Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Aug 25.
Published in final edited form as: J Psychiatr Res. 2024 Aug 27;178:388–396. doi: 10.1016/j.jpsychires.2024.08.040

Sleep disorder symptoms and suicidal urges among US Marines seeking suicide treatment: Findings from an intensive daily assessment study

Lily A Brown 1, Yiqin Zhu 1,2, Bridget Feler 1, Marin Kautz 1, Daniel J Taylor 3, Kristi E Pruiksma 4, Justin Baker 5, Johnnie Young 5, Lauren Khazem 5, Craig J Bryan 5, Joshua Wiley 6
PMCID: PMC12371704  NIHMSID: NIHMS2103795  PMID: 39214033

Abstract

Background:

Rates of suicide in United States Marines are among the highest in the military, and sleep disorder symptoms are a known risk factor for suicide in the military. Intensive ecological momentary assessments (EMA) might improve the ability to detect periods that are characterized by increased suicidal ideation. Marines who were at high risk for suicide were intensively assessed for one month on sleep, suicidal urges, posttraumatic stress disorder (PTSD) and depression symptoms.

Methods:

U.S. Marines (N = 40) who had a past month suicide attempt or suicidal urges with intent were sent EMA for 28 days. Mixed effects models explored associations among daily sleep, suicidal urges, PTSD, and depression symptoms.

Results:

Worsened sleep indicators on a given night significantly predicted higher maximum values of suicide urges the following day. Worse sleep quality the prior night was moderately associated with more severe PTSD symptoms and depression symptoms. Greater severity of PTSD symptoms and depression symptoms were strongly associated with both the maximum value and the range of suicide urges. PTSD and depression symptoms mediated the relationship between sleep quality and suicidal urges. Participants reported that 0000 – 0300 had the greatest elevation in endorsement of highest suicide urges.

Limitations:

This study had a small sample size may not generalize beyond active duty Marines.

Conclusions:

Poor sleep quality and other sleep markers were an important risk factor for suicidal urges among U.S. Marines. This relationship was mediated by exacerbations in PTSD and depression symptoms. Interventions are needed to interrupt suicide risk during and following nights with poor sleep.

Keywords: Sleep, suicidal ideation, military, posttraumatic stress disorder

Introduction

Active duty service members are at particularly high risk for suicide despite decades of work to identify risk factors, protective factors, and prevention strategies for suicide (Franklin et al., 2017; Pruitt et al., 2019). From 2019 to 2021, the average rate of suicide for US active duty service members was 26.4 per 100,000, with the highest rates among the Army and the Marine Corps (34.3 and 27.9 per 100,000, respectively; Department of Defense, 2022). From 2011 to 2022, suicide rates rose for active duty service members with Army and Marine Corps service members reporting the most prevalent rates (Department of Defense, 2023). Active duty service members present with unique distal risk factors for suicide, including high levels of exposure to unexpected or morally challenging events, which may have a lasting impact on their lives and further increase their risk for suicide across time (LeardMann et al., 2021; Schmied et al., 2023). Despite identifying unique demographic and circumstantial risk factors, the field continues to lack an understanding of the proximal risk factors that account for fluctuations in suicide risk for this population. Using ongoing, naturalistic data collection of active duty service members’ lived experiences, such as ecological momentary assessment (EMA), can help researchers and clinicians capture proximal risk factors for suicidal ideation so that timely and individualized prevention efforts can be delivered almost in real-time.

