Abstract Introduction Healthy sleep is critical for the physical, emotional, and cognitive development of adolescents. Capturing the multidimensional components of sleep behavior in this age group remains a challenge, particularly when relying on retrospective questionnaires. This study compares sleep data collected via youth-report, caregiver-report and from Fitbit devices in a large, diverse sample of early adolescents. Methods This study analyzed data from the Adolescent Brain Cognitive Development (ABCD) Study, comprising 11,879 US adolescents (Year 2 age: 11−14 years). The participants self-reported their sleep period (from falling asleep to wake-up time) via the Munich Chronotype Questionnaire, while their caregivers completed the Children's Sleep Disturbance Scale. A subset (N = 4,282, Mean Age = 11.97 years, 51.19% female) also wore Fitbit Charge 2 devices for sleep tracking for 21 days, directly following the annual assessment. We evaluated the internal consistency of the questionnaires and employed Bland-Altman and interclass correlation analyses to compare self-reported and Fitbit measures of sleep period. Results We found acceptable internal consistency in the youth-reported (α = 0.71) and caregiver-reported (α = 0.83) sleep questionnaires. There was a greater discrepancy between caregivers and adolescents when adolescents reported sleep durations less than 7 hours: 38% of caregivers reported 7-8 hours and 35% reported 8-9 hours whereas only 15% reported < 7 hours. Compared to Fitbit measures adolescents generally estimated their sleep period with reasonable accuracy (ICC = 0.182 [0.15, 0.21] p<.001), displaying an average discrepancy of 40.5 minutes on weekdays. Conclusion The findings indicate that there is reasonable agreement between youth-report and Fitbit measures of sleep period in adolescents, opening up the possibility for assessing the complexity of sleep behavior. In this age group, caregivers tend to overestimate the adolescents' sleep duration, which highlights the importance of youth-reported sleep along with objective measures. This study contributes valuable insights into the methodology of sleep research and underscores the need for multi-dimensional approaches in assessing sleep patterns in adolescents. Support (if any) National Institute of Health: U01DA041022

Abstract Introduction Poor sleep and mental health problems commonly emerge in adolescence, with a higher prevalence in females and those who suffer from obesity. Here, we evaluated the complex associations between poor sleep, obesity, and biological sex and how they interact to influence wellbeing in a large, diverse cohort of adolescents in the US. Methods Data were analyzed from 7,261 adolescents (Year 2: Mean age=11.94 years, range: 10-14 years, 47.3% female), collected as part of the ongoing ABCD Study®. Sleep duration was assessed with the Munich Chronotype questionnaire (youth-report), and sleep quality was assessed with a question about having trouble falling or staying asleep in the past two weeks from the Kiddie Schedule for Affective Disorders and Schizophrenia (KSADS-5) (youth-report). Internalizing and externalizing problems were assessed using the Child Behavior Checklist (caregiver-report). Positive affect was assessed with the NIH-toolbox measures. Regression models examined associations between obesity (>95 body mass index percentile), sleep behavior and mental health, considering sex differences, age, and socio-demographic characteristics. Results Short sleep duration and poorer sleep quality, as well as obesity were associated with higher internalizing (p<.01) and externalizing (p<.01) problems. Adolescents with obesity and prolonged sleep duration had higher internalizing (p=.01) and externalizing problems (p<.01). Shorter sleep duration (p<.01) and poorer sleep quality (p=.01) were associated with lower positive affect, with stronger effects in female adolescents. Obesity partially mediated the association between sleep problems and positive affect (p<.01). Conclusion This study demonstrates a strong link between sleep, obesity, and mental health outcomes in adolescents, highlighting significant sex differences. Poor sleep quality and obesity is related to more internalizing and externalizing problems and lower positive affect, with notable with notable impacts on female adolescents. These findings highlight the importance of addressing sleep and obesity in adolescent mental health interventions, recognizing the unique vulnerabilities and needs of male and female adolescents and those with obesity. Support (if any) National Institutes of Health: U01DA041022

Abstract Introduction Numerous physiological processes display menstrual cycle variations, including body temperature. The advances in quality and accessibility of wearables facilitate collecting time series of physiological data, including skin temperature during sleep. The cosinor method, frequently used in circadian rhythms biology, may be a useful tool to assess if a menstrual cycle is ovulatory, based on a biphasic temperature rhythm. It could also be used to derive metrics about the cycle, in turn allowing the investigation of rhythm characteristics of females at different reproductive stages. Methods Here, 67 females in the early reproductive (age: 25.5 ± 5.4 years (mean ± SD)) and 53 females in the late reproductive/menopausal transition (age: 47 ± 2.9 years) stages tracked sleep and temperature with an Oura ring 2 across a menstrual cycle. They also reported menses and used an ovulation kit that detects a rise in luteinizing hormone. A cosinor method was fitted to daily skin temperature points extracted during the sleep period, and the fit quality was compared with the ovulation kits results. The cycle metrics were extracted and compared between the two groups with Wilcoxon tests. Results With the cosinor method, a cycle was considered ovulatory when the fit had a r2 > .25, a method that agreed with the ovulation kit in 82% of cases. When the fit quality was r2 > .4, the model was considered sufficiently good to calculate derived metrics. There was no difference in the fit quality, acrophase relative to menses, or amplitude of the rhythm between the early and late reproductive/menopausal transition groups. However, the latter had a higher mesor compared with the early reproductive stage group (p = 0.03). Conclusion The cosinor method can be used to model not only circadian but also menstrual rhythms, allowing identification of ovulatory cycles, and derivation of metrics about menstrual cycle rhythms that can be used to track characteristics within and between individuals over time. The overall higher skin temperature rhythms found in the late reproductive/early menopausal transition group could reflect shifted temperature regulation and more heat dissipation during this stage. Support (if any) National Institutes of Health (NIH) grant RF1AG061355 (Baker/Mednick)

