizing a whole day approach. It presents sleep, not only as an important behaviour in its own right, but also as highly integrated with the other movement behaviours of physical activity and sedentary behaviour. While there is mounting evidence of the importance of sleep as a protective factor for childhood overweight and obesity, there remains uncertainty surrounding the optimal role of parental support for child sleep [
15].
Objectives
The objective of this study is to determine the relative contribution of different types of parental support behaviours for predicting the likelihood that children meet sleep guidelines on weekdays and weekends.
Discussion
A major finding of the current study is that only parental support in the form of weekday bedtime rule enforcement contributed significantly to children meeting sleep guidelines; however, findings surrounding child age and weekday/weekend differences also warrant further discussion. Our results suggest that the proportion of children meeting Canadian sleep guidelines decreases as they enter and progress through adolescence. This discovery is consistent with contemporary sleep duration estimates by Chaput and Janssen [
31], which reveal shorter sleep durations in Canadian children and adolescents as they grow older. The sleep duration trend may be an indication of later bed times, rather than earlier wake up times. For instance, almost half of 16- to 17-year-olds reported going to bed after midnight on weekends [
31]). A number of factors contributing to insufficient sleep in adolescents have been identified, including: (1) biological processes (i.e., an evening-type circadian phase; delayed melatonin production), (2) social obligations (i.e., early school start times), (3) consumption of caffeine (i.e., soda, coffee, and energy drinks), and (4) media use (i.e., direct displacement of sleep; disruption of circadian rhythms by light; or increased sleep-disrupting arousal) [
36].
The present study compared the proportion of children meeting sleep guidelines on weekdays and weekends, and found the greatest difference in 15-year-olds. Interestingly, results from a 2014 multicohort study of over 270,000 American adolescents observed that over the previous 20 years, the largest decrease in the proportion getting ≥7 h of sleep was also in 15-year-olds [
29]. As a potential high-risk group, the biological and social factors at play for this particular age may warrant further elucidation.
The larger discrepancies between weekday and weekend sleep habits seen from middle childhood through adolescence echo the work of others [
7]. It is well established that adolescents accumulate sleep debt from regularly going to bed later during the week, which may lead to later wake times on weekends (also known as catch-up sleep) [
29‐
31,
36,
37]. Yet, what appears inconsistent is whether this sleep debt leads to longer [
11] or shorter (as found in the present study) weekend sleep duration in adolescents.
The proportion of parents engaging in support behaviours for sleep was lower for the two
regulatory support behaviours, relative to the two
motivational support behaviours. Similarly, the largest decline in parental support for adolescents of increasing age was seen for the two
regulatory support behaviours. As children get older, parents are usually less involved with their sleep routines [
17]. Nevertheless, parental rules—including an earlier bedtime—have been shown to increase adolescent total sleep [
14], highlighting the need for increased
regulatory parental support.
In the weekday model, two parental support behaviours contributed significantly to predictions of children meeting sleep guidelines. First, parents who reported encouraging their child to go to bed at a specific time were less likely to have children meeting sleep guidelines on weekdays. Given that
motivational support behaviours are defined by “prompting others to engage in the behaviour of interest” [
35], it was unexpected to find that encouragement was a negative predictor of sufficient sleep in children. Because of this study’s cross-sectional design, we cannot discern the direction of the relationship between parental support and child sleep. Accordingly, one possible explanation for this finding is that “encouraging one’s child to go to bed at a specific time” is a parent’s reaction to their child’s already-established poor sleep habits. In other words, encouragement may lend itself to being a reactive behaviour, while other behaviours—such as rule-setting—may be more proactive. Given the lack of research on parental support for child and adolescent sleep, the proposed explanations would require further investigation in order to be substantiated.
Second, parents who reported enforcing rules about their child’s bedtime were more likely to have their child meeting sleep guidelines on weekdays. Family rules have been shown to be associated with earlier bedtimes and greater total hours of sleep on weekdays for older children (12 to 19 years old) [
14]. One reason for this association proposed by Adam, Snell, & Pendry [
14] is that parents more strictly enforce rules regarding bedtimes during the week; however, their results also suggest that it is not merely stricter bedtime rules—but rather more general parental expectations, structure, or monitoring—that affect child sleep on weekdays. In a meta-analysis of risk and protective factors associated with adolescent sleep, parent-set bedtime had the greatest positive correlation with sleep duration [
38]. However, there is evident dissonance between parental bedtime rule-setting as an important determinant of adolescent sleep, and as a parent support behaviour reported less frequently as children enter adolescence. The present study includes a range of child ages—not one particular age—when looking at the effect of bedtime rule enforcement on child sleep. Age-specific investigations are warranted to determine if these findings hold true in different age groups. Future studies should also consider which specific rules parents can set and how parents can enforce these rules to best support child sleep.
