Adolescents experience chronic partial sleep deprivation primarily due to late bedtimes and early school start times. Late bedtimes are partly driven by late (delayed) circadian rhythms. Morning bright light (MBL) advances (shifts earlier) circadian rhythms; however, extending evening light by staying awake late may delay the circadian clock and limit effectiveness of MBL. We investigated how increasing durations of late evening light coupled with sleep restriction affects the advancing ability of MBL in adolescents (14–17 years). Bedtime was delayed by 0 h (n = 11), 1.5 h (n = 9), 3 h (n = 12), or 4.5 h (n = 8), resulting in sleep opportunities of 10 h, 8.5 h, 7 h, or 5.5 h. Participants received 90 min of MBL on 3 consecutive mornings while the sleep/dark episode gradually shifted earlier. Participants in the 10-h and 8.5-h sleep groups phase advanced by 2.0 ± 1.1 h and 0.5 ± 0.6 h. Participants in the 7-h and 5.5-h sleep groups delayed by 0.8 ± 1.1 h and 2.6 ± 1.2 h. Thus, advance shifts from MBL were reduced in a step-wise manner with increasing duration of evening light and decreasing sleep opportunity, ultimately producing phase delays. These results suggest that MBL efficacy is reduced in adolescents with restricted sleep opportunities associated with late bedtimes and early school start times. These factors need to be considered in circadian-based treatment approaches.

OBJECTIVES:Adolescents in the United States do not typically attain enough sleep at night due to a combination of factors, including biological maturation and academic and social commitments. This lack of sleep leads to poor cognitive, mental, and physical health outcomes. Naps may compensate for inadequate sleep, however, little is known about how napping affects nocturnal sleep and mood over time among adolescents who are poor sleepers. METHODS:The present study used daily diary and actigraphy data collected over 15days in a sample of late- and short-sleeping adolescents (n=99; Mage=15.9years; 55% Female; 46% White, 30% Black, and 24% Multiracial/Other). Multilevel modeling was utilized to examine daily associations between daytime napping and subsequent nocturnal sleep outcomes, controlling for previous night sleep. Sex was explored as a moderator. RESULTS:The majority of participants (76%) napped at least once during the study period, and almost one-fifth of the sample demonstrated "habitual" or frequent napping. Nap occurrence was associated with a decrease in total sleep time by 27 minutes as well as later sleep onset time that night. Later nap offset time was associated with later nocturnal sleep onset time, and longer nap duration was associated with later nocturnal sleep onset time and lower sleep efficiency. Napping was unrelated to affect. Sex did not moderate any relations between napping and nocturnal sleep or affect. CONCLUSIONS:The findings indicate that napping is common and generally associated with worse nocturnal sleep outcomes among a sample of late- and short-sleeping adolescents.

Most adolescents experience chronic sleep restriction and circadian misalignment due to delayed sleep propensity combined with early school start times. The present analysis aims to examine adolescent mood under conditions of sleep restriction and circadian misalignment. 74 (37 female) adolescents (14.1-18.0 years) completed a 15-day study. Participants spent days 1-7 (baseline) at home following individualized 10-h sleep schedules. Participants lived in the laboratory on days 8-14. On days 9-13, sleep was restricted by delaying bedtime 1.5h (n=20), 3h (n=23), or 4.5h (n=20). Between baseline bedtime and delayed bedtime, 34 participants wore amber-lensed glasses (16±2 lux), and 29 participants remained in room light without glasses (104±11 lux). Eleven participants continued the 10-h sleep opportunity followed during baseline. On day 11, sleep/dark shifted 1.5h earlier. On days 12-13, sleep/dark shifted 1h earlier each day. All participants received 1.5h of bright light (8595±834 lux) within 5 minutes of waking on days 12-14. Circadian phase (dim light melatonin onset (DLMO)) was measured on days 8 and 14. The protocol produced a wide distribution of day 14 DLMOs (17:30–4:10). Throughout the study, participants completed visual analog scales before bed to assess mood. An individual’s mood on day 13 (after 2.5-h sleep/dark advance) was subtracted from their baseline week average. Mood change scores were compared between participants with day 14 DLMOs before (n=43; “aligned”) or after (n=31; “misaligned”) day 13 scheduled bedtime. Day 14 DLMOs ranged from 4.5h before to 3.3h after day 13 bedtime. Misaligned teens reported feeling more tense (p=0.02), less energetic (p=0.02), and less calm (p=0.03) than aligned teens. The use of amber-lensed glasses only impacted feelings of calmness; participants wearing glasses were calmer than individuals without (p=0.05). Sleep duration only impacted feelings of sleepiness; the 10-h group felt less sleepy than the 8.5-h (p=0.05), 7-h (p=0.08), and 5.5-h (p=0.02) groups. Changes in happiness, temper, irritability, and concentration did not differ in any analyses. These results indicate that circadian misalignment plays a role in adolescent mood. Subsequent analyses will expand the current focus beyond the evening by evaluating mood at different times of day. R01 HL146772 (Crowley)

