Positive childhood experiences and insufficient sleep among children and adolescents in the United States: A nationally representative study

Abstract

Objectives:

Positive childhood experiences (PCEs), defined as positive social experiences and environments which foster healthy child development, are associated with better mental and physical health outcomes and moderate the negative effects of Adverse Childhood Experiences (ACEs). However, knowledge of the associations between PCEs and childhood sleep deficiency is limited. We conducted cross-sectional analyses of the 2022–2023 National Survey of Children’s Health (NSCH) to evaluate associations between PCEs and insufficient sleep (defined as sleeping the recommended number of hours for age each weeknight).

Methods:

Parents of children ages 6–17 years old (n = 56,540) reported on their child’s exposure to 7 PCEs, including having a resilient and connected family, having an adult mentor, participating in after school activities or volunteering in the community, and living in a supportive and safe neighborhood. Associations between PCEs and insufficient sleep were evaluated using weighted, multivariate logistic regression analyses adjusted for sociodemographic factors.

Results:

We found that PCEs had dose-dependent associations with insufficient sleep; children and adolescents exposed to higher numbers of PCEs (5–7 PCEs) had the lowest reported rate of insufficient sleep (29.5%), while those exposed to 2 or fewer PCEs had the highest rate of insufficient sleep (48.0%). Adjusted analysis confirmed that youth experiencing 5–7 PCEs had significantly lower odds of insufficient sleep relative to youth experiencing only 0–2 PCEs (adjusted odds ratio (aOR): 0.57, 95% confidence interval (CI): 0.50–0.68, p < .0001). PCEs moderated associations between ACEs and insufficient sleep.

Conclusions:

In conclusion, youth with greater PCEs exposure had lower prevalence rates of insufficient sleep. Furthermore, PCEs were associated with reduced prevalence of insufficient sleep among youth exposed to ACEs.

Introduction

Insufficient sleep in children and adolescents is a common problem, affecting approximately 35.5% of youth in the United States.¹ This puts youth at increased risk for a multitude of adverse health outcomes, including a greater prevalence of obesity, worse mental health, and higher rates of persistent and chronic pain.² Existing literature has identified adverse childhood experiences (ACEs) as being associated with poorer health outcomes, including sleep deficiency. ACEs include potentially traumatic events that a child experiences, witnesses, or which may negatively impact their sense of well-being including abuse, neglect, and household challenges.³ In a study by Lin et al. examining the 2016–2017 National Survey of Children’s Health (NSCH), ACEs appear to have a cumulative detrimental effect on sleep duration, with youth who reported 2 or more ACEs experiencing the highest rates of insufficient sleep.⁴ This link between ACEs and sleep in children and adolescents invites an exploration of what interventions could effectively temper the observed negative outcomes. In particular, positive childhood experiences (PCEs) have been identified as potential protective factors against poor sleep and as moderators of the negative health impact of ACEs on sleep.

PCEs, also referred to as Benevolent Childhood Experiences, refer to those experiences which foster successful child development including having safe, stable, and nurturing relationships and environments.⁵ PCEs include family factors, namely feeling connected to and supported by one’s family, community factors, including having an adult mentor or supportive neighborhood, and larger societal factors, such as access to after school opportunities. Most of the existing literature surrounding PCEs examine the relationship between positive experiences and mental health outcomes across the lifespan.^6–9 To our knowledge, only 2 prior studies have examined the association between PCEs and insomnia in Southeast China and Taiwan, finding PCEs had a significant inverse relationship with persistent insomnia. Furthermore, a high number of PCEs mitigated the negative effect of ACEs on persistent insomnia and a high number of ACEs decreased the protective effect of PCEs.^10,11 However, it remains unclear whether this relationship extends to sleep duration more broadly and applies to other populations of youth. Given high rates of ACEs among youth in the United States, it is important to understand whether PCEs could help mitigate the effects of ACEs on sleep deficiency.

The primary aim of this study was to determine whether exposure to PCEs is associated with lower prevalence rates of insufficient sleep in a nationally representative sample of youth between 6 and 17 years of age in the United States. We hypothesized that exposure to PCEs would be associated with decreased likelihood of insufficient sleep after controlling for sociodemographic characteristics. This is the first study, to our knowledge, that examines associations between PCEs and insufficient sleep in a national sample of children and adolescents. The secondary aim was to determine whether PCEs were associated with a reduction in the well-known risk of insufficient sleep among youth exposed to ACEs by evaluating associations between PCEs and insufficient sleep stratified across different levels of ACEs exposure. We hypothesized that the presence of PCEs reduces associations between ACEs and insufficient sleep.