In the general population and among active duty service members, sleep disorder symptoms are an important risk factor for suicidal ideation and attempts (Bernert et al., 2015; Fawcett et al., 1990; Geoffroy et al., 2022; McCall et al., 2010; Mysliwiec et al., 2013; Wang et al., 2019). In cross-sectional and longitudinal studies, insomnia is an independent risk factor for suicide ideation and attempts, above and beyond the influence of depression symptoms (Pigeon et al., 2016). For many patients with insomnia, their objective sleep may be statistically equivalent to patients without insomnia, but their subjective distress about sleep is high (Orff et al., 2007). Nightmares have also emerged as a critical risk factor for suicidal ideation and attempts among civilians (Bernert et al., 2015; Nadorff, Fiske, et al., 2013; Nadorff, Nazem, et al., 2013; Tanskanen et al., 2001) and active duty military personnel (Paxton Willing et al., 2021). Lastly, nocturnal wakefulness is an understudied topic area. Still, suicide mortality studies demonstrated that the middle of the night is a very high-risk time for suicide death in the general population (Perlis et al., 2016) and among Veterans (McCarthy et al., 2019). In qualitative research, individuals with a history of suicidal thoughts and behaviors report that the middle of the night is an ideal time for a suicide attempt because it is less likely to be interrupted by others and because social support is reduced during this window, plus problems with sleep were perceived to make the next day more challenging because of increasing depression and negative affect (Littlewood et al., 2016). Empirical research is needed to verify this important qualitative feedback. Prior research explored this in civilians over one week (Brüdern et al., 2022; Littlewood et al., 2019) or two weeks (Rogers & Bozzay, 2024), but research over a longer period of time may reveal different associations. These associations may also differ among active duty military personnel who have career duties which may interfere with sleep.

Sleep disorders might increase the risk for suicidal ideation because of their impact on psychiatric symptoms characterized by negative affect which is an independent risk factor for suicidal ideation (Blais & Geiser, 2019; Rubio et al., 2020) and is elevated in psychiatric syndromes such as posttraumatic stress disorder (PTSD) and depression. A robust literature base highlights how sleep disorder symptoms often precede depression and PTSD symptoms (Gehrman et al., 2013; Mellman et al., 2007; Riemann & Voderholzer, 2003; Taylor et al., 2003). In turn, PTSD symptoms (Brown et al., 2016, 2018, 2020) and depression (Rugo et al., 2020) are each independently associated with suicidal ideation and attempts. PTSD and depression mediate the relationship between sleep disorder symptoms and suicidal ideation among active duty military personnel (Allan et al., 2017; Morgan et al., 2017). However, the prior research on this topic has primarily been conducted in cross-sectional or longitudinal studies with infrequent assessments and has not utilized prospective measures of sleep efficiency, sleep duration, number of awakenings, length of nocturnal wakefulness, sleep quality, or nightmare frequency. To eventually design just-in-time interventions to interrupt the association between sleep disorder symptoms, psychiatric symptoms (PTSD symptoms and depression symptoms), and suicidal ideation, there is a need to understand the patterns of influence in daily fluctuations in these symptoms.

The goal of this project was to conduct intensive longitudinal assessments of sleep, psychiatric symptoms (depression and PTSD symptoms), and suicidal urges (an indicator of wanting to act on suicidal ideation) among Marines who were at higher risk for suicide (i.e., recently hospitalized for heightened suicide risk, recent suicide attempt, or reporting suicidal ideation with intent or plan). Marines completed 28 days of daily sleep diaries (i.e., sleep efficiency, sleep duration, number of awakenings, length of nocturnal wakefulness, sleep quality, nightmare frequency, PTSD and depression symptoms, and suicidal urges). We hypothesized that sleep disturbances on a given day (t) would predict suicidal urges on the subsequent day (t+1). Based on prior research (Brown, Zhu, et al., 2022; McCarthy et al., 2019; Perlis et al., 2016), we hypothesized that suicidal urges would be most severe between 0200 – 0300. Finally, we hypothesized that PTSD and depression would mediate the associations between sleep disturbances and suicidal urges.

Methods

Participants.

Participants were Marines (N = 40) recruited from Marine Corps Base Camp Lejeune, North Carolina, as part of a parent trial (PI: C. J. Bryan, W81XWH1820022; Khazem et al., 2021). Inclusion criteria were age 18 years or older; reporting current suicide ideation with intent to die and/or a suicide attempt within the past month; ability to understand and speak the English language; and ability to complete the informed consent process. Exclusion criteria were any psychiatric or medical conditions that precluded the ability to provide informed consent or participation in outpatient treatment (e.g., psychosis, mania, acute intoxication), anticipated discharge from the military, deployment, or change of duty station within 90 days. For full details, see our published protocol (Brown, Taylor, et al., 2022). Demographic information is presented in Table 1.

Table 1.

