Abstract Introduction Sleep has been shown to slow the rate of forgetting compared to a period of wake, but no studies to date have explored whether total sleep deprivation (TSD) also moderates the rate of forgetting compared to rested wakefulness. We investigated how well participants could retain information throughout the day on the day before, during, and the day after TSD. Methods 16 individuals (ages 18-38; mean age 25.8y ± 7.9y; 8 males, 8 females) completed a 4-day/3 night in-laboratory sleep study. Following a baseline 10h sleep opportunity, participants underwent a 38h TSD period followed by a 10h recovery sleep opportunity. Sleep periods were 22:00-08:00. Each day in the lab, participants were presented with a new list of words at around 09:30 and were asked to recall the words immediately, 4h, and 6h after study. Results A mixed-effects ANOVA with fixed effects of session (1, 2, or 3), test (immediate, 4h, or 6h later), and their interaction, with a random intercept over subjects, revealed a significant effect of test (F=22.23, p < 0.001), but no significant effect of session or session x test interaction (F< 2.26, p>0.10). Relative to immediate testing, recall decreased 27.3% at 4h and 26.1% at 6h. Post-hoc tests on the effect of test and the session x test interaction revealed that accuracy on the immediate recall test was better than accuracy 4h or 6h later (p< 0.001), but only in sessions 2 and 3. The pattern was similar in session 1, but not statistically significant. Conclusion In line with the established forgetting curve, participants had better memory for the words at immediate recall than at 4 and 6h later. Contrary to expectations, sleep deprivation did not appear to exacerbate the forgetting rate, even though participants had been awake for 25.5h at study. Future research should include tests at greater delays. Additionally, it should investigate whether chronic and acute sleep deprivation have differential effects on the forgetting rate. Support (if any) Google

Abstract Introduction Menstruation has been argued to affect the cognitive performance of women, but the available research is limited and its findings remain mixed. In this study, we investigated the effect of menstrual phase on one aspect of cognitive performance (i.e., sustained attention) throughout a period of total sleep deprivation (TSD). Methods A total of 31 healthy women (10 controls, mean age 27.6y ± 6.6y; 21 TSD, mean age 28.2y ± 5.9y) completed one of two highly controlled in-laboratory sleep studies each with a 38h TSD period or matching control condition with a 10h sleep opportunity. Performance was measured using a 10min version of the Psychomotor Vigilance Task (PVT). Test bouts administered at identical times in each study were used for analysis. Onset of menses and menstrual duration was determined using self-reported cycle length and regularity, most recent period, and next predicted period. The follicular phase was defined as the beginning of menstruation to the start of ovulation. Ovulation was defined as 14 days prior to the predicted date of the next period. The luteal phase was defined as the start of ovulation to the onset of menstruation. In our sample, 11 women were in the follicular phase, 9 were in the luteal phase, and 11 were menstruating. No women were in the ovulation phase during experimentation. Results There was a significant increase in both the number of lapses of attention (RT>500ms) and mean reaction time in the TSD group (all F>4.50, P< 0.01). Performance on the PVT was not significantly affected by menstrual phase (all F< 1.3, P>0.30). Conclusion As expected, TSD degraded performance on the PVT as evidenced by an increase in the number of lapses and mean RT. Although we did not have any women in the ovulation phase during experimentation, it does not appear that menstrual phase impacted performance on a sustained attention task beyond the effects of TSD itself. Future research will examine the impact of menstrual phase on subjective ratings of sleepiness and cognitive tasks measuring other facets of cognition administered throughout TSD. Support (if any) ONR grant N00014-13-C-0063 and USAMRDC W81XWH-20-1-0442

OBJECTIVE:Individuals with type 1 diabetes (T1D) experience poor subjective sleep quality and are at an increased risk for obstructive sleep apnea (OSA), yet the contribution of diabetes-related characteristics remains unclear. METHODS:This study evaluated associations between diabetes characteristics, affect, and sleep in 210 adults with T1D (M age=45.6 years, SD=15.6; 52.9% female) from the Glycemic Variability and Fluctuations in Cognitive Status in Adults with Type 1 Diabetes study. Participants completed measures of sleep quality (Pittsburgh Sleep Quality Index), OSA risk (STOP-Bang), and affect (Generalized Anxiety Disorder-7, Perceived Stress Scale, Patient Health Questionnaire-8). Blinded continuous glucose monitoring data were collected for up to 20 days (M=18.8, SD=2.5). Bivariate correlations and hierarchical regressions evaluated relationships among subjective sleep quality and OSA risk with affect and diabetes variables. RESULTS:Overall, 47% and 37% of participants scored above clinical cutoffs on the Pittsburgh Sleep Quality Index and STOP-Bang. Those with microvascular complications reported poorer sleep quality (β=0.13, t(188)=2.23, p=0.03). Microvascular complications (β=0.15, t(200)=3.10, p=0.002) and cardiovascular disease (β=.014, t(200)=2.89, p=0.004) were associated with a higher OSA risk. Continuous glucose monitoring metrics were not associated with the outcomes. CONCLUSIONS:In this sample of adults with T1D, nearly half reported poor sleep quality, and more than one-third were at elevated risk for OSA, with microvascular complications associated with poorer sleep quality and both microvascular and cardiovascular complications associated with a higher OSA risk. Aggregate glycemic metrics were not associated with sleep outcomes. Future research should evaluate the relationships between daily variation in sleep and glucose to inform targeted clinical interventions.