Participants and methods

Population and data

This cross-sectional investigation utilized data from the 2022–2023 National Survey of Children’s Health (NSCH), accessible at https://www.childhealthdata.org. The NSCH, a comprehensive national survey sponsored by the Maternal and Child Health Bureau and conducted by the US Census Bureau, compiles national-level data of the physical and mental health status of children aged 0–17 years within the United States, thereby ensuring that the study cohort accurately represents the broader pediatric population of the United States. Households in the United States were randomly selected and contacted by mail to identify households with children 0–17 years of age. From each identified household, 1 child was selected at random to be the focal point of the survey, with the primary data being furnished by the parents or guardians. The initial participant pool comprised 109,265 individuals, however we excluded children under the age of 6 years (n = 41,265) as the survey queries pertinent to Positive Childhood Experiences (PCEs) were exclusively directed to caregivers of children aged 6 years and older. In addition, we excluded participants based on the absence of data regarding the variables of interest (n = 11,460), culminating in a final analytic sample size of 56,540 participants. Informed consent was procured from all participants. Given that the NSCH data are publicly accessible and de-identified, the Institutional Review Board at Stanford University deemed this study exempt from review. We adhered to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines for cross-sectional studies.

Insufficient sleep duration

Youth were classified as having insufficient sleep duration based on the recommendation from the American Academy of Sleep Medicine that children 6–12 years of age should regularly sleep 9–12 h a day and children 13–18 should sleep 8–10 h a day for optimal health.¹² The number of hours each child slept on an average weeknight in the past week was reported by parents of youth in the study. Children 6–12 years of age whose parents reported they slept less than 9 h on an average weekday were categorized as having insufficient sleep. Teens 13–17 years of age whose parents reported they slept less than 8 h on an average weekday were categorized as having insufficient sleep. Existing research has demonstrated high agreement between child and maternal reported total sleep duration, even among teens and adolescents.¹³ This methodology has previously been used to examine insufficient sleep using data from the 2016–2017 NSCH.⁴

Positive childhood experiences

We used 7 NSCH items to assess exposure to PCEs during the child or adolescent’s lifetime, aligning our methodology with prior research.^14–19 PCEs included the child (1) volunteering in the community, (2) participating in after-school activities, (3) having an adult mentor other than their parent or caregiver, (4) being a member of a resilient family, (5) feeling connected with their family, (6) living in a supportive neighborhood, and (7) living in a safe neighborhood. PCE survey questions, response options, and categorization of responses are detailed in Table 3.

Table 3.

Associations of each positive childhood experience (PCE) and insufficient sleep duration

Positive childhood experience	Insufficient sleep (%)	95% CI	aOR*	95% CI	p-value
Volunteers in community
No	36.7	[35.7,37.7]
Yes	31.0	[29.8,32.2]	0.90	[0.84,0.97]	< .0001
Participates in after school activities
No	42.3	[40.5,44.1]
Yes	32.2	[31.4,33.0]	0.85	[0.78,0.93]	< .0001
Has an adult mentor
No	42.3	[39.4,45.1]
Yes	33.7	[32.9,34.4]	0.82	[0.73,0.92]	< .0001
Resilient family
No	45.2	[43.0,47.4]
Yes	32.8	[32.0,33.6]	0.65	[0.59,0.71]	< .0001
Connected family
No	39.5	[38.2,40.8]
Yes	31.8	[30.9,32.8]	0.70	[0.65,0.75]	< .0001
Lives in a supportive neighborhood
No	39.7	[38.5,41.0]
Yes	31	[30.0,32.0]	0.79	[0.73,0.85]	< .0001
Lives in a safe neighborhood
No	40.1	[38.6,41.5]
Yes	32.1	[31.2,33.0]	0.80	[0.75,0.87]	< .0001

aOR indicates adjusted odds ratio; CI, confidence interval.

