Conference Agenda

Given that 1 in 5 US high school students report seriously considering suicide in the past year alone, it is essential to identify who is acutely at risk and when youth are most at risk for thinking about suicide. The current study leveraged an intensive, real-time monitoring design for two months to examine sleep (quality, onset latency, nightmares, duration, timing) as modifiable risk factors for suicidal thoughts among adolescents. The current sample of adolescents were recruited nationwide and were unselected for current or past suicide risk. A total of 128 adolescents (Mean Age =15.9 years; 60% female; 45% White) completed a baseline clinical interview and self-reported surveys, along with two months of ecological momentary assessment (EMA) with 3 prompts per day, including morning sleep diaries and evening reports of suicidal ideation (occurrence and intensity). Adolescents wore actiwatches for two months to capture behavioral sleep metrics and installed an application for smartphone sensing to capture phone use and app patterns. Multilevel modeling was used to examine overall associations between sleep and suicidal ideation over the two-month period (between-person), as well as individual fluctuations in sleep and next-day active suicidal thoughts. Findings indicate that youth who reported worse sleep quality (B=-.04, p=.02) and longer sleep onset latency (B=.03, p<.01) were more likely to endorse suicidal thoughts over the study period (between-person). There were only within-person findings for nightmares, such that youth were more likely to endorse suicidal thoughts on days after they had nightmares (B=.57, p=.04). Actigraphic results will also be presented for timing and duration. This study highlights the importance of evaluating modifiable risk factors, such as sleep, within-person and using intensive monitoring designs to inform interventions and prevent suicide.

Objective:

Growing evidence links sleep health impairments, including irregular sleep, with suicide risk across the lifespan, including adolescence. Much of the evidence linking sleep to suicidality during adolescence is based on cross-sectional self-report data, providing little information about the temporal and dynamic processes by which sleep may lead to increasing suicidality.

Methods:

To date, we examined data in 240 youth ages 13–23 (Mean age=16.2, SD=2.0) across two separate cohorts enrolled in an Intensive Outpatient or Acute Hospitalization Program for suicidality. Daily sleep data were collected via actigraphy and morning diary; suicidal ideation, depression severity, and reactivity to positive/negative interpersonal events were collected in evening diary. We examined weekly variability in actigraphy-derived measures of sleep timing and duration (measured with coefficient of variation) and with 24-hour rest activity rhythms (i.e., interdaily stability). We also conducted a small sleep/circadian intervention pilot study in suicidal youth.

Results:

Greater variability in actigraphy-assessed sleep timing and duration was associated with the rates and intensity of concurrent week suicidal ideation, and greater variability in sleep timing predicted rates and intensity of suicidal ideation the following week. In analyses combined across both cohorts, we found that within-person decreases in weekly interdaily stability (a measure of RAR regularity over multiple days) was associated with the occurrence and intensity of acute SI. In our pilot intervention, interdaily stability significantly improved pre- to post-treatment.

Conclusions:

We found prospective associations between sleep regularity and suicidal ideation in ultra-high risk adolescents and young adults. Effective behavioral sleep interventions can improve these targets (e.g., improve erratic and short sleep/wake patterns), and as such, these are modifiable risk factors. By understanding the temporal relationship between sleep and suicidality, new or refined interventions targeting sleep health may improve mental health and lessen suicidality.

Self-reported sleep difficulties are linked with suicidal ideation, suicide attempts and death. However, the underlying mechanism between the neurobiology of sleep and suicide risk remains unclear. Sleep electroencephalogram (EEG) captures electrical activity in the brain during sleep and represents the gold standard for sleep assessment and staging. Therefore, sleep EEG represents a valuable tool to understand the interplay between sleep disruption and suicide risk at the electrophysiological level, particularly by focusing on alpha (relaxed wakefulness) and beta (alert wakefulness) and delta power (slow wave or deep sleep). Building on initial results that linked nocturnal wakefulness and suicidal ideation, we examined patterns of nocturnal alpha and beta activity with next-day suicidal thoughts. Functional analytic results from sleep EEGs from 33 unmedicated participants with treatment-resistant depression demonstrated that oscillations in beta power were linked to next-day suicidal thoughts. We then evaluated the impact of ketamine, a glutamatergic modulator with known impacts on depression and suicidal thoughts. In a sample of 36 unmedicated participants with treatment-resistant depression, ketamine was linked with increased delta power earlier in the night and increased alpha and delta power later in the night, as compared to saline placebo. Differences in delta and alpha power did not mediate ketamine’s antidepressant or suicide ideation effects. Implications for future research on the relationship of sleep EEG, suicide risk and treatment response will be discussed.

