Purpose: An increasing number of elementary age youth own smartphones. Cross-sectional studies have linked smartphone use with childhood obesity, but longitudinal evidence is lacking.  Smartphones may be particularly detrimental for children with low sleep resilience (i.e. the ability to re-establish typical sleep following a night of atypical sleep). Given that children are at risk for accelerated weight gain over the summer, the current study examined if smartphone ownership was associated with summer weight gain in children. This association was hypothesized to be stronger among children with low sleep resilience.

Methods: Timeseries data from Fitbits (Charge-2) were collected on 6,642 days from 196 children (median 22 days/child) as part of a larger school-based study.  The sample was 53% female, 5-10yrs old with a mean zBMI of 0.75 (SD = 1.2). Dynamic Structural Equation Modeling (DSEM) was used to estimate sleep resilience from Fitbit data, defined as previous night sleep regressed on current night sleep (i.e., autoregression). Parents reported smartphone ownership and zBMI was measured in spring and fall of 2018. Linear regression was used to test sleep resilience as a moderator of smartphone ownership and zBMI change.

Results: 29% of children owned a smartphone. Sleep resilience was unrelated to zBMI change or smartphone ownership. Children who owned a smartphone had greater increases in zBMI over the summer (β = 0.22, 95%CI 0.05 to 0.38). Sleep resilience moderated (β = -2.10, 95%CI -4.14 to -0.06) such that smartphone ownership was associated with zBMI increases among children with low sleep resilience. Smartphone ownership was not associated with zBMI change for children with higher than average sleep resilience.  

Conclusions: Owning a smartphone is a risk for accelerated summer weight gain in the context of average or low sleep resilience. Future longitudinal research should explore potential mechanisms and confounders of this association (i.e. if smartphone ownership engenders or exacerbates existing sleep dysregulation perhaps through blue light exposure). Alternatively, high sleep-resilience may indicate environmental context which supports sleep consistency, such as household rules or routines which may protect against negative health sequelae associated with children’s smartphone usage. 

Purpose: Changes in obesogenic behaviors (physical activity, sedentary time, screen-time, diet, sleep) during summer break (SB) may explain accelerated summer weight gain. This preliminary study examined changes in obesogenic behaviors of children attending year-round and traditional schools during SB.

Methods: Primary school children (48.7% male, 57.4% African American) attending one year-round (n=73) or two traditional schools (n=117) participated in this study. Fitbits (sleep, activity, sedentary) and daily diaries (screen time, diet) measured obesogenic behaviors over five measurement waves during school (spring 2018, fall 2018, spring 2019) and SB (2018 & 2019). SB was 11 weeks in traditional schools and 5 weeks in year-round schools. Multilevel regressions with schedule (yearround vs. traditional), time (SB weekdays vs. school days), and schedule-x-time as independent variables and engagement in obesogenic behaviors as the dependent variable estimated changes in obesogenic behaviors during SB.

Results: Children provided a mean 67.3 and 44.9 days of Fitbit and 13.7 and 12.9 daily diaries during school and SB, respectively. During SB moderate-to-vigorous physical activity (MVPA) increased for traditional children (5.3min, 95CI: 1.9, 8.7) but decreased for year-round children (-5.9min, 95CI: -10.8, -1.0). Sedentary time increased for traditional (36.1min, 95CI: 28.9, 43.3) and year-round (72.9min, 95CI: 62.6, 83.3) children. Screen time increased for traditional (49.3min, 95CI: 41.1, 57.5) and year-round (54.0min, 95CI: 43.1, 64.9) children. The number of unhealthy foods/beverages consumed increased for traditional children (0.3, 95CI: 0.2, 0.5) while healthy foods/beverages consumed decreased for year-round children (-0.2, 95CI: -0.3, -0.1). Sleep midpoint shifted later for traditional (88.5min, 95CI: 83.9, 93.1) and year-round children (113.9min, 95CI: 107.4, 120.5). Year-round children experienced more unfavorable changes in sedentary (+30.3min, 95CI: 18.6, 42.1), MVPA (-9.3min, 95CI: -14.2, -4.4), sleep midpoint (25.3, 95CI: 17.3, 33.3) and healthy food (-0.2, 95CI: -0.3, -0.1) consumed during SB compared to traditional children.