^*adjusted odds ratio, adjusted for age, biological sex, race & ethnicity, family income, insurance, family structure, primary household language, parent education, and census geographical region

Source: National Survey of Children’s Health, 2022–2023

All PCEs were categorized as binary (yes or no). In addition, PCEs were summed to calculate a composite score reflecting the total number of PCEs each participant experienced. The summary score was then categorized into the groupings: 0–2 PCEs, 3–4 PCEs, and 5–7 PCEs to 1) to assure adequate statistical power to detect meaningful associations, and 2) simplify reporting of results by narrowing the number of comparison groups requiring reporting as documented in previous publications.^15,20–23 Higher PCE scores indicated more positive experiences during childhood. We previously reported on associations between PCEs and chronic pain using data from NSCH.¹⁷

Adverse childhood experiences

The NSCH identifies 9 ACEs, occurring during the youth’s lifetime, that have been reported in several studies to date.^19,24–27 Five of these ACEs were adopted from the Behavioral Risk Factor Surveillance System ACE Module (Centers for Disease Control and Prevention), including questions regarding (1) divorce or separation of parents, (2) incarceration of parent, (3) witnessing of domestic violence by child, (4) child cohabitated with someone who is mentally ill and/or suicidal, (5) child cohabitated with someone with substance abuse problems. Additional survey questions included categories surrounding (6) child being treated or judged unfairly due to race or ethnicity, (7) child having experienced death of a parent, (8) child having been victim of or witnessed violence within their neighborhood, and (9) hardship suffered by child because of family with low income. These additional survey items were developed from expert panel input to capture potentially stressful life-course events and experiences. Some of the questions within the survey used a Likert Scale, for which responses categorized as “somewhat often” or “very often” were analyzed as a definitive positive (yes) response and “rarely” or “never” as a definitive negative (no) response. This meant that all ACEs were categorized as binary (yes or no).

The ACEs were summed to calculate a composite score reflecting the total number of ACEs each participant experienced. We created ACE cumulative score groupings of 0 ACEs, 1 ACE, and 2+ ACEs to (1) assure adequate statistical power to detect meaningful associations, and (2) simplify reporting of results by narrowing the number of comparison groups requiring reporting, consistent with previous publications.^15,17,28

Covariates

NSCH collected on the following socio-demographic variables: age (6–17 years; responses from parents of youth outside of this range were excluded), sex (biologically male vs. female), race or ethnicity (reported by parents or caregivers and categorized by NSCH as: White, non-Hispanic; Black, non-Hispanic; Hispanic, and Other, non-Hispanic), household income based on federal poverty status (0%−99% Federal Poverty Level (FPL), 100%−199% FPL, 200%−399% FPL, 400% or greater FPL), insurance type (private, public only, uninsured), family structure (2 parent, single parent household, or other), primary household language (English vs. non-English), parent education (less than high school, high school and some college or associates degree, college degree or higher), and census geographical region (Northeast, Midwest, South, West).

Data analysis plan

Data were analyzed using Stata V18.0. Hypothesis testing was 2-sided, and p-values set to < .05. We adjusted for the complex survey design of NSCH and therefore all estimates presented here are nationally representative of the non-institutionalized youth population in the United States. We also conducted descriptive analyses to estimate overall rates of PCEs in our sample.

To address our primary aim, we estimated prevalence rates of insufficient sleep stratified by PCE cumulative score grouping (0–2 PCEs, 3–4 PCEs, and 5–7 PCEs). Differences in unadjusted prevalence rates of insufficient sleep based on PCEs score groupings were determined using design-adjusted Pearson chi-square analysis. We then conducted multivariable logistic regression analysis to estimate associations between PCEs and insufficient sleep adjusting for covariates, including age, biological sex, race and ethnicity, family income, insurance, family structure, primary household language, parent education, and census geographical region. These covariates were included based on existing epidemiological research finding strong associations with adverse health outcomes, including insufficient sleep, PCEs, and ACEs.^4,19,29–31 We did not include mediators potentially residing in the causal pathways between PCEs and insufficient sleep to limit overadjustment bias.³⁰