Demographic information of participants (N = 40)

Variable
Age, M (SD), year 22.53 (2.85)
Gender
 Male n (%) 30 (75.0%)
 Female n (%) 10 (25.0%)
Marital Status
Divorced/Separated 3 (7.5%)
Married/Living with Partner 15 (37.5%)
Single/Not living with Partner 20 (50.0%)
Cohabitating 2 (5.0%)
Ethnicity (% Hispanic or Latino) 16 (40.0%)
Race
White 21 (52.5%)
Black/African American 5 (12.5%)
Other/Multiracial 10 (25.0%)
Asian 2 (5.0%)
Native Hawaiian/Pacific Islander 2 (5.0%)
American Indian/Alaskan Native 4 (7.5%)
Education
High school diploma or equivalent 27 (67.5%)
Some college, no degree 11 (27.5%)
Associate’s degree 1 (2.5%)
Bachelor’s degree 1 (2.5%)
Years of active duty, M (SD), year 3.75 (2.41)
Rank
Enlisted-1 1 (2.5%)
Enlisted-2 6 (15.0%)
Enlisted-3 17 (42.5%)
Enlisted-4 9 (22.5%)
Enlisted-5 5 (12.5%)
Enlisted-6 2 (5.0%)
Open in a new tab

Measures.

Daily Sleep Diary.

Participants were instructed to complete the consensus sleep diary daily before going to sleep at night and upon awakening, capturing subjective assessments of the preceding night’s sleep. The metrics gathered included sleep efficiency, sleep duration, number of awakenings, length of nocturnal wakefulness, sleep quality, and nightmare frequency. Length of nocturnal wakefulness was evaluated using the following item: “In total, how long did these awakenings last in minutes?” Sleep duration and efficiency were assessed by calculating nocturnal wakefulness and responding to three further items: 1) “What time did you attempt to fall asleep last night?”, 2) “What time did you arise for the day?”, and 3) “How long did it take you to fall asleep in minutes?”. Sleep duration was computed by deducting sleep latency and nocturnal wakefulness from the total time spent in bed. Sleep efficiency was calculated by dividing sleep duration by the total time in bed. Participants’ perception of sleep quality was measured by one item, “How would you rate your quality of sleep?” on a 5-point Likert scale ranging from 0 (“Very Poor”) to 4 (“Very Good”). Finally, the frequency of nightmares was measured using the item “Last night, how many nightmares did you have that woke you up?”

Suicide-Visual Analogue Scale (S-VAS; Bryan, 2019).

Participants were instructed to reflect on their suicide urges using the S-VAS. Item five of the S-VAS assesses the “urge to kill myself” using a horizontal sliding scale ranging from 0 (none) on the left anchor and 100 (extreme) on the right anchor. Initially, the S-VAS was presented with the slide indicator on the “none” position, and participants were instructed to indicate their response by moving the slide indicator, if relevant. The S-VAS was administered four times daily for 28 days: during the waking day (i.e., 0800 to 2300) real-time momentary suicide urges were randomly assessed twice, as well as once when awakening in the morning sleep diary, and once at night before sleep in the nighttime sleep diary. The morning sleep diary assessed the strongest suicide urges in the prior night (from the point of getting into bed) alongside the timing of the strongest urges. The nighttime sleep diary assessed the strongest suicide urges in the day (from the point of awakening) alongside the timing of the strongest urges. The maximum value and range of the four daily S-VAS surveys were computed for each participant. The S-VAS has strong convergent and divergent validity and is a strong predictor of subsequent suicide attempts (Bryan, 2019).

Posttraumatic Checklist for DSM-5-Reduced (PCL-5-R; Ringer et al., 2018).

The PCL-5-R is a self-report measure derived from the PCL-5 that assesses the severity and presence of PTSD symptoms using a 1 (not at all) to 5 (extremely) point Likert scale (range 5 to 40). The PCL-5 has good internal consistency (α = 0.96) and test-retest reliability (r = 0.84), as well as good convergent and discriminant validity (Bovin et al., 2016). Eight items from the PCL-5 were administered once daily through EMA to assess PTSD symptoms, and participants were asked to reflect on how much they have been experiencing them in the past 24 hours on a scale from not at all to extremely. These items are included in the Military Suicide Research Consortium Common Data Elements and have 1) intrusive memories; 2) nightmares; 3) flashbacks; 4) physical reactions to reminders; 5) avoiding thinking or talking about it; 6) avoidance of activities or situations; 7) hyperarousal; and 8) startle. Measure timing was adapted to the past 24 hours and sent to participants daily. The internal consistency of the eight-item measure was excellent (Cronbach’s alpha = .94).