Conclusions: During SB children experienced unfavorable changes in some obesogenic behaviors with the magnitude of change differing by SB length. Understanding obesogenic behavior changes during SB is essential to interventions targeting accelerated summer weight gain.

Purpose: Although the relationship of unhealthy dietary patterns with depression and some mental disorders is well documented, little is known about whether dietary behaviours are associated with sleep problems in adolescents. This study aimed to examine the associations of fast-food and carbonated soft drink intake with anxiety-induced sleep disturbance among adolescents around the world.

Methods: We analysed the Global School-based Student Health Survey data, collected between 2007 and 2016, from 162,526 adolescents aged 12–15 years (49% girls) in 59 countries. Adolescents reported frequency of being so worried that they could not sleep during the past 12 months, and having carbonated soft drinks during the past 30 days and fast-food during the past 7 days. We used multivariable logistic regression to examine country-level associations. Multilevel logistic regression analysis was used to examine the relationships, taking into account the hierarchical structure of the data. Pooled estimates were derived from country-level estimates by using meta-analysis with random effects.

Results: Overall, 7.3% of adolescents reported having anxiety-induced sleep disturbance most or all of the time during the past 12 months (boys: 6.5%; girls: 8.3%). The prevalence of sleep disturbance increased linearly with increased soft drinks and fast-food intake across gender. Adolescents having soft drinks at least daily had 29% higher odds of sleep disturbance than those who had soft drinks less than daily (OR=1.29, 95% CI 1.25–1.34). Adolescents who had fast-food on at least two days/week had 31% higher odds of sleep disturbance than those who had fast-food on at most one day/week (1.31, 1.26–1.36). No apparent gender differences were observed in the association estimates. Meta-analysis showed that adolescents with frequent intake of soft drinks (at least daily) and fast-food (at least two days/week) had 48% higher odds of sleep disturbance than those with infrequent intake (1.48, 1.33–1.57; I²=41.6%). Sensitivity analyses supported the meta-analysis results. 

Conclusions: Our findings suggest positive associations between frequent intake of carbonated soft drinks and fast-food with anxiety-induced sleep disturbance among adolescents. Prospective studies are needed to understand the causal pathways of these relationships, which can inform strategies to reduce sleep disturbance among adolescents. 

Purpose Evidence on the relationship between sleep and bone health in youth is still scarce. The present study aims to examine cross-sectional and longitudinal associations between self-reported sleep characteristics and bone stiffness index (SI) in European children and adolescents.

Methods 4901 children aged 2-11 years from IDEFICS study and 839 children aged 6-15 years from the subsequent I.Family study providing data on all variables of interest were included. Self-reported sleep characteristics were measured as sleep duration (i.e. nocturnal sleep and daytime napping) and sleep quality (i.e. unregularly bedtime routine, have difficulty falling asleep and trouble getting up in the morning). The calcaneus SI of both feet was measured using quantitative ultrasound. Self-reported sex, age, pubertal status, family socioeconomic status, country, physical activity, screen time and body mass index calculated from measured height and weight were used as covariates in all analyses. Linear mixed-effects models were used to analyse cross-sectional and longitudinal associations between sleep duration, sleep quality and SI with taking confounders into consideration.

Results At baseline, there were 11.20% participants defined as short sleep duration, 6.02% participants defined as long sleep duration, and 44.60% participants defined as poor sleep quality. Daytime napping (every 10min/day) was observed to be positively associated with SI percentiles (β=0.74, 95%CI: 0.39, 1.10, p<0.001). Longitudinal results suggested that only in children who had poor sleep quality at baseline, extreme sleep duration predicted lower SI percentiles after 4 years (β=-7.75, 95%CI: -13.01, -2.48, p=0.004). While only in children who had extreme sleep duration at baseline, poor sleep quality predicted lower SI percentiles after 4 years (β=-11.68, 95%CI: -20.98, -2.39, p=0.014).