Phase preference (PP) measures include several questionnaires, the utility of which may vary with age. Here, we undertake a secondary analysis (Crowley et al., 2014) of PP questionnaires in preteens and teenagers to evaluate associations with circadian phase and sleep patterns. The preteens comprised 44 adolescents (ages 9&10, 19F); teenage cohort includes 74 adolescents (ages 15&16, 34F). PP measures included: Carskadon et al. (1993, Owl), Smith et al. (1989, Smith), and Horne Östberg (1976, HOQ). Preteens completed all three PP measures (Owl, n=43; Smith, n=41; HOQ, n=35); teenagers completed HOQ (n=48) and Smith (n=73). Self-selected sleep-wake schedules at home were monitored via actigraphy for all participants. Sleep onset (SO) was estimated with standard scoring procedures (Acebo et al. 1999). At week’s end, dim light melatonin onset (DLMO) phase was measured from saliva samples. Thirty-eight preteens and 56 teenagers completed questionnaires, actigraphy, and DLMO. Correlations among PP scores were strong in preteens (Owl vs. Smith r=.690, p<.001; Smith vs. HOQ r=.660, p<.001; Owl vs. HOQ r=.418, p=.012) and teenagers (Smith vs. HOQ r=.784, p<.001). PP scores were not correlated with DLMO phase for preteens or teenagers, though HOQ showed a trend (r= -.25, p=.090) with the teenagers. In preteens, PP was not correlated with SO, whereas both Smith (r= -.34, p=.012) and HOQ (r= -.46, p<.001) were correlated with SO in teenagers. PP measures completed by preteens did not correlate with DLMO phase or SO; however, in teenagers, HOQ achieved near significant correlation with DLMO phase, and both PP measures were correlated with SO in the teens. These findings indicate that PP scores in teenagers may be more highly reflective of sleep pattern than circadian phase. Further, we conclude that age-related differences in PP ratings reflect different biobehavioral signals in preteens compared to teenagers. Finally, because of the stronger association of Owl to Smith than to HOQ, we suggest the Smith PP scale may be preferred for use in preteens. R01AA013252; P20GM139743

BACKGROUND:Sleep health is emerging as a public health priority due to its strong associations with several key domains of health. However, most of the existing literature are from studies located in high income settings and may not be representative of low-middle income settings. Leveraging the Modeling the Epidemiologic Transition Study, a study of cardiometabolic disease risk in five diverse African-origin populations, we explored differences in objectively measured sleep behavior across cohorts from Ghana, South Africa, Jamaica, Seychelles, and the United States. METHODS:Data from 809 participants (35-55 years old, 63% women) from the 5 Modeling the Epidemiologic Transition Study research sites were included. Objectively-measured sleep, using actigraphy, was scored according to the criteria of Patel and colleagues. For those with at least 5 nights of valid data, ecological mean sleep onset time, wake-up time, sleep duration, wake after sleep onset, and sleep efficiency were examined. FINDINGS:Adjusted models indicate that sleep onset was earlier in all sites when compared to US (p<.005). Sleep efficiency varied by locations, being lower in participants from Ghana, South Africa, and Jamaica when compared to United States (Ghana β: -3.7, South Africa: -5.8, Jamaica: -1.3, p<.05 for all) and higher in Seychelles when compared to United States (Seychelles β: 1.6; p=.02). Women presented with shorter sleep duration but with higher sleep efficiency. INTERPRETATION:Sleep duration, timing (wake time, midsleep time and sleep onset), and efficiency differ by country and sex, likely driven by socio-economic settings. Understanding sleep patterns in different contexts is needed to make informed and culturally appropriate health recommendations.