Patient Health Questionnaire – 2 (PHQ-2; Kroenke et al., 2003).

The PHQ-2 comprises the first two items of the PHQ-9, which was developed to make a criteria-based diagnosis of major depressive disorder. It initiates with a guiding query: “In the past 24 hours, how often have you been bothered by the following problems?” The subsequent two items include 1) “little interest or pleasure in doing things” and 2) “feeling down, depressed, or hopeless.” Participants rated each item on a 4-point scale from 0 (“Not at all”) to 3 (“Nearly all the time”). The PHQ-2 has good sensitivity (.82) and specificity (.87) for depression screening and was assessed once daily via EMA, with anchors changed to reflect the past 24 hours (Levis et al., 2020).

Procedures.

Following their baseline assessment, Marines downloaded a mobile app on their phone and completed daily assessments of suicidal urges using the Suicide-Visual Analogue Scale (S-VAS; Bryan, 2019) and their daily sleep diary. See Brown et al. (2022) for additional details. In addition to the daily assessments of suicidal urges, Marines also completed a daily evaluation of PTSD symptoms (using the 8 MSRC CDE items from the PTSD Checklist over the prior 24 hours; Blanchard et al., 1996; Ringer et al., 2018) and a daily assessment of depression symptoms (using the PHQ-2; Kroenke et al., 2003). During the 28-day assessment window, participants were randomized to one of two treatment conditions: Brief Cognitive Behavioral Therapy (BCBT) or present centered therapy (PCT) with a frequency of once per week sessions and up to twelve sessions (for more details, see Khazem et al., 2021). Thus, the assessments occurred during approximately the first month of treatment of a three month treatment.

Using methods that we previously published (Brown, Zhu, et al., 2022) and used by Perlis et al. (2016), we compared the timing of the highest suicidal urges (using the S-VAS during the morning and nighttime sleep diary) and compared the observed frequency of each time to the expected frequency (based on the American Time Use Survey, ATUS; U.S. Department of Labor Statistics n.d.; Perlis et al., 2016), accounting for the likelihood of being awake based on estimates from the general population, for each hour in a 24-hour window across the study. When a non-zero score was reported on the S-VAS during the morning and nighttime sleep diary, Marines were asked to report the timing of their highest urge.

Data Analysis.

Outliers in the data were Winsorized to three standard deviation values prior to analyses. To test the daily-level associations among five sleep metrics (sleep efficiency, sleep duration, length of nocturnal wakefulness, sleep quality, and nightmare frequency) and suicide urges, we conducted the following linear mixed-effect models using the lmer function of the lme4 package (Bates, 2016) in R (version 4.2.1): 1) sleep metrics in the prior night predicting S-VAS (in terms of its range and maximum value) in the next day; and 2) the reversed direction, with the range and maximum value of S-VAS on a given day predicting sleep metrics on that night. Following these, we conducted two series of sensitivity analyses to address potential bias introduced by missing data: 1) whether missingness of sleep metrics (0 = not missing, 1 = missing) in the prior night predicted S-VAS (in terms of its range and maximum value) on the following day; 2) whether missingness of S-VAS (in terms of its range and maximum value) on a given day predicted sleep conditions on the subsequent night. The lmer function uses a method called “restricted maximum likelihood” (REML) to estimate model parameters. It inherently handles missing data under the assumption that the data are missing at random (MAR), which means that the probability of missingness is related to the observed data but not the unobserved data.

To examine the potential mediating role of PTSD/depression symptoms in the relationship between sleep and suicide, we adhered to the procedures outlined by Tingley et al. (2014), which enabled the estimation of Average Causal Mediation Effects (ACME) using the ‘mediation’ package in R. The ‘mediation’ package supports causal mediation analysis of multilevel data via the lmer and glmer functions in the lme4 package (Bates, 2016). Our first step involved constructing a series of mediator models to explore whether the prior night’s sleep metrics predicted PTSD and depression symptoms on the following day. Subsequently, we evaluated specific outcome models for sleep metrics that showed significance in the mediator models. These models were constructed to investigate whether PTSD/depression symptoms predicted S-VAS in terms of its range and maximum value. Notably, each model included the corresponding significant sleep metric as a covariate. The ACME was estimated using the ‘mediate’ function, which provides an empirical assessment of the extent to which sleep impacts suicide risk through the potential mediation of PTSD/depression symptoms.