Conclusions Daytime napping was positively associated with SI in children and adolescents. Furthermore, the long-term detrimental effect of extreme sleep duration on SI was only observed in children who had poor sleep quality. Similarly detrimental effect of poor sleep quality was only observed in children who had extreme sleep duration.

Purpose During the transition from primary to secondary school adolescents’ sleep disturbances and insufficient sleep duration increases. However, the factors underlying this are largely unknown. Therefore, we conducted a participatory needs assessment to identify 1) adolescent-perceived factors that negatively influence sleep, and 2) adolescent-perceived changes in these factors after the transition from primary to secondary school.

Methods During one school year weekly participatory meetings were held with one group of 10-12-year-olds and one group of 12-14-year-olds (n=5-8 per group) facilitated by an academic researcher. During these meetings adolescents actively participated as co-researchers in exploring factors influencing their sleep behaviour and that of their peers (e.g. using interviews, questionnaires, focus groups with peers). Of all identified factors, adolescent-co-researchers selected the most important factors that may negatively influence sleep, and subsequently indicated potential causal pathways.

Results As most important factor influencing sleep, adolescents mentioned social media: the accompanying notifications (e.g. messages from apps or friends on a mobile device) make them curious and anxious to miss out on something and keep them awake. Being energized by this, some adolescents subsequently game at night. Second, adolescents mentioned that a lack of sleep rules from their parents results in later bedtimes and an increased use of social media or gaming at night. Third, adolescents indicated that when experiencing stress at night - for example due to academic pressure - they lie awake more and have more nightmares. Additionally, environmental factors such as lights and ambient noise, and making homework or being physically activity close to bedtime were mentioned. Last, adolescents mentioned that they like to stay awake at night because it is fun; they consume candy and energy drinks to accomplish this. After the transition from primary to secondary school adolescents perceived less sleep rules from their parents, resulting in later bedtimes and an increased use of social media at night.

Conclusions This participatory study identified factors negatively influencing sleep from the perspective of adolescents themselves. This knowledge may be helpful for the development of effective interventions targeting healthy sleep behaviour for this age group.

Purpose

Movement behaviours (e.g., sleep, sedentary behaviour, and physical activity) in isolation have demonstrated benefits to preschool-aged children’s development. However, little is known on the integrated nature of movement behaviours and their relationship to healthy development in this age range. Thus, the objective of this study was to examine the relationships between accelerometer-derived movement behaviours and indicators of physical, cognitive, and social-emotional development using compositional substitution analyses in a sample of preschool-aged children.

Methods

Children (n=95) were recruited from Edmonton, Canada. Movement behaviours were measured with ActiGraph wGT3X-BT accelerometers, worn 24 hours/day. Physical (i.e., BMI z-scores, percent of adult height, and motor skills), cognitive (i.e., working memory, response inhibition, and vocabulary), and social-emotional (i.e., sociability, externalizing, internalizing, prosocial behaviour, and cognitive, emotional, and behavioural self-regulation) development were measured. Objective height and weight were measured for BMI z-scores and percent of adult height, and the Test of Gross Motor Development-2 was used to assess motor skills. The Early Years Toolbox was used to assess all cognitive and social-emotional development indicators. Compositional substitution models were conducted in R for each development outcome.

Results

Children accumulated 11.1 hours of sleep, 6.1 hours of stationary time, 5.1 hours of light-intensity physical activity (LPA), and 1.8 hours of moderate- to vigorous-intensity physical activity (MVPA). Consistent favourable associations were found when adding MVPA while subtracting other movement behaviours for locomotor, object, and total motor skills; except one null association when removing stationary time for locomotor skills. The majority of associations for stationary time, LPA, and sleep were non-significant. However, some additional favourable associations were found when adding MVPA and subtracting other behaviours (i.e., sociability when subtracting stationary time and sleep; cognitive self-regulation when subtracting LPA), and when adding stationary time and subtracting other behaviours (i.e., vocabulary when subtracting sleep; BMI z-scores when subtracting MVPA).

Conclusions

When considering the composition of movement behaviours in this sample, the findings confirmed the importance of MVPA for motor skills. Favourable results were found for some other substitutions, but further research is needed in larger and more representative samples to confirm these findings.