Chronic pain is a prevalent condition often treated with opioid medication. Opioids, such as methadone, are also used to treat opioid use disorder (OUD). As central nervous system depressants, opioids can exacerbate sleep problems, particularly those due to respiratory issues. This study investigates sleep quality and respiratory distress measures in adults with Chronic Pain (CP) receiving opioid medication and those with OUD receiving methadone. Adults (18years+) in the CP group had a “moderate” daily level of pain (≥5 on a 0-10 Numeric Pain Scale [NPS]) and were prescribed opioids. Adults (18years+) in the OUD group were enrolled in an opioid treatment program and taking methadone daily. All participants rated their pain intensity and interference scales using the PROMIS. NightOwl mini disposable home sleep tests (Ectosence, Leuven, Belgium) were used to record total sleep time (TST), time in bed (TIB), sleep efficiency (SE), and apnea-hypopnea index (AHI) for all participants. In total, N= 17 (12 women) provided data, N=9 (6 women 60y±12y [Mean ± SD]) in the CP group and N=8 (6 women 45.5y±12.3y) in the OUD group. Average pain intensity and interference T-scores were 69.4±8.1 and 69.72±6.5 for CP and 59.6±7.9 and 58.4±9.1 for OUD (t>-2.45, p< 0.02). CP participants spent 8.6±2.8h TIB and 5.08±2.3h asleep, resulting in an SE of 61.4±24.9% and a predicted AHI of 6.6±6.4, indicative of mild sleep apnea. OUD participants spent 8.27±2.93h TIB and 4.19±2.03 hours asleep, resulting in an SE of 54.86±19.35% and a predicted AHI of 21.4±27.7, indicative of moderate sleep apnea (all p>0.05). Although the CP group reported significantly higher pain intensity and interference scores, objective sleep data showed the OUD group had higher measures of sleep apnea and worse sleep efficiency. This preliminary data warrants further exploration of sleep quality and respiratory status among adults with chronic pain and OUD receiving maintenance treatment with opioids.

Maternal sleep disruption is common throughout the first postpartum year, and particularly severe in the first postpartum week. As nearly 75% of all pregnancy-related maternal deaths occur between postpartum days 1 and 7, maternal sleep disruption after birth could be a risk factor for maternal morbidity and mortality. Here, in a sample of first-time mothers, we quantified objectively measured sleep during the first 7 postpartum days. N=41 first-time mothers (ages 26–43y) recorded their wrist activity (Fitbit) continuously across postpartum days 1-7. Sleep data were analyzed in 5min bins, using ≥10min consecutive sleep and wake as criteria for onset and offset of sleep periods, respectively. Sleep duration (including naps) and longest stretch of sleep (LSS) were calculated for each 24h day. Off-wrist detection was based on absence of heart rate data and controlled for in analyses. The incidence of acute total sleep deprivation (>24h without sleep) was 20% (n=8) on day 1, 20% (n=8) on day 2, and 10% (n=4) on days 3-7. Total 24h sleep duration ranged from a low of 2.7±2.2h (mean±SD) on day 1 to a high of 5.0±2.0h on day 7. LSS ranged from a low of 1.7±1.4h on day 1 to a high of 2.9±1.2h on day 7. These results show that acute total sleep deprivation is common in new mothers after giving birth. The results also show severe chronic sleep restriction during the first postpartum week, with the average first-time mother in our sample obtaining no more than 5h total sleep duration per 24h and consolidated sleep periods shorter than 3h – much less even during the first few postpartum days. Our findings provide objective evidence of considerable maternal sleep disruption, making sleep disruption a plausible contributor to maternal health risks in the first postpartum week. Limitations of the study include that the sample was predominantly white, relatively affluent, and generally in good health. The sleep data nonetheless suggest that studies should investigate whether there is a causal relationship between maternal sleep disruption and health risks, and whether interventions to protect maternal sleep during the first postpartum week may improve health outcomes. trackthatsleep LLC