Despite extensive research on the effects of sleep restriction on adolescent health, the field lacks experimental methods to study the health effects of mistimed sleep, which is also common among adolescents. This paper describes a novel 3-week experimental protocol that was designed to compare sleep restriction, like what many adolescents experience on school nights, against sleep that meets the recommended duration but is timed to be relatively aligned or misaligned with their circadian phase. Healthy 14-18-year-olds, classified as early ("Lark") and late ("Owl") chronotypes, entered a six-night chronotype-aligned stabilization condition, followed by five nights of sleep restriction, a return to the stabilization schedule, and five nights of healthy sleep duration (HS). During HS, participants were randomly assigned to early-to-bed versus late-to-rise arms, intended to align with or misalign with their circadian phase. Actigraphy monitored sleep, and weekly dim-light melatonin onset (DLMO) assessed circadian phase. Analyses confirmed that the protocol met five key validation metrics related to differential attrition, sleep timing, circadian phase, and experimental induction of HS that is timed to be relatively aligned vs. misaligned with circadian phase. This protocol appears useful for future research into how misaligned sleep patterns, which occur regularly for many adolescents, may impact health.

Staying awake late and rising early for school produces partial sleep deprivation and circadian misalignment in adolescents. We tested whether wearing sunglasses when staying up later than usual can help with daytime performance on reaction time tests that measure attention. Adolescents (n=74, 36F; 14.1-18.0 years-old) completed a 14-day protocol. On days 1-7 (baseline), participants slept at home on individualized 10-h sleep/dark schedules. On days 8-14, they lived in the laboratory. Sleep opportunity remained at 10h (n=11) or was restricted to 8.5h (n=19), 7h (n=23), or 5.5h (n=21) by delaying bedtime 1.5h, 3.0h, or 4.5h, respectively on days 9 and 10. Thirty-three participants wore amber-lensed glasses (transmit ~14% visible light) during the extra 1.5-4.5h time awake before bed; 41 remained in room light without glasses. Participants completed simple reaction time tests on day 5 (baseline) and on day 11 after the two sleep restriction nights. Daily difference-from-baseline scores for lapses (≥ 500 msec), median reaction time, and mean response rate were calculated. Two (glasses vs no glasses) by four (sleep opportunity groups) analyses of variance were completed for each outcome; post hoc t-tests compared outcomes between glasses groups within each sleep opportunity. For all reaction time measures, attention deficits increased as sleep opportunity shortened [sleep opportunity main effect: p’s< 0.01]; however, when participants wore the amber-lensed glasses these deficits were no longer observed and remained close to baseline levels [glasses main effect: p’s< 0.001]. With 8.5-h, 7-h and 5.5-h sleep opportunities and amber-lensed glasses, adolescents showed fewer lapses, faster reaction times and greater throughput (mean response rate) compared to no glasses (p values ranged from 0.001 to 0.009). Reducing the intensity of evening light when staying up later by wearing amber-lensed glasses protected against daytime attention deficits in adolescents. Evening light has a delaying effect on the circadian clock. Reducing the intensity of this light may reduce phase delays and, in turn, circadian misalignment. Therefore, this simple intervention has potential to improve academic performance and enhance safety during the school commute. Whether these improvements are maintained after more than two nights of sleep restriction is to be determined. R01 HL146772 (Crowley)

Most adolescents experience circadian misalignment due to a mismatch between their delayed circadian physiology and early school start times. This review considers cognitive functioning and school behaviors, associated with delayed circadian timing and subsequent circadian misalignment. Insights from special populations and lessons learned during the COVID-19 pandemic are also highlighted. When adolescents experience circadian misalignment, there are implications for their cognitive functioning and school behaviors. Findings emphasize the need to align school start times with adolescents’ delayed circadian physiology to provide more conducive learning environments. Indeed, adaptations in learning schedules during the COVID-19 pandemic may have benefited adolescents’ sleep. Ultimately, adolescents who experience greater circadian misalignment and those that have later sleep preferences are most vulnerable to cognitive and school-related difficulties. Establishing later school schedules or developing individualized strategies to shift the circadian system earlier could help improve cognitive functioning and school behaviors at a critical time in youth’s development.