Abstract Introduction In today’s digital landscape, social media features prominently in adolescents’ social interactions and information consumption, influencing their development, potentially altering brain processes. Research shows a bidirectional relationship between social media use and both sleep health and brain activities, especially in executive control and reward processing. These neural developments are pivotal in adolescent behavioral and psychological growth, with sleep being crucial yet underexplored in the context of reward, social (media) behaviors. Methods This study investigated the reciprocal links between social media use, self-reported sleep duration, and brain activation in 6,516 adolescents (ages 10-14 years, 46.2% female) from the Adolescent Brain Cognitive Development (ABCD) Study®. Sleep duration was assessed from the Munich Chronotype questionnaire, and recreational social media use through the Youth Screen Time Survey. Brain activities were analyzed from fMRI scans during the Monetary Incentive Delay (MID) task, targeting regions associated with reward processing. The study used three different sets of models, switching predictors and outcomes each time, to examine the reciprocal relationships between sleep duration, brain activation, and social media use, and their interactions. Age, COVID-19 pandemic timing (before/during), and socio-demographic characteristics were included in the models. Results Shorter sleep duration correlated with greater social media usage (p<.001). Notably, interactions between sleep duration and brain activation in the cingulate gyrus (p=.021), inferior frontal gyrus (p=.009), and precuneus (p=.008) predicted social media use. In predicting brain activity, interactions between sleep duration and social media use emerged as significant for the inferior (p=.008) and middle frontal gyrus(p=.003). For sleep duration predictions, interactions were important between social media use and brain activation spanning seven areas, including the cingulate gyrus (p=.039), hippocampus (p=.005), insula (p=.015), inferior frontal gyrus (p=.001), middle frontal gyrus (p=.003), precuneus (p=.038) and the superior frontal gyrus (p=.028). Conclusion These results highlight distinct relationships between sleep, social media, and activity across frontolimbic brain regions, key for executive control and reward processing. Such insights deepen our understanding of how individual's unique neural sensitivities to digital technology use and sleep necessity interact in adolescents. Future longitudinal studies based on these findings could pave the way for developing more nuanced, individualized interventions. Support (if any) NIH: U01DA041022 and R01MH128959.

Abstract Introduction Adolescence in girls is a vulnerable transitional period, characterized by substantial hormonal and physical development, the onset of menarche, dramatic brain development and changes in behavior, including sleep. Menstrual problems, including painful menses, are common among female adolescents, and contribute to school absenteeism. They may also be symptoms of gynecological conditions, which can negatively affect reproductive and general health of women. Limited work has linked sleep disturbances and menstrual problems in adolescents. Here, we examined the association between sleep behavior and menstrual problems in a large sample of adolescent girls in the Adolescent Brain Cognitive Development (ABCD) Study®. Methods We fitted linear mixed effect models to examine the associations between sleep behavior and menstrual problems in 1837 post-menarcheal girls (Year 3, Mean-age = 13.03, [12-15] years), assessed as part of the ongoing ABCD Study®. Sleep was assessed with the Munich Chronotype questionnaire (youth-report) and the sleep disturbance scale for children (caregiver-report). Girls answered questions about menstrual cycles and associated problems. We considered age, time since menarche, BMI, use of hormonal contraceptives, and socio-demographic characteristics in the models. Results Short sleep duration was related to higher menstrual pain intensity (p<.01), irregular menstrual cycles (p<.01) and more premenstrual symptoms (p=.01). Higher total sleep disturbance score (caregiver-reported) was also associated with higher menstrual pain intensity (p<.01) and greater impact of menstrual pain on usual activities (p<.01). Participants with a later chronotype (p=.04) and later wake-up time (p=.02) were more likely to experience an irregular menstrual cycle. Later wake-up time was also related to higher menstrual pain intensity and impact on daily life (p<.01). Girls with recent menarche reported heavier menstrual flow (p=.02), higher menstrual pain intensity, more premenstrual symptoms, and a greater overall impact of menstrual pain on their usual activities (p<.01). Conclusion Our results indicate multiple associations between sleep behavior and menstrual problems in adolescent girls. Both menstrual problems and insufficient sleep/sleep disturbances are important for female adolescent health and should be routinely screened for by healthcare providers and school health education providers. Future work is needed with longitudinal analyses to determine directionality of the associations we found between sleep and menstrual problems. Support (if any) NIH U01DA041022