Work in healthcare routinely involves night and rotating shifts, early morning starts, and variable call schedules, resulting in short and/or disrupted sleep. We conducted a laboratory-based experiment in both normal sleepers and those with chronic insomnia to better understand implications of total sleep deprivation (TSD) on decision making using a reversal learning task (i.e., go/no go; GNG) with unannounced reversal contingences. 28 individuals completed the sleep study, 15 with sleep-onset insomnia (ages 22-40, 11 females), of whom 7 underwent TSD, and 13 normal sleeper controls (ages 22-40, females), of whom 7 underwent TSD. The insomnia group met International Classification of Sleep Disorders (3rd edition; ICSD-3) criteria for chronic insomnia with no other clinically relevant condition contributing to their sleep disturbance. Subjects were in the laboratory for 5 days/4 nights. The first 2 days and nights were baseline days, each with a 10h opportunity for sleep (22:00—08:00). This was followed by 38h of TSD or another nighttime sleep opportunity for the control group. The last day was a recovery day, with a 10h sleep opportunity for all participants. Unique versions of a GNG task were administered in random, counterbalanced order at baseline and 24h later during TSD (or 2h awake in the control group). There was a significant main effect of day (F1,69= 10.37, P=0.002), condition (F3,69 = 5.94, P = 0.001), phase (F1,69 = 10.03, P = 0.002), and day by condition by phase interaction (F10,69 = 4.73, P < 0.001). In general, participants performed better before TSD and pre-reversal. While both TSD groups showed poorer performance post-reversal during TSD, only the TSD insomnia group showed relatively intact performance pre-reversal. TSD led to significant impairment on a reversal learning task for both normal sleepers and those with chronic sleep-onset insomnia. This has important implications for healthcare workers who are routinely exposed to disrupted sleep which may leave them more susceptible to making potentially life-threatening errors. Hyperarousal, the widely accepted underlying mechanism perpetuating insomnia, may provide those chronically exposed to sleep loss some level of protection on decision making tasks compared to normal healthy sleepers during TSD. ONR grant N00014-13-C-0063

Acute total sleep deprivation (TSD) is common due to extended work hours or other time demands. In controlled laboratory settings, individuals exposed to TSD display highly reproduceable profiles of neurobehavioral impairment. In real-world settings, however, TSD is often combined with substance intake, including alcohol and cannabis. In this pilot study, we investigated the effect of substance intake on the neurobehavioral response to TSD. Six healthy individuals (ages 23-37y, all male) completed two 24h in-laboratory study visits separated by ≥1 week. During each visit, they were kept awake from 15:00 until 06:00 the next day, and they completed a 10min PVT every 2-3h. This was followed by an 8h recovery sleep opportunity. For the second visit, participants were randomized – three per group – to receive oral administration of cannabis at 22:30 (10mg) or alcohol at 23:30 (peak blood alcohol concentration of 0.043±0.006% at 00:13, decaying to 0.005±0.007% by 03:55). PVT mean RT, number of lapses (RT>500ms) and false starts were analyzed using mixed-effects regression. As expected, PVT performance deteriorated through the second half of the extended waking period (after midnight), with slower mean RT, more lapses, and more false starts (F>3.6, P< 0.002). Alcohol increased mean RT (F=4.9, P=0.002) and lapses (F=7.7, P< 0.001), but did not significantly affect false starts (F=1.0, P=0.41). Cannabis did not significantly affect mean RT (F=0.9, P=0.46), lapses (F=1.1, P=0.37), or false starts (F=0.8, P=0.56) during TSD. Alcohol exacerbated neurobehavioral impairment during TSD, even at blood alcohol levels below the legal limit of most countries and states, reaching statistical significance despite our small sample size. Cannabis at the dose provided did not significantly further degrade PVT performance during TSD. Importantly, no attempt was made to make the alcohol and cannabis doses equipotent; therefore, these results should not be interpreted as evidence of their relative effect sizes or safety profiles. Yet, our findings provide preliminary evidence suggesting that commonly used drugs such as alcohol may amplify neurobehavioral impairment from sleep loss, which may have critical implications for automobile driving and other safety-sensitive activities. National Safety Council