Morning bright light is most effective in shifting circadian phase earlier; however, our adolescent PRC to bright light unexpectedly showed phase advances in response to bright light timed in the afternoon (Crowley & Eastman, JBR 2017 32(4):334-344). The current ongoing study is examining phase shifts in response to three bright light schedules: morning bright light (MBL), afternoon bright light (ABL), and a combination of morning and afternoon bright light (MBL+ABL) compared to a room light (RL) control. So far, 51 adolescents (27F) aged 18.3-20.9 years completed a two-week study. During the first week, participants followed a rigid individualized 9-hour sleep schedule at home. On days 8-13, participants lived in the lab. On day 8, participants completed initial phase assessments to determine Dim Light Melatonin Onset (DLMO). On day 9, they followed their baseline sleep schedule. On days 10-12, their sleep (dark) advanced 1-hour earlier each day. DLMO was measured again on day 13. Three groups received bright light from light boxes (8542±820 lux, 45-min exposures totaling 3h/day) on days 10-13 at the following times: all 4 exposures started 5h after wake (ABL; n=18), all 4 exposures started within 5 minutes of wake (MBL; n=7), or 2 exposures started within 5 minutes of wake and 2 exposures started 7 h after wake (MBL+ABL; n=8). A fourth group (RL; n=17) completed the same protocol but remained in room light (45±22 lux). DLMO advanced in all groups (RL: 0.8±0.5h; ABL: 0.9±0.6h; MBL: 1.9±0.9h; MBL+ABL: 0.9±0.6h). A main effect of group (F(3,46)=5.86, p=.002) was explained by the MBL group showing larger phase advances compared to all other groups (p’s<.02). Gradually advancing the sleep (dark) schedule by 1h/day for 3 days advanced the circadian clock of adolescents by ~1h. Adding 3h of morning bright light increased the advance to ~2h. Afternoon bright light, even when combined with morning light, did not significantly change the phase advance obtained by advancing sleep (dark) only. R01 HL151512 (Crowley)

Adolescence is an important developmental time, with specific sleep health needs for optimal functioning. Juveniles residing in correctional facilities have important developmental milestone to hit, which requires adequate sleep health and hygiene. However, very little is known about the sleep health of juveniles residing in justice facilities. Even less is known regarding the day-to-day operation management of juvenile justice facilities and how decisions concerning sleep are made (e.g., light, noise, schedule, and medication). We deployed a qualitative method for collecting information from three key personnel at youth correctional facilities: superintendents, night staff, and medical staff. Semi-structured interviews were transcribed and coded to understand the management of youth facilities with an emphasis on protocols that affect sleep during detention. Policy recommendations are discussed.

Study Objectives Little is known about sleep health among staff in the US juvenile justice system. Poor sleep health is associated with negative mental and physical health, which may impact daily interactions and treatment of detained youth. The current study explored sleep-wake patterns and sleep health knowledge of Department of Juvenile Services (DJS) staff in Maryland (MD).Methods DJS Staff (N = 218) were invited to complete a survey that queried staff on their own sleep-wake patterns, job role and schedule, and knowledge of youth sleep needs. Descriptive analyses and multivariate analyses of variance (MANCOVA) were conducted to summarize workers' sleep-wake patterns and examine differences by staff position and schedule.Results Fifty-one percent of staff served as RAs who directly supervise the youth. Just over half (55%) worked in detention and 45% in treatment facilities. Staff reported sleeping 7.24 hours (SD = 4.10) on workdays and 8.59 hours (SD = 2.69) on non-workdays. RA staff working night/rotating versus day shifts reported the most sleep irregularity with larger weekend oversleep times. A little more than half of the staff (53.9%) were knowledgeable regarding youth sleep health with differences by position type.Conclusions Findings show that DJS staff are meeting recommended sleep duration guidelines but are still experiencing sleep schedule and time in bed irregularity. Knowledge variability of youth sleep health across staff may necessitate focused educational programming. Overall, this study may inform future development and prioritization of sleep and circadian health interventions and educational campaigns for staff who work with detained juveniles. This paper is part of the Sleep and Circadian Health in the Justice System Collection.