To address our secondary aim, we first conducted multivariate logistic regression analysis, using interaction terms, to formally test whether PCEs moderate the associations between ACEs and insufficient sleep. We considered both linear and non-linear models (including adding polynomial terms, using spline terms, and categorizing variables). Using Stata’s svylogitgof command we estimated that the best model fit was a non-linear model including a non-linear term for ACEs. This indicates that the effects of ACEs and PCEs on sleep are non-linear. It appeared that the predicted probability of insufficient sleep clearly increased as the number of ACEs increased, while it decreased with greater number of PCEs. These associations are presented as a contour plot (Fig. 1). The nature of this association was then further explored by stratifying our sample according to cumulative ACEs score (0 ACEs, 1 ACE, 2+ ACEs). After stratification, we estimated associations between PCEs and insufficient sleep within each ACEs category using multivariable logistic regression, controlling for covariates, including age, biological sex, race and ethnicity, household income, insurance, family structure, primary household language, parent education, and census region. These analyses are consistent with methods from previous studies examining whether PCEs reduce risk for adverse health outcomes associated with ACEs.

Fig. 1.

Predicted probability of insufficient sleep by number of PCEs and ACEs reported by children and adolescents in the United States. PCE indicates positive childhood experiences; ACE, adverse childhood experiences

Source: National Survey of Children’s Health, 2022–2023.

Given that sleep-related outcomes differ by sex and age in children and adolescents, we conducted sensitivity analyses to determine whether associations between sleep and PCEs or ACEs differed by sex and age differences.

Results

Sample characteristics

The sample included 56,540 parent responses for child and adolescent participants 6–17 years of age (Table 1). Approximately half (51.4%) of the sample was between 12–17 years of age, 51.2% were male (48.8% were female), 48.6% were White, non-Hispanic, and 85.3% were from primarily English-speaking households. Over half (51.6%) of parents had a college degree or higher and 71.3% of families were a part of a 2 parent household. The majority (57.8%) of youth had experienced 5–7 PCEs, 31.4% were exposed to 3–4 PCEs, and 10.8% reported 0–2 of the PCEs surveyed. Over a third (34.7%) of youth reported insufficient sleep duration.

Table 1.

Sample sociodemographics

Variable	Number in Sample n=56,540	%	95% CI
PCE Categorical
0–2 PCEs	4545	10.8	[10.2,11.4]
3–4 PCEs	15,984	31.4	[30.7,32.2]
5–7 PCEs	36,011	57.8	[57.0,58.6]
Insufficient sleep for age
No	38,932	65.3	[64.5,66.0]
Yes	17,608	34.7	[34.0,35.5]
Age category
6–11 years	26,438	48.6	[47.8,49.4]
12–17 years	30,102	51.4	[50.6,52.2]
Biological sex of child
Male	29,777	51.2	[50.4,52.0]
Female	26,763	48.8	[48.0,49.6]
Race and ethnicity
White- non-Hispanic	36,804	48.6	[47.8,49.4]
Black- non-Hispanic	3449	12.5	[11.9,13.1]
Hispanic	8300	26.5	[25.7,27.3]
Other - non-Hispanic	7987	12.4	[11.9,12.8]
Income based on federal poverty level status
0%−99% FPL	6683	17.3	[16.6,18.0]
100%−199% FPL	8830	19.5	[18.8,20.1]
200%−399% FPL	16,750	29.4	[28.7,30.2]
400% FPL or greater	24,277	33.8	[33.1,34.5]
Insurance type
Private	42,268	64.7	[63.9,65.5]
Public only	11,863	28.6	[27.9,29.4]
Uninsured	2409	6.7	[6.2,7.3]
Family structure
Two parent household	42,741	71.3	[70.6,72.1]
Single parent household	11,771	24.4	[23.6,25.1]
Other	2028	4.3	[4.0,4.7]
Primary language
English	51,791	85.3	[84.6,86.1]
Non-English	4456	14.7	[13.9,15.4]
Parent education
Less than High School	1478	8.7	[8.1,9.4]
High School and some college or associates degree	18,938	39.2	[38.4,39.9]
College degree or higher	36,124	51.6	[50.8,52.3]
USA census region
Northeast	10,064	15.8	[15.4,16.3]
Midwest	13,923	20.9	[20.4,21.4]
South	15,192	39.2	[38.5,40.0]
West	17,361	24.0	[23.4,24.5]

Source: National Survey of Children’s Health, 2022–2023

Positive childhood experiences and insufficient sleep

Rates of insufficient sleep duration were lowest (29.5%) among youth who had experienced all or almost all (5–7) PCEs (Table 2). Among those exposed to 3–4 PCEs, rates increased to 39.8%. Insufficient sleep was most prevalent among youth who had experienced very few or no (0–2) PCEs (48.0%).