Maternal sleep is disrupted during the postpartum period, but the nature and severity of sleep disruption has not been adequately documented. Studies of sleep duration have reported only modest sleep loss beyond the first week after giving birth, leaving first-time mothers poorly prepared for what kind of sleep disruption to expect. Using sleep data from first-time mothers’ personal wearables (Fitbit), we examined sleep duration and the longest stretch of sleep (LSS) – a sleep consolidation metric commonly used for infant sleep – to quantify maternal sleep during the first 13 postpartum weeks. N=41 first-time mothers (ages 26–43y) provided their sleep/wake wearable data from a full year before childbirth to the end of the first postpartum year. Sleep data were analyzed in 5min bins, using ≥10min consecutive sleep and wake as criteria for onset and offset of sleep periods, respectively. Off-wrist detection was based on absence of heart rate data. Daily sleep duration and LSS were calculated for each 24-hour day and compared between the first 13 postpartum weeks and the equivalent days of the prior year (preconception baseline). During postpartum week 1, daily sleep duration was 4.4±0.2h (mean±SEM) compared to 7.8±0.2h at preconception baseline. Daily LSS was 2.2±0.2h versus 5.6±0.2h preconception. 31.7% of participants went >24h without sleep. Across postpartum weeks 2-7, daily sleep duration increased to 6.7±0.1h versus 7.7±0.1h preconception. However, daily LSS stayed low at 3.2±0.1h versus 5.5±0.1h preconception. Across postpartum weeks 8-13, daily sleep duration was 7.3±0.1h versus 7.9±0.1h preconception. Yet, daily LSS was still reduced at 4.1±0.1h versus 5.6±0.1h preconception. All differences were significant (F>29.8, P< 0.001). Sleep duration was greatly reduced during the first postpartum week, but gradually returned to near baseline levels thereafter. However, sleep consolidation, as captured by LSS, stayed considerably below preconception baseline throughout the first 13 postpartum weeks. This suggests that in postpartum weeks 2-13, sleep discontinuity – more so than sleep loss – contributed most prominently to first-time mothers’ sleep disruption. Sleep discontinuity may be a risk factor and intervention target for postpartum depression and other postpartum-related health issues. trackthatsleep LLC

Autonomic hyperarousal is thought to be a pathogenic mechanism in chronic insomnia, characterized by elevated heart rate (HR) and reduced heart rate variability (HRV), indicating greater sympathetic nervous system activity. Spectral analysis links HR changes to central nervous system activity, with low frequency (LF) reflecting sympathetic activity and high frequency (HF) indicating parasympathetic activity. This study examined differences in HR and HRV between individuals with insomnia and healthy control sleepers during total sleep deprivation (TSD). 7 individuals with chronic sleep-onset insomnia (M=29.0y, SD=6.2y, 6 females) and 7 age-matched healthy control sleepers (M=29.0y, SD=6.6y, 4 females) completed a 5-day (4-night) laboratory study. After an adaptation day and baseline day (each 10h time in bed, 22:00-08:00), participants underwent 38h of TSD followed by a recovery day (10h time in bed; 22:00-08:00). They were fitted with a 5-lead Holter monitor (DMS 300-3A Digital Holter Recorder) for continuous ECG recording. Mean HR was extracted in 5-minute bins, and power in the LF (0.04–0.15 Hz), HF (>0.15–0.40 Hz), and the LF/HF ratio components of HRV were determined by spectral analysis. Data were analyzed using mixed-effects ANOVA with a fixed effect of group (insomnia vs. controls) and period (daytime 08:00-22:00 vs nighttime 22:00-08:00) and their interaction with a random effect of subject on the intercept. Overall, the insomnia group exhibited lower LF power (F1,1299=5.38, P=0.021) or greater parasympathetic activity as compared to the healthy control sleepers. There was a significant effect of period for all measured parameters, indicating a shift toward greater parasympathetic activity at night (all F>17.7, P< 0.001). However, the significant period-by-group interactions revealed that this shift was less pronounced in the insomnia group for all measured parameters (all F>2.24, P< 0.02). Our findings suggest an interplay of homeostatic, circadian, and sleep related effects on sympathovagal balance, with more pronounced sympathetic effects observed in the healthy controls compared to individuals with insomnia. This dampened response in the insomnia group may be a result of their chronic sleep deficiency, making them less responsive to sleep and TSD due to underlying hyperarousal. ONR grant N00014-13-C-0063