Abstract Introduction Puberty encompasses a multitude of physical, hormonal, and psychological changes. Additionally, adolescents also experience a delay in circadian rhythm, increasing their vulnerability to insufficient sleep. Even though puberty and sleep share a strong association, few studies have explored the shift in sleep after the onset of menstruation (i.e., menarche). This longitudinal, multimethod study examines sleep health parameters before and after menarche among a community sample of females. Methods Biological females (n = 47; 74% white) from a Midwestern metropolitan area were recruited as part of a larger study on menarche. Participants completed two lab visits: Time 1 (T1: prior to menarche; Mage = 11.3, SDage = 1.0) and Time 2 (T2: 3-9 months post-menarche; Mage = 12.6, SDage = 1.0). An actigraph wristwatch was worn for 7 days following each lab visit to measure dimensions of sleep health including bedtime, wake time, time in bed, and total sleep time. Phillips Actiware Software was used to score actigraphy data; SPSS 27 was used to perform paired sample t-tests to explore changes over time. Results Following menarche, there were significant differences in several parameters of sleep health including: 1) a 1.7-hour delay in bedtime (T1M = 21:45, T1SD = 1:02; T2M = 23:29, T2SD = 3:07, t(46) = (3.37), p = (.002), 2) less time in bed (T1M = 542, T1SD = :48; T2M = 518, T2SD = :58), t(46) = (2.67), p (.01), and 3) a 27-minute decrease in total sleep time (T1M = 482, T1SD = :40; T2M = 455, T2SD = :51), t(46) = (3.32), p = (.002). Overall, these results indicate participants are sleeping significantly less in the months following menarche. Conclusion Findings confirm that biological females experience changes in sleep behaviors within 9 months of menarche. Chronic sleep loss and daytime sleepiness worsens throughout adolescence, furthering the risk for cognitive and behavioral issues. Data collection is ongoing and future analyses will examine sleep variability across this transitional point in development. This is one of the first studies to examine the sleep-wake experience of females across menarche. Support (if any)

Study Objectives:Our sleep extension intervention in adolescents showed that gradually shifting weekday bedtime earlier plus one weekend of morning bright light advanced circadian phase and increased weeknight sleep duration. Here, we examine at-home maintenance of these changes. Methods:Fourteen adolescents (15.3-17.9 years; 7 female) completed a 7-week study. After usual sleep at home (2-week baseline), intervention participants (n = 8) gradually advanced weekday bedtime (1 hour earlier than baseline during week 3; 2 hours earlier in week 4) and received bright light (~6000 lux; 2.5 hours) on both mornings of the intervening weekend. During three maintenance weeks, intervention participants were instructed to maintain their school-day wake-up time on all days, keep their early week four bedtimes, except on weekends when they could go to bed up to 1 hour later, and get a 2.5-hour light box exposure within 5 minutes of waking on one morning (Saturday or Sunday) of both weekends at home. Control participants (n = 6) slept as usual at home and did not receive weekend bright light. Dim light melatonin onset (DLMO) was measured after the 2-week baseline, 2-week intervention, and 3-week maintenance in all participants. Actigraphic sleep-wake was collected throughout. Results:After the 2-week intervention, DLMOs advanced more compared to control (37.0 ± 40.0 minutes vs. -14.7 ± 16.6 minutes), weekday sleep duration increased by 69.7 ± 27.8 minutes and sleep onset was 103.7 ± 14.2 minutes earlier compared to baseline. After three maintenance weeks, intervention participants showed negligible DLMO delays (-4.9 ± 22.9 minutes); weekday fall-asleep times and sleep durations also remained stable. Conclusions:Early circadian phase and extended sleep can be maintained with at-home weekend bright light.