Table 2.

Prevalence and adjusted odds ratios of associations between positive childhood experiences (PCEs) and insufficient sleep among children and adolescents in the USA

	Prevalence of Insufficient sleep			Adjusted Odds Ratios for Associations between PCEs and Insufficient Sleep
Number of PCEs	Weighted %	95% CI	p-value	aOR*	95% CI	p-value
0–2 PCEs	48.0	[45.1,50.9]	< .0001	ref
3–4 PCEs	39.8	[38.4,41.3]		0.78	[0.69,0.90]	< .0001
5–7 PCEs	29.5	[28.6,30.4]		0.57	[0.50,0.68]	< .0001

aOR indicates adjusted odds ratio; CI, confidence interval.

^*adjusted odds ratio, adjusted for age, biological sex, race & ethnicity, family income, insurance, family structure, primary household language, parent education, and census geographical region

Source: National Survey of Children’s Health, 2022–2023

Multivariable analysis confirmed that among youth experiencing 5–7 PCEs, the odds of insufficient sleep were significantly lower as compared to those experiencing 0–2 PCEs (adjusted odds ratio (aOR): 0.57, 95% confidence interval (CI): 0.50–0.68, p < .0001). Likewise, youth experiencing 3–4 PCEs had lower odds of insufficient sleep as compared to the same reference group (aOR: 0.78, 95%CI: 0.69–0.90, p < .0001). Similar variations in insufficient sleep prevalence were observed when each of the 7 PCEs were evaluated separately (Table 3). Other than having an adult mentor, each individually measured PCE was associated with a significantly lower odds of insufficient sleep compared to individuals who had not experienced that PCE.

Positive childhood experiences and insufficient sleep: associations across adverse childhood experiences levels

In our contour plot, we plotted the probability of insufficient sleep associated with PCEs for different levels of ACEs as estimated by multivariate logistic regression analysis (Fig. 1). This analysis revealed that children and adolescents who were not exposed to any ACEs and were exposed to 5–7 PCEs had the lowest predicted probability for insufficient sleep (20%−30%). In contrast, children and adolescents who experienced both 3 or more ACEs and 4 or fewer PCEs had a predicted probability of insufficient sleep ranging above 60%. Youth who experienced 7 or more ACEs and 1or fewer PCEs had an even greater predicted probability of insufficient sleep, ranging above 70%. Adding interaction terms into our multivariable model did not reveal a significant interaction between ACEs and PCEs, However, when the model was stratified by level of PCEs, we found a significant interaction between ACEs and PCEs when children had 5–7 PCEs (p < .04). Therefore, we then conducted stratified multivariate analysis by different levels of ACEs (Table 4). As expected, for each level of ACEs, as the number of PCEs reached 5–7, the probability of having insufficient sleep significantly decreased.

Table 4.

Prevalence of insufficient sleep by positive childhood experiences (PCEs) scores for each of 3 adverse childhood experiences (ACEs) exposure levels

ACEs category	Weighted %	95% CI	aOR*	95% CI	p-value
0 ACEs
0–2 PCEs	43.8	[38.5,49.2]	ref
3–4 PCEs	35.1	[33.0,37.2]	0.8	[0.6,0.9]	.027
5–7 PCEs	26.6	[25.5,27.7]	0.6	[0.5,0.7]	< .0001
1 ACE
0–2 PCEs	45.1	[39.2,51.0]	ref
3–4 PCEs	40.1	[37.4,43.0]	0.8	[0.6,1.1]	.152
5–7 PCEs	33.4	[31.3,35.5]	0.6	[0.5,0.8]	.001
2 or more ACEs
0–2 PCEs	53.8	[49.6,58.0]	ref
3–4 PCEs	47.8	[44.9,50.7]	0.8	[0.6,1.0]	.119
5–7 PCEs	36.8	[34.3,39.4]	0.6	[0.4,0.7]	< .0001

aOR indicates adjusted odds ratio; CI, confidence interval.