Postpartum maternal sleep is disrupted, even after stabilizing following the first three months. Evidence suggests that breastfeeding women wake more often during the night than those who use infant formula, while sleep duration is similar. Here we investigated the effects of breastmilk versus formula feeding on maternal sleep across postpartum weeks 14-52. N=41 first-time mothers (26-43y) recorded their sleep (Fitbit) and reported infant milk type – breastmilk or formula – with 16 participants also logging daily infant feeding events (mobile app). Daily sleep duration and continuity (LSS: longest stretch of sleep) were assessed by 24h day. Effects of milk type were analyzed with mixed-effects ANCOVA, controlling for general trends across days. Mean daily sleep duration was 7.4h throughout postpartum weeks 14-52, whereas LSS increased from 4.3h to 5.3h. In week 14, 64.7% of participants fed breastmilk, 17.1% formula, and 18.2% mixed. By week 52, 47.2% fed breastmilk, 43.5% formula, and 9.3% mixed. Milk type did not affect sleep duration (F=2.6, P=0.11), but those feeding breastmilk had 0.4h shorter LSS than those feeding formula or mixed (F=10.2, P=0.001). Among those logging feeding events, milk type did not affect sleep duration (F=0.8, P=0.38), but LSS was shorter by 6.2min per feeding event for breastmilk compared to formula (F=12.0, P< 0.001). There were no significant relationships with breastmilk pumping, breastfeeding versus breastmilk bottle feeding, infant weight gain, or infant sleep training. On average there were 9.1 daily feeding events for exclusive breastmilk feeding, compared to 7.3 for formula or mixed (F=10.9, P=0.001). In this predominantly white and relatively affluent sample, maternal sleep duration in postpartum weeks 14-52 did not vary by infant milk type. However, participants who fed their babies formula had fewer daily feeding events and greater sleep continuity than participants who exclusively fed their babies breastmilk. As there was no effect of breastfeeding versus breastmilk bottle feeding, use of infant formula per se may have led to increased maternal sleep continuity. Whether a difference in infant sleep continuity is involved remains to be investigated. Regardless, current recommendations of exclusive breastmilk feeding may come at a cost to maternal sleep. trackthatsleep LLC

Sleep problems occur at a higher rate in autism spectrum disorder (ASD) than in typical development (TD). Polysomnography (PSG) is the gold-standard for collecting objective sleep data, but it is labor-intensive, intrusive, requires patients to stay overnight in an unfamiliar environment, and typically limited to one night of recording decreasing its validity and scalability to larger cohorts of ASD individuals. We tested a non-contact biomotion sensor, the SleepScore Max (SSM) to determine the acceptability of longitudinal use and feasibility of estimating sleep characteristics in ASD and TD youth in the home. 7 ASD (M=13.0y, SD=2.2y, with community ASD diagnosis) and 6 TD (M=13.3y, SD=2.3 y) individuals completed 2 weeks of at-home sleep monitoring using the device. Participants were between age 8-15 years, living with a caregiver, sleeping alone, and >80lbs. Caregivers completed an acceptability questionnaire at the end of the study. Sleep metrics were analyzed using repeated-measures ANOVA with fixed effects of group (ASD vs TD), day, and their interaction. Pairwise comparisons of standard deviations determined differences in night-to-night variability between groups. The ASD group exhibited more deep sleep (F=21.52, p< 0.001), longer sleep onset latency (F=5.13, p=0.025), and increased total wake time (F=4.56, p=0.034), whereas the TD group had more light sleep (F=5.58, p=0.019), and total interruptions (F=12.79, p< 0.001). Pairwise comparisons determined that the ASD group had greater night-to-night variability in REM sleep (t=-2.7, p=0.021). Caregivers reported the device to be highly acceptable (M=6.44; SD=1), reasonable to use (M=6.31; SD=1) and causing little to no discomfort (M=1.7; SD=1.2) on a 1-7 Likert scale. Our results agree with previous PSG studies showing an increase in sleep onset latency and an increase in deep sleep in autistic individuals. These results provide preliminary evidence that a non-contact biomotion sensor can accurately measure and differentiate sleep between ASD and NT individuals in a naturalistic and sensory friendly manner with high levels of acceptability and low participant burden. Non-contact sensors show great promise for long-term longitudinal characterization of sleep in the home to evaluate outcomes of sleep focused interventions.