Lack of sleep is common among adolescents, with nearly half of Swedish students sleeping less than the recommended 8 hours on school nights. This presentation describes converging evidence linking insufficient sleep to increased suicide risk and explores a promising structural intervention: delayed school start times.

We analyzed cross-sectional data from 8,449 adolescents, finding 46% slept less than 8 hours on school nights—significantly higher among girls (53% vs 38%). Short weekday sleep duration and poor sleep quality remained significant predictors of depression beyond demographic factors and students with suicidal ideation slept 60 minutes less on weekdays compared to peers without ideation. Even after controlling for depression, each 30-minute increase in weekday sleep duration was associated with 10% lower odds of suicidal ideation (OR=0.810, p<.001), suggesting that sleep has a direct effect on ideation.

In a different study we explored implementation feasibility through interviews with school administrators and student focus groups across Stockholm County. While administrators recognized widespread student sleep deprivation, most attributed it to screen time and parenting rather than biological factors. Key barriers included scheduling constraints, anticipated resistance from teachers preferring early hours, and coordination with after-school activities. However, schools with existing "sleep-in mornings" reported positive outcomes.

Preliminary data from an ongoing implementation study examining the effects of one-hour delayed start times on student sleep, wellbeing, and academic performance will be presented.

These findings and their consequence for educational policies, particularly given the alignment between adolescents' biological sleep patterns and mental health needs will be discussed.

Rising rates of suicidal thoughts and behaviors highlight the need for empirically supported interventions. Sleep and circadian rhythm disturbances are associated with suicidal ideation (SI) and may serve as state-sensitive targets for clinical intervention. Studies suggest that interventions targeting sleep and circadian rhythms, including cognitive behavioral therapy for insomnia (CBT-I), pharmacotherapy, and chronotherapy, can improve mental health broadly, but their potential to reduce SI requires further investigation. This study provided a quantitative synthesis of the existing randomized controlled trials (RCTs) assessing the impact of sleep- and circadian-focused interventions on suicidal ideation (SI). A total of 25 unique effect sizes from RCTs evaluating sleep- and circadian-focused interventions on SI were identified. The pooled effect of sleep- and circadian-focused interventions on SI immediately post-intervention was small but significant (g=–0.16, 95% CI =–0.25 to –0.08). Four studies suggested that treatment effects were maintained at a one-year follow-up (g=–0.15, 95% CI=–0.30 to –0.01). Sub-analyses evaluating different types of interventions revealed that behavioral sleep medicine (BSM) interventions (k=15; 60%) produced a small but significant reduction in SI (g=–0.17, 95% CI = –0.26 to –0.08), whereas the effects of chronotherapy (k=5) and pharmacotherapy (k=5) interventions were non-significant. Moderator analyses indicated that participant characteristics, treatment length, and delivery modality did not impact the pooled effect size. However, BSM studies with baseline suicidality as an inclusion criteria (k=5) had larger effects (g = –0.45, 95% CI = –0.70 to –0.21) than those without this criterion (g = –0.13, 95% CI = –0.23 to –0.03). These findings suggest that sleep- and circadian-focused interventions may reduce SI, particularly BSM interventions for individuals with existing suicidality, but further study specifically with suicidal patients is required. Targeting sleep and circadian functioning may represent a promising, modifiable target for addressing suicidality among individuals with elevated SI.