^*adjusted odds ratio, adjusted for age, biological sex, race & ethnicity, family income, insurance, family structure, primary household language, parent education, and census geographical region

Source: National Survey of Children’s Health, 2022–2023

Sensitivity analyses found that associations between sleep and PCEs or ACEs did not differ by age category or biological sex.

Discussion

We examined associations between positive childhood experiences and insufficient sleep using a large national sample of youth in the United States. Our study had 2 main findings: greater exposure to PCEs was associated with a lower prevalence of insufficient sleep, and PCEs were associated with reduced risk of insufficient sleep among youth exposed to ACEs.

While, to our knowledge, no previous studies have examined the relationship between PCEs and sleep duration, our findings that greater exposure to PCEs was associated with significantly lower rates of insufficient sleep builds upon 2 previous studies examining the role of PCEs as a protective factor against insomnia. In a sample of Taiwanese young adults, Wu et al. found a significant inverse relationship between PCEs and persistent insomnia, surveyed using a 5-point Likert Scale measuring difficulty falling asleep.¹¹ Similarly, in a longitudinal cohort study of mental health among Chinese emerging adults, Geng et al. found a higher number PCEs were significantly associated with decreased rates of all severity levels of insomnia using the Youth-Self-Rating Insomnia Scale.¹⁰ Direct comparisons with these studies is challenging because the domains of sleep examined were not consistent between studies. Moreover, both studies involved adult participants and enrolled local community samples not generalizable to national populations. Indeed, the National Survey of Children’s Health is one of the few population-based studies that measures both PCEs and sleep duration among youth in the United States. Extending these efforts to other population health surveys will greatly advance our understanding of the protective effects of PCEs.

While the impact of PCEs on insufficient sleep in a large national sample has not previously been evaluated, researchers have examined the role of PCEs on other childhood mental and physical health conditions. PCEs are associated with lower rates of childhood depression, anxiety, and obesity among youth in the United States.³² Regarding psychosocial processes, from a compensatory model of resilience, PCEs may serve as promotive factors that foster positive adjustment and mental health.^33,34 For instance, it has been proposed that PCEs may improve mental health through a decrease in stress and an increase in self-esteem. In fact, previous longitudinal research has shown a bidirectional relationship between psychological well-being and good sleep.^35–37 The demonstrated positive impact of PCEs on childhood psychosocial development is an additional pathway that may contribute to the positive association between number of PCEs and sleep duration.

Additionally, we found PCEs moderated the negative associations of adverse childhood experiences and insufficient sleep. Previous work found strong associations between ACEs and insufficient sleep.^4,38,39 In our study, for each level of ACEs exposure, greater number of PCEs was associated with lower prevalence of insufficient sleep, based on parental reports of how many hours their child slept on an average weeknight. Children and adolescents with a high number of positive experiences and no ACEs had the lowest rates of insufficient sleep duration. Inversely, youth with few or no PCEs and 2 or more ACEs had the highest rates of insufficient sleep duration. The protective factor model of resilience theory posits that enhancing promotive factors provides youth with resources necessary for healthy child development, thereby helping youth overcome risk by modifying the relationship between risk factors and outcomes.³³ Previous studies have demonstrated the negative impact of ACEs on sleep duration.^38,39 High rates of PCEs may moderate the relationship between insufficient sleep and ACEs by serving as a buffer against the social and psychological impact of trauma, therefore reducing the rates of adverse health consequences. Indeed, it has previously been shown that PCEs moderate the negative associations of ACEs in children and adolescents within social, psychological, and biological domains of health, including but not limited to substance abuse, depression, anxiety, and chronic pain.^{17,32,40–42}

This study has several limitations that should be acknowledged. Firstly, a causal relationship cannot be drawn between PCEs and insufficient sleep duration due to the cross-sectional study design. While we hypothesize PCEs are protective against the development of insufficient sleep duration in youth, it has also been proposed that this relationship is bidirectional.⁴³ If insufficient sleep also leads to poor emotional regulation and behavioral problems, this could result in youth having less ability to participate in PCEs and decreased resilience in the face of traumatic experiences. Secondly, since the retrospective questionnaire was administered to parents, reporting bias may lead to underreporting of negative information and over-reporting of positive information compared with youth self-reports. For instance, parents are less likely to report their teen has experienced an ACE than their teen.⁴⁴ Thirdly, previous research demonstrates parents overestimate their child’s sleep duration compared with actigraphy.⁴⁵ This suggests the true prevalence of insufficient sleep amongst children may be greater than reported. Interestingly, multiple studies suggest children overestimate their sleep duration to a greater extent compared with their parents.^13,45 Fourthly, our findings on the impact of both PCEs and ACEs on sleep are limited by the variables assessed on the NSCH. However, this is a known limitation of large-scale national surveys.