Conventional sleep staging is the standard for diagnosing most sleep disorders but has little diagnostic yield for chronic insomnia. Despite this, there has been a large effort to identify unique polysomnographic (PSG) patterns in chronic insomnia. Current consensus of PSG markers of chronic insomnia includes increased sleep onset latency (SOL) and wake time after sleep onset (WASO) and decreased total sleep time (TST) and sleep efficiency (SE). Less is known about how acute total sleep deprivation (TSD) may impact these same sleep markers. We investigated PSG patterns at baseline and after TSD in individuals with chronic sleep-onset insomnia as compared to healthy sleeper controls. 11 individuals with chronic insomnia and 11 healthy controls (ages 22-40y, 14 females) completed a 5-day laboratory study with an adaptation night, baseline night, assignment to 38h TSD (n=6 insomnia, n=5 control) or equivalent non-TSD control (n=5 insomnia, n=6 control), and recovery night. Sleep periods were 10h (22:00-08:00) and measured with PSG. Data were analyzed using mixed effects ANOVA with fixed effects of condition (TSD or control) and night (baseline and recovery) and their interaction, with a random effect over subjects on the intercept. There was a significant interaction of condition by night with increases in TST (F3,18=5.61, P< 0.01), SE (F3,18=5.53, P< 0.01), and total N3 (F3,18=12.76, P< 0.001), and a decrease in SOL (F3,18=3.91, P=0.026), N1 (F3,18=3.82, P=0.028), and WASO (F3,18=4.32, P=0.018) in both the insomnia and healthy controls following TSD. No significant difference between conditions were observed in any measured sleep metric at baseline (all P>0.13). After TSD, both groups demonstrated the expected increased SE and total N3 alongside decreases in WASO and SOL and were not significantly different from each other. These findings suggest that conventional sleep staging lacks the sensitivity to identify distinct features of insomnia, both with and without TSD. This underscores the need for alternative methods of measurement such as spectral analysis, the results of which are forthcoming. Additionally, these findings highlight the importance of incorporating subjective sleep assessments in studies of insomnia as objective metrics may not always align with subjective reports of poor sleep. ONR grant N00014-13-C-0063

Sustained attention is important for optimal neurobehavioral performance, but many biological and environmental factors (e.g., circadian rhythm, distraction, etc.) may cause sustained attention deficits. It has been suggested that mastication (chewing) may ameliorate such deficits. We used a randomized, within-subjects, cross-over design to investigate the effect of mastication (gum chewing) on levels of sustained attention. To initially provide data that was ecologically valid for the average person, participants were not sleep deprived or otherwise challenged. Fifty-eight healthy adults (aged 18-45 years; 38 females) completed a 5 h in-laboratory daytime study during which time they completed two, 40 min test bouts. Participants completed the Psychomotor Vigilance Test (PVT), the Sustained Attention to Response Task (SART), the Karolinska Sleepiness Scale (KSS) and the Positive and Negative Affect Schedule (PANAS). During one of the two test bouts, participants were instructed to chew a piece of gum at a steady, comfortable rate. The statistical analyses were conducted blind. The primary outcome variable used for analyses was PVT lapses using the transformation square root of lapses plus square root of lapses plus 1 in addition to PVT mean reciprocal response time. Secondary outcome variables were PVT time-on-task slope and SART error score. Using rested participants and moderately fatiguing tasks, we were unable to detect any significant improvement in PVT or SART performance, or in KSS or PANAS ratings. A follow-up study under conditions of sleep deprivation and/or with longer task duration may provide further insight into the countermeasure potential of mastication.

Abstract Introduction Methadone treatment for opioid use disorder (OUD) reduces drug cravings and promotes abstinence, but sleep disturbances can negatively impact OUD recovery outcomes. Irregular sleep architecture and respiratory disturbance in outpatients administered methadone as medication for OUD (MOUD) are common, but inter-individual differences may be substantial. We investigated the magnitude and stability of inter-individual differences in sleep architecture and respiratory measures in outpatients in their first 90 days of MOUD treatment. Methods N=6 adults (42.5±10.4y, four females) enrolled in local MOUD programs participated in this research study. The daily methadone dose varied between participants (75.8±23.2mg, range 50-120mg). All participants reported past use of multiple substances; four reported current mental health disorders; and four reported chronic pain. As part of the study, participants underwent two consecutive nights of cardiorespiratory PSG, with lights off at 21:49 on average (range 21:17-22:16) on both nights. Sleep stages and apnea-hypopnea index (AHI) were analyzed for systematic inter-individual differences using variance components analysis. The stability of inter-individual differences was quantified with the intraclass correlation coefficient (ICC, ranging from 0 to 1 for negligible to complete stability). Results On night one, mean±SD was 441.9±21.1min for total sleep time (TST), 9.0±5.2min for sleep latency (SL), 35.3±22.4min for wakefulness after sleep onset (WASO), 28.8±26.6min for N1, 220.8±58.6min for N2, 114.0±77.2min for N3, and 78.3±65.6min for REM. AHI scores were 16.4±9.1, indicating moderate sleep apnea on average. Compared to similarly aged healthy adults, N1 was low, and TST, WASO, N2, REM, and AHI were high. Stability was significant and substantial for N1, N2, REM, and AHI (ICC≥0.68, F≥5.19, P< 0.05). Conclusion Despite small sample size, we found substantial, stable inter-individual differences in N1, N2, REM, and AHI, whereas TST, SL, WASO, and N3 did not reach statistical significance. Varying methadone doses, prior substance use, physical and mental health, chronic pain, and (epi)genetic predisposition may partially account for these inter-individual differences. Our PSG data were collected in-laboratory, not naturalistically; results should be interpreted accordingly. Given the substantial, stable inter-individual differences in sleep architecture, a personalized approach to sleep management may be crucial to help improve OUD recovery outcomes. Support (if any) WSU ORAP and HSCL