Study objectives: Atopic dermatitis (AD) is a chronic inflammatory skin disorder in children. AD worsens at night, particularly in severe disease. Low light exposure contributes to inflammation, poor sleep and misalignment between circadian (24-hour) rhythms (biologic clocks) and social clocks (weekday-versus-weekend sleep timing), but has not been evaluated in AD. Our objective was to perform a cross-sectional study to determine whether there is an association between AD severity, recorded light exposure (RLE), and sleep measures in participants with AD and healthy controls.Methods: Secondary data analysis from two prospective observational studies of 74 participants ages 5-17 y.o. with severe AD compared to others (healthy controls and mild/moderate AD). Participants wore actigraphy watches for at least 1 weekday and one weekend. Rest/activity and RLE (lux) were obtained from the watches and were analyzed to estimate duration and quality of sleep/light exposure.Results: Participants (n=74) were on average 10.9 +/- 3.6 y.o., with 45% female, 17% no AD, 27% mild, 32% moderate, and 24% severe AD. On weekends, severe AD participants versus others fell asleep at a similar time (23:52 +/- 1:08 vs. 23:40 +/- 1:29 mean clock-time hours +/- SD; p=0.23), had similar sleep onset latency (8.2 +/- 8.7 vs. 12.7 +/- 16.9 minutes; p=0.28), but woke later (09:12 +/- 1:04 vs. 08:13 +/- 1:14 minutes; p<0.01) resulting in a later sleep-midpoint (04:32 +/- 0:53 vs. 03:49 +/- 1:08 minutes; p=0.02). Severe AD participants had lower levels of daytime RLE than others (mean-over-all-days: 1948.4 +/- 2130.0 vs. 10341.3 +/- 13453.8 lux; p=0.01) and throughout seasons, weekdays, or weekend, yet had similar nighttime RLE.Conclusion: Severe AD is characterized by low RLE and sleep disturbance. Low RLE could potentially induce circadian misalignment, contributing to inflammation and worse disease in severe AD. Low RLE can also reflect altered lifestyle and behavior due to atopic disease burden. Prospective studies are needed to test causality and the potential of bright light as an adjuvant therapy for severe AD.

IntroductionThe propensity for sleep shifts later as puberty progresses. The present analysis examines whether the circadian-dependent wake maintenance zone, or forbidden zone for sleep observed in the evening just before habitual bedtime is more pronounced in late to post-pubertal adolescents compared to adults and may partly explain late sleep onset in maturing adolescents.MethodsForty four healthy late/post-pubertal adolescents (aged 14.3–17.8 years, 23 female) and 44 healthy adults (aged 30.8–45.8 years, 21 female) participated in an ultradian light/dark protocol for 3 days cycling between 2-h wake periods (~20 lux) and 2-h nap periods (~0 lux) without external time cues. The dim light melatonin onset (DLMO), a measure of circadian phase, was measured immediately before the ultradian protocol by sampling saliva every 30 min in dim light. Wrist actigraphs were used to assess sleep onset latency and total sleep time during the naps that occurred during the ultradian sleep/wake schedule. Sleep episodes were grouped into 2-h bins relative to individual DLMOs (28–56 naps/bin). Sleep onset and total sleep time were compared between adolescents and adults as well as between males and females within each age group.ResultsAdolescents took significantly longer to fall asleep compared to adults during naps that occurred in the 4 h window surrounding the DLMO [2h before DLMO t(50) = 2.13, p = 0.04; 2 h after DLMO t(33) = 3.25, p = 0.003]. Adolescents also slept significantly less than adults during naps that occurred in the 4-h window surrounding DLMO [2 h before DLMO t(51) = −2.91, p = 0.01; 2 h after DLMO t(33) = −1.99, p = 0.05]. Adolescent males slept less than adolescent females in naps that occurred in the 2 h window after the DLMO [t(14) = −2.24, p = 0.04].DiscussionCompared to adults, late/post-pubertal adolescents showed greater difficulty falling asleep and maintaining sleep around the time of their DLMO, which usually occurs a few hours before habitual sleep onset. A greater amplitude in the circadian-driven forbidden zone for sleep could be an additional physiological mechanism explaining why maturing adolescents find it difficult to fall asleep early, increasing the risk for restricted sleep in the context of early school start times.