Results

Descriptive Statistics

The total possible and valid observation counts, means, medians, standard deviations, and range for all EMA variables were presented in Table 2. Their distributions were presented in Figure 1. The valid observation counts ranged from 369 to 775 for 40 participants over 28 days. Missing proportions were 65.4% for suicide urges in random surveys, 54.7% for morning surveys covering morning S-VAS, 42.41% for all sleep variables, 64.2% for S-VAS in night surveys, 67.1% for depression symptoms, and 66.9% for PTSD symptoms. Given that participants could report up to four S-VAS per day, we derived daily maximums and ranges for S-VAS to align with other once-a-day assessed variables. Consequently, the missing rates for the daily maximum and range of S-VAS stood at 42.3%. Missing data rates were higher than those of prior EMA suicide studies (e.g., 37.25% missingness for a study of suicidal inpatients; Kleiman et al., 2017), which might be due to the sample (active duty military personnel with acute suicide risk).

Table 2.

Descriptive statistics.

Variable N Valid n Mean Median SD Range ICC
S-VAS random survey 2240 775 (34.60%) 16.87 0 25.40 0–100 0.788
S-VAS morning 1120 507 (45.27%) 18.04 0 27.31 0–100 0.771
S-VAS night 1120 401 (35.80%) 18.74 0 27.60 0–100 0.766
S-VAS daily maximum 1120 646 (57.68%) 22.19 5 29.44 0–100 0.762
S-VAS daily range 1120 646 (57.68%) 6.40 0 12.16 0–49.14 0.465
PTSD 1120 371 (33.13%) 18.87 17 9.73 8–32 0.802
Depression 1120 369 (32.95%) 2.59 2 1.97 0–6 0.706
Sleep Quality 1120 479 (42.77%) 1.82 2 0.87 0–4 0.389
Nightmare Times 1120 479 (42.77%) 0.73 0 1.61 0–13.43 0.293
Sleep time (hours) 1120 475 (42.41%) 7.10 7.08 2.33 1–13.38 0.277
Sleep Efficiency 1120 475 (42.41%) 0.87 0.92 0.13 0.44–0.57 0.534
Awakening Length (hours) 1120 479 (42.77%) 0.20 0.08 0.29 0–1.14 0.387
Open in a new tab

Notes. N refers to the number of possible observations. ICC = Intraclass Correlation Coefficient. S-VAS = Suicide-Visual Analogue Scale, PTSD = posttraumatic stress disorder.

Data were Winsorized to three standard deviations.

Figure 1. Violin Plots.

Figure 1.

Open in a new tab

Notes. SU = Suicide Urges, PTSD = posttraumatic stress disorder.

Daily-level associations between depression/PTSD symptoms and sleep

Worse sleep quality on the prior night was associated with both more severe PTSD symptoms (ß = −1.03, p = .009) and more severe depression symptoms (ß = −.30, p = .016; Table 3). The other sleep variables were not associated with PTSD or depression severity. Specifically, nightmares (ßPTSD = 0.28, p = .077; ßDep = 0.05, p = .272), length of nocturnal wakefulness (ßPTSD = 1.73, p = .175; ßDep = −0.45, p = .252), sleep time (ßPTSD = −0.18, p = .260; ßDep = −0.04, p = .508), or sleep efficiency (ßPTSD = −1.80, p = .554; ßDep = −1.97, p = .063) were not associated with PTSD or depression severity.

Table 3.