In the weekend model, none of the parental support behaviours contributed significantly to predictions of children meeting sleep guidelines. While non-significant results for
all support behaviours was unexpected, this finding parallels that of prior research showing family variables as better predictors of weekday, as oppose to weekend, sleep behaviours [
14]. A potential explanation for these results is that parents exercise less control over children’s schedules on weekends, giving children more discretionary time to choose what activities they engage in [
39]. Time spent watching television, using the computer or video games, socializing, playing sports, working, and participating in religious activities all predict fewer hours of sleep on weekends [
14]. Our results do not rule out parental support behaviours as important determinants of child sleep on weekends; instead, they point to the need to explore the child’s social and household environments to gain a full picture of the many factors affecting weekend sleep behaviours.
A noteworthy finding of the current study is that, while the use (and even presence) of screens in the bedroom can negatively affect children’s sleep [
40,
41], parental rule enforcement and encouragement to limit screens in the bedroom did not predict children meeting sleep guidelines on weekdays or weekends. One possible explanation is that use of electronic media or “screen time” is an unstructured leisure activity, with no clear starting and ending point [
27]. Use of screens in the bedroom may represent a child behaviour over which parents have limited control [
26]. Another contributing factor is that restricting media may make it all the more enticing [
42], and the child’s bedroom may be a location where screen use is less supervised. Finally, it is important to acknowledge the role of not only a parent, but the entire family, in the creation and perpetuation of a child’s sleep-displacing screen use [
26]. Our findings draw attention to the notion that traditional parenting behaviours (rule-setting, encouragement) may not be sufficient when it comes to changing screen time behaviours in the bedroom; the sleep and screen use behaviours and expectations of the entire family should considered in future studies.
Limitations
The present study has several limitations that warrant discussion. First, all information was obtained from survey data, and therefore, measures of parental support and child sleep were parent-reported. Both measures are therefore subject to recall bias, as well as social desirability bias stemming from the desire to appear as a supportive parent. In particular, child sleep duration may be overestimated if parents’ responses were based on when their child was
in bed, as opposed to when their child was actually
asleep. Despite the drawbacks of using subjective versus objective measures, current sleep duration recommendations are based largely on self-reported sleep data [
43]. A second, related limitation is the study’s use of sleep duration as measure of children’s sleep health. While duration is an important component of sleep hygiene, other aspects including sleep quality, consistency, continuity, and timing play an important part in one’s overall health and well-being [
32]. The
Canadian 24-Hour Movement Guidelines for Children and Youth specify “uninterrupted” sleep; however, because sleep continuity could not be measured, the use of sleep duration data allowed for clear comparisons with these sleep recommendations in order to construct the primary outcome variable. It should still be noted that because this study’s measure of child sleep is based on parent report, we cannot know whether the hours of sleep are fragmented or truly uninterrupted. Third, parents were asked the extent to which they agreed with statements starting with “I encourage…” and “I enforce rules…” (Table
2) without being provided with definitions of encouragement and rule enforcement. Thus, parents’ perceptions of these words informed the measures of parental support for child sleep. Fourth, while the logistic regression models of the present study controlled for factors such as time since immigration, education, employment status, and total household income, other cultural factors were not included. Although it was beyond the scope of this study to ascertain cultural backgrounds and implications for sleep habits, it is important to acknowledge that cross-cultural variation exists for sleep timing and duration [
44]. Fifth, variables exploring the whole day of the child (e.g., physical activity and sedentary time) were not able to be included in the analyses. Due to the length of the survey, parents were asked to answer at least two of the four health behaviour modules (physical activity, healthy eating, recreational screen time, and sleep). These were randomly selected at the time of the interview. Participants were also invited to complete additional, optional modules at the end of the survey. Given that the sample of parents responding to the sleep module was not the same sample that responded to the physical activity module, for instance, analyses could only be performed on variables within the same module, not across modules. Sixth, the adapted framework for classifying parental support behaviours [
34] was developed after the survey questions were designed. Accordingly, only
motivational and
regulatory support behaviours were measured and analyzed in the current study. Future studies should consider also including
instrumental and
conditional support behaviours to better understand the relationship between parental support and child sleep. Finally, the complex relationship between parental support behaviours, child sleep, and child age is important to consider in future studies. This study does not attempt to form conclusions about the role of child age; for example, we cannot assert that the influence of parental support on child sleep varies across ages. More research on this relationship, with a main focus on child age, is warranted.
Acknowledgements
The authors would like to thank Jocelyn Jarvis and Karen Deng for their contribution to this research. They would like to acknowledge Public Health Ontario’s Research & Ethics Services for their support throughout this study’s development, and Library Services for their invaluable contribution to the literature search conducted for this study. The authors would also like to acknowledge Dr. Mark Tremblay (chair) and the members of the Healthy Kids Community Challenge Scientific Reference Committee (SRC) for their ongoing scientific guidance.