OBJECTIVE:Individuals with type 1 diabetes (T1D) have increased risk for cognitive dysfunction and high rates of sleep disturbance. Despite associations between glycemia and cognitive performance using cross-sectional and experimental methods few studies have evaluated this relationship in a naturalistic setting, or the impact of nocturnal versus daytime hypoglycemia. Ecological Momentary Assessment (EMA) may provide insight into the dynamic associations between cognition, affective, and physiological states. The current study couples EMA data with continuous glucose monitoring (CGM) to examine the within-person impact of nocturnal glycemia on next day cognitive performance in adults with T1D. Due to high rates of sleep disturbance and emotional distress in people with T1D, the potential impacts of sleep characteristics and negative affect were also evaluated. METHODS:This pilot study utilized EMA in 18 adults with T1D to examine the impact of glycemic excursions, measured using CGM, on cognitive performance, measured via mobile cognitive assessment using the TestMyBrain platform. Multilevel modeling was used to test the within-person effects of nocturnal hypoglycemia and hyperglycemia on next day cognition. RESULTS:Results indicated that increases in nocturnal hypoglycemia were associated with slower next day processing speed. This association was not significantly attenuated by negative affect, sleepiness, or sleep quality. CONCLUSIONS:These results, while preliminary due to small sample size, showcase the power of intensive longitudinal designs using ambulatory cognitive assessment to uncover novel determinants of cognitive fluctuation in real world settings, an approach that may be utilized in other populations. Findings suggest reducing nocturnal hypoglycemia may improve cognition in adults with T1D.

Abstract Introduction Alcohol and opioids disturb sleep, but the impact of substance use on recovery sleep after sleep deprivation is largely unknown. Here we piloted a study protocol to assess the combined impact of total sleep deprivation (TSD) and substance administration on recovery sleep. Methods N=6 healthy normal sleepers (ages 28.2±5.6y; 4 males) completed two 24h laboratory study sessions. During each session, which began at 15:00, participants were kept awake for 15h until 06:00 the next day. They were then given an 8h recovery sleep opportunity (06:00–14:00) and went home. During the second study session, participants were randomly assigned to receive alcohol (n=3; peak BAC of 0.043±0.008% at 00:55, decayed to zero by 06:00) or an opioid (n=3; 10mg oxycodone administered at 22:30). Recovery sleep was recorded polysomnographically and scored using AASM criteria; analyses focused on total sleep time (TST), sleep efficiency, sleep latency, sleep stages N1–N3 and REM, and latency to each of the sleep stages. Results Session 1 (pre-substance) TST was considerably shorter in the opioid group (282±28min) than the alcohol group (403±24min). Therefore, we analyzed sleep variables for the two groups separately, using mixed-effects ANOVA with a fixed effect for study session (TSD vs. TSD+drug) and a random effect over subjects on the intercept. For the opioid group, latency to N3 sleep was significantly longer by 10±2min after opioid administration compared to TSD alone (F=33.8, p=0.028). There were no significant effects of TSD+alcohol compared to TSD alone. Conclusion Opioid administration during TSD delayed N3 onset, but no other effects of opioid or alcohol administration were seen during recovery sleep. Our sample was small, but the within-subjects study design provided considerable statistical power – substance effects on neurobehavioral performance during TSD (reported elsewhere) were readily detectable. However, BAC at the onset of recovery sleep had decayed to zero, and elevated homeostatic sleep pressure from TSD may have negated any residual substance effects other than the opioid-induced delay in N3. Investigating the impact of substances on recovery sleep after TSD, as well as post-recovery neurobehavioral functioning, is important for safety and health in today’s sleep-deprived society. Support (if any) National Safety Council