South Africans living in low socioeconomic areas have self-reported unusually long sleep durations (approximately 9–10 h). One hypothesis is that these long durations may be a compensatory response to poor sleep quality as a result of stressful environments. This study aimed to investigate whether fear of not being safe during sleep is associated with markers of sleep quality or duration in men and women. South Africans (n = 411, 25–50 y, 57% women) of African-origin living in an urban township, characterised by high crime and poverty rates, participated in this study. Participants are part of a larger longitudinal cohort study: Modelling the Epidemiologic Transition Study (METS)–Microbiome. Customised questions were used to assess the presence or absence of fears related to feeling safe during sleep, and the Epworth Sleepiness Scale, Pittsburgh Sleep Quality Index (PSQI) and Insomnia Severity Index were used to assess daytime sleepiness, sleep quality and insomnia symptom severity respectively. Adjusted logistic regression models indicated that participants who reported fears related to safety during sleep were more likely to report poor sleep quality (PSQI > 5) compared to participants not reporting such fears and that this relationship was stronger among men than women. This is one of the first studies outside American or European populations to suggest that poor quality sleep is associated with fear of personal safety in low-SES South African adults.

Abstract Introduction Previously, we reported that evening room light and sleep restriction due to late bedtimes can reduce the expected circadian phase advances in response to morning bright light in adolescents or can shift rhythms later (delay). Here, we examine whether reducing evening light with sunglasses can rescue this effect and help rhythms shift earlier (advance). Methods So far, 57 adolescents (29 females; 14.1-18.0 years) completed a 14-day protocol. On days 1-7, they kept individualized 10-h sleep/dark schedules at home. On days 8-14, they lived in the laboratory. On day 8, we measured their Dim Light Melatonin Onset (DLMO). On days 9-10, they went to bed 1.5h, 3.0h, or 4.5h later. Wake time was unchanged; thus, sleep opportunity was 8.5h, 7h, and 5.5h, respectively. In each sleep restriction group, participants remained awake in room light (104±11 photopic lux; 8.5h n=9; 7h n=12; 5.5h n=8) or wore amber-lensed sunglasses (transmitted 14% of light and 10% of short wavelength light; 8.5h n=5; 7h n=4; 5.5h n=9). A control group (n=10) kept their baseline bedtime, so was not exposed to additional evening light or sleep restriction. On days 11-13, sleep/dark was gradually shifted earlier, and adolescents received three 30-min exposures of bright light (8599±885 photopic lux) upon waking. Final DLMO was measured on day 14. Multiple linear regression examined sleep restriction dose, sunglasses, and the dose-by-sunglasses interaction effects on phase shift. Results The control group advanced +2.1±1.0h. Greater sleep restriction reduced phase advances (dose: β=-1.31, p<.001), but evening sunglasses attenuated this effect (dose-by-sunglasses: β=1.01, p<.05). When sleep/dark was 8.5h, sunglasses produced larger advances (+1.3±0.6h) compared to no sunglasses (+0.5±0.6h) [t(12)=-2.4, p=.04]. Compared to no sunglasses, evening sunglasses prevented delays when sleep/dark was 7.0h (-0.8±1.1h vs +0.5±0.9; t(14)=-2.2, p=.04) and 5.5h (-2.6±1.2h vs +0.0±1.1h; t(15)=4.8, p<.001). With evening sunglasses and 7.0h or 5.5h sleep/dark, advances were still smaller than control (p’s≤.01). Conclusion Reducing evening light with amber-lensed glasses during a phase-advancing protocol with morning bright light facilitates larger advances or prevents delays in adolescents. However, ≤7h sleep/dark reduces phase advances in response to morning bright light even when evening light is reduced. Support (if any) R01HL146772 (Crowley)