Prospective associations between Sleep Variables and Next-day PTSD and Depression

Outcomes Predictors β SE Standardized Coefficient 95% CI of (Standardized) t p
PTSD Nocturnal Wakefulness 1.73 1.27 0.05 [−0.02, 0.13] 1.36 .175
Sleep Quality −1.03 0.39 −0.09 [−0.17, −0.02] −2.63 .009
Nightmare frequency 0.28 0.16 0.06 [−0.01, 0.12] 1.78 .077
Total Sleep Time −0.18 0.16 −0.04 [−0.10, 0.03] −1.13 .260
Sleep Efficiency −1.80 3.05 −0.02 [−0.10, 0.06] −0.59 .554
Depression Nocturnal Wakefulness −0.45 0.39 −0.07 [−0.19, 0.05] −1.15 .252
Sleep Quality −0.30 0.12 −0.14 [−0.24, −0.03] −2.43 .016
Nightmare frequency 0.05 0.05 0.06 [−0.05, 0.16] 1.10 .272
Total Sleep Time −0.04 0.06 −0.04 [−0.14, 0.07] −0.66 .508
Sleep Efficiency −1.97 1.05 −0.11 [−0.22, 0.01] −1.88 .063
Open in a new tab

Notes. SE = Standard Error, CI = Confidence Interval, PTSD = posttraumatic stress disorder.

Daily-level associations between sleep and suicidal urges.

Better sleep quality on the prior night significantly predicted lower maximum values of suicide urges (ß = −2.18, p = .017) but not their range (ß = −.19, p = .752) on the following day. Nightmare frequency, length of nocturnal wakefulness, total sleep time, and sleep efficacy in the prior night did not significantly predict the maximum value or range of next-day suicide urges. However, when controlling for prior-day suicidal urge (Table 4), less sleep time on the prior night significantly predicted greater maximum values of suicide urges (ß = −.84, p = .007) but not their range (ß = −.33, p = .147) on the following day. Similarly, lower sleep efficiency on the prior night significantly predicted greater maximum values of suicide urges (ß = −14.23, p = .031) but not their range (ß = −7.03, p = .148) on the following day. Nightmare frequency, length of nocturnal wakefulness, and sleep quality in the prior night did not significantly predict the maximum value or range of next-day suicide urges. Reversed analyses for the significant relationships indicated that neither maximum value (ßtime = 4.68E−03, ptime = .43; ßefficiency = 1.14E−05, pefficiency = .97) nor range of suicide urges (ßtime = 1.36E−03, ptime = .90; ßefficiency =−1.50E−04, pefficiency = .77) predicted sleep time or sleep efficiency the next day.

Table 4.

Prospective associations between Sleep Variables and the Maximum and Range of Next-day S-VAS after controlling for auto-correlation of suicidal urge the previous day

Outcome Predictors β SE Standardized Estimate 95% CI (Standardized) t p
Max of next-day S-VAS Nocturnal Wakefulness −1.94 2.61 −0.02 [−0.07, 0.03] −0.75 .457
Sleep Quality −0.68 0.87 −0.02 [−0.07, 0.03] −0.78 .436
Nightmare frequency 0.46 0.41 0.03 [−0.02, 0.08] 1.11 .267
Total Sleep Time −0.84 0.31 −0.07 [−0.11, −0.02] −2.71 .007
Sleep Efficiency −14.23 6.54 −0.05 [−0.11, 0.01] −2.18 .031
PTSD 0.44 0.13 0.15 [0.06, 0.24] 3.29 .002
Depression 3.33 0.54 0.24 [0.16, 0.31] 6.18 <.001
Range of next-day S-VAS Nocturnal Wakefulness −2.16 1.65 −0.05 [−0.14, 0.03] −1.11 .267
Sleep Quality −0.28 0.65 −0.02 [−0.11, 0.07] −0.44 .662
Nightmare frequency −0.16 0.31 −0.02 [−0.11, 0.06] −0.53 .595
Total Sleep Time −0.33 0.23 −0.06 [−0.14, 0.02] −1.45 .147
Sleep Efficiency −7.03 4.85 −0.07 [−0.17, 0.02] −1.45 .148
PTSD 0.24 0.10 0.19 [0.04, 0.33] 2.49 .014
Depression 1.49 0.37 0.24 [0.12, 0.35] 3.98 <.001
Open in a new tab

Notes. SE = Standard Error, CI = Confidence Interval, S-VAS = Suicide-Visual Analogue Scale, PTSD = posttraumatic stress disorder.

Timing of suicidal urges.