Current evidence and professional guidance recommend sleeping between 7 and 9 h in a 24-h period for optimal health. The present study examines the association between sleep duration and mortality and assesses whether this association varies by racial/ethnic identity for a large and diverse sample of United States adults. We use data on 274,836 adults, aged 25 and older, from the 2004-2014 waves of the National Health Interview Survey (NHIS) linked to prospective mortality through 2015 (23,382 deaths). Cox proportional hazards models were used in multi-variable regressions to estimate hazard ratios for mortality by sleep duration and racial/ethnic identity, controlling for sociodemographic, socioeconomic, and psychological distress variables. We find elevated risks of mortality from any cause for adults who sleep less than 5 h or more than 9 h in a 24-h period after all adjustments. Further, we find evidence that these elevated risks for mortality are more pronounced for some racial/ethnic groups and less pronounced for others. Improved understanding of differences in sleep duration and sleep health can facilitate more effective and culturally-tailored interventions around sleep health, improving overall well-being and enhancing longevity.

Abstract Introduction The hypothalamic-pituitary-adrenal (HPA) and sympathetic-adrenal-medullary (SAM) axes activate in response to stressors. Real-world exposure to stressors often co-occurs with total sleep deprivation (TSD). The SAM response to a single stressor appears unchanged by TSD, but salivary alpha-amylase (sAA) has shown that TSD blunts the SAM response to repeated stressor exposure. Using salivary cortisol concentration (SCC), we investigated the HPA response to repeated stressors under well-rested and TSD conditions. Methods N=10 healthy adults (ages 28.3±5.78; 5f) completed a 4-day/3-night in-laboratory study with 38h TSD preceded and followed by 10h sleep opportunities. On day 2 (well-rested) and day 3 (TSD), participants completed two stressor sessions in a high-fidelity shooting simulator, separated by 30min. Acting as police officers, civilian participants verbally interacted with emergency response scenarios and decided whether to use (simulated) deadly force. Seven saliva samples were collected each day: pre-stressor, and 0min, 15min, and 30min after each session. Samples were assayed for SCC and normalized against each day’s pre-stressor sample. Results Mixed-effects ANOVA showed a significant effect of sample time (F[5,99]=19.85, p< 0.001), with SCC peaking 15min after the first stressor session and steadily declining thereafter. The SCC peak was significantly blunted during TSD compared to well-rested (t[99]=2.84, p=0.006). Correlations with previously reported, simultaneously assessed sAA concentrations, which peaked right after the first stressor session, were not significant (p>0.2). Additionally, whereas sAA showed a second peak after the second stressor session when participants were well-rested, no second peak was found in SCC. Conclusion The SCC response after one stressor session with simulated emergency response scenarios was blunted during TSD, unlike the sAA response. However, while sAA peaked twice in response to repeated stressor exposure when participants were well-rested (though not during TSD), SCC continued to decline after the first stressor exposure, potentially indicating a HPA refractory mechanism or habituation. Also, over participants there was no significant relationship between the magnitude of the stressor response between SCC and sAA. Taken together, our results suggest fundamentally distinct SAM and HPA axis responsivity to repeated acute stressors, with differential impact of TSD. Support (if any) ONR N00014-13-1-0302, PRMRP W81XWH-20-1-0442.

Abstract Introduction Ample research supports the need for adequate sleep to develop normal physical and mental health throughout childhood. While objective sleep monitoring is ideal, it is not always practical in real-world settings. Therefore, subjective reports provide a more realistic alternative. The validated Children’s Sleep Habits Questionnaire (CSHQ) requires parents/caregivers to retrospectively report sleep quality in school-aged children. In a nation-wide study, we compared CSHQ ratings to determine any relationship with sleep quality and demographic variables. Methods CSHQ and demographic questionnaires were completed for n=152 children (ages 3-12y, 63% Male, 37% female), 45% African American and 55% Caucasian. Means of CSHQ scores and demographic variables were compared using independent t-tests and chi-square test of independence. Results Overall, African American children had lower reported sleep quality compared to Caucasian children (t150= -3.77, p< 0.001). Additionally, reported sleep quality was lower in households earning less than 100K compared to those earning more than 100K (t146= 2.69, p< 0.01), with more African American families earning less than 100K compared to Caucasian families (Χ2(1) = 11.99, p < 0.001). Conclusion Our results are consistent with previous research indicating that African American children have lower reported sleep quality than their Caucasian counterparts. While families with lower income report higher rates of sleep disturbance, this appears to be more prevalent in African American households. Although significantly different, both groups reported high levels of sleep disturbance, highlighting the need for general sleep education as well as the development of interventions tailored to different ethnic groups and demographic levels. Support (if any) Elson S. Floyd College of Medicine Dean’s Excellence Award Scholarship