While there are some weaknesses, this study does have significant strengths including the use of a nationally representative database, a large sample of children and adolescents with insufficient sleep, inclusion of comparison groups, and a measurement of multiple types of adverse and positive childhood experiences, all of which allow for greater reciprocity of results across the national pediatric population.

Our findings are useful in evolving nationwide efforts to decrease the burden of insufficient sleep in children and adolescents, highlighting the need for increased public health attention in the assessment and promotion of positive childhood experiences and child resilience resources more broadly. Much attention has been given to the negative health impacts of ACEs, and relatively less, although growing, attention has been given to the relationship between promotive experiences and health, both independently and as protective factors against childhood adversity. With respect to sleep, research on the role of socio-ecological factors has focused primarily on risk factors for poor sleep. Amongst children and adolescents, research does suggest that resiliency is a mediator of the impact of sleep behavior.⁴⁶ However, research investigating the role of resilience factors in reducing the rates of insufficient sleep is virtually nonexistent. Our findings highlight the importance of seeking to reduce rates of childhood adversity in tandem with actively promoting positive childhood experiences to most effectively promote healthy child development. Frameworks which have described and categorized these promotive factors, such as the Health Outcomes from Positive Experiences framework developed by Sege and Browne⁴⁷ at the Center for the Study of Social Policy, highlight the need for interventions at the family, community, and societal levels.

Promoting PCEs at each level of these socio-ecological systems has the untapped potential to improve insufficient sleep and the lives of children and adolescents. Modifiable PCEs examined in this study include connection and resilience at the family level, having an adult mentor and a supportive neighborhood at the community level, and access to volunteer and after school opportunities at the societal level. Interventions at the family level include screening for PCEs in addition to ACEs during healthcare visits. Primary care providers should ideally screen for ACEs and encourage PCEs throughout childhood to potentially mitigate adverse health events. Highlighting and encouraging PCEs during routine clinical encounters allows families to recognize the protective role of resilience factors and fosters motivation to engage in these behaviors and experiences.⁴⁸ While familial factors, such as programs to bolster child confidence and to improve parenting may be more easily targeted by clinicians, opportunities to promote PCEs across socio-ecological levels merits further examination.

For instance, at the community level, peer-based support groups allow for parents and their children to have increased social support.⁴⁹ Parent participation in these interventions has been associated with improved sleep behaviors in early childhood.⁵⁰ Strong social networks have been shown to promote perceived well-being in adulthood at all levels of ACEs exposure.⁵¹ At the societal level, it is essential to increase funding for programs that seek to increase opportunities for PCEs to occur. In schools, social-emotional learning programs which incorporate resilience-oriented approaches show potential to promote engagement in positive behaviors, forge social connectedness, and decrease internalizing problems in school aged children and adolescents.⁵² Holistic approaches to decreasing rates of insufficient sleep in youth should allocate resources to enhancing PCEs in addition to reducing risk of ACEs.

Conclusion

In conclusion, in a nationally representative sample of children and adolescents in the United States, we found that PCEs were associated with lower prevalence rates of insufficient sleep duration. Furthermore, PCEs moderated the known negative associations between ACEs and insufficient sleep duration. Efforts to translate these findings into clinical practice and public health interventions aimed at developing and engaging youth in positive experiences across family, community, and policy domains may lead to a decrease in the rates of insufficient sleep duration. Overall, these findings highlight the need for identifying and proactively promoting PCEs in children and adolescents in order to reduce the risk of insufficient sleep and associated negative health consequences.

Data sharing

The National Survey of Children’s Health methodology, code-books, and datasets are publicly available and de-identified through the Data Resource Center for Child & Adolescent Health (www.childhealthdata.org), a project of the Child and Adolescent Health Measurement Initiative.