During the nighttime sleep diary, where participants were asked to report the timing of their highest suicidal urges over the prior night (for any non-zero suicidal urges), participants reported that 0200 (2 AM, Standardized Incidence Ratio (SIR) = 14.646), 0000 (12 AM, SIR = 10.547), 0300 (3 AM, 8.698), and 0100 (1 AM, SIR = 7.937) were the time periods with the greatest elevation in endorsement of highest suicide urges, after accounting for the likelihood that participants would be awake at each hour (see Table 5). This is consistent with the hypothesis described in our protocol, wherein we proposed that 0200 would have the highest likelihood of endorsement.

Table 5.

Timing of suicidal ideation during the morning sleep diary (assessed for the prior night)

Time ATUS % Awake Scaled % Awake Expected awake Scaled Expected Awake Observed Observed % Adjusted % Scaled % Standardized Incidence Ratio (SIR) Lower 95% CI for SIR Upper 95% CI for SIR
12:00 AM 14.47% 0.92% 1.447 2.181 23 21.905 15.895 16.521 10.547* 0.810 1.167
1:00 AM 8.36% 0.53% 0.836 1.260 10 9.524 11.962 12.433 7.937* 0.782 1.187
2:00 AM 5.89% 0.37% 0.589 0.888 13 12.381 22.071 22.941 14.646* 0.810 1.167
3:00 AM 5.34% 0.34% 0.534 0.805 7 6.667 13.109 13.625 8.698* 0.737 1.217
4:00 AM 10.19% 0.65% 1.019 1.536 3 2.857 2.944 3.060 1.954 0.586 1.299
5:00 AM 21.89% 1.39% 2.189 3.299 14 10.526 6.396 6.647 4.244* 0.817 1.161
6:00 AM 47.40% 3.01% 4.74 7.143 16 12.030 3.376 3.508 2.240* 0.830 1.152
7:00 AM 69.92% 4.45% 6.992 10.537 12 11.429 1.716 1.784 1.139 0.802 1.173
8:00 AM 82.66% 5.26% 8.266 12.457 14 13.333 1.694 1.760 1.124 0.817 1.161
9:00 AM 89.80% 5.71% 8.98 13.533 18 17.143 2.004 2.083 1.330 0.840 1.144
10:00 AM 93.82% 5.97% 9.382 14.139 18 17.143 1.919 1.994 1.273 0.840 1.144
11:00 AM 95.63% 6.08% 9.563 14.412 12 11.429 1.255 1.304 0.833 0.802 1.173
12:00 PM 96.34% 6.13% 9.634 14.519 9 8.571 0.934 0.971 0.620 0.770 1.195
1:00 PM 95.92% 6.10% 9.592 14.455 2 1.905 0.209 0.217 0.138 0.482 1.340
2:00 PM 96.01% 6.11% 9.601 14.469 2 1.905 0.208 0.217 0.138 0.482 1.340
3:00 PM 96.46% 6.13% 9.646 14.537 3 2.857 0.311 0.323 0.206 0.586 1.299
4:00 PM 96.81% 6.16% 9.681 14.589 1 0.952 0.103 0.107 0.069 0.236 1.406
5:00 PM 97.26% 6.18% 9.726 14.657 5 4.762 0.514 0.534 0.341 0.686 1.248
6:00 PM 97.43% 6.20% 9.743 14.683 7 6.667 0.718 0.747 0.477 0.737 1.217
7:00 PM 96.80% 6.16% 9.68 14.588 8 7.619 0.826 0.859 0.548 0.755 1.205
8:00 PM 93.09% 5.92% 9.309 14.029 9 8.571 0.967 1.005 0.642 0.770 1.195
9:00 PM 79.91% 5.08% 7.991 12.043 7 6.667 0.876 0.910 0.581 0.737 1.217
10:00 PM 53.28% 3.39% 5.328 8.029 14 13.333 2.628 2.731 1.744* 0.817 1.161
11:00 PM 27.96% 1.78% 2.796 4.214 10 9.524 3.577 3.717 2.373* 0.782 1.187
SUM 1572.64 100.00 157.264 237 237 219.700 96.211 100
Open in a new tab

Daily-level associations among PTSD, depression symptoms, and suicidal urges

Greater severity of PTSD symptoms and depression were strongly associated with both the maximum value (ßPTSD = 0.44, p = .002; ßDep = 3.33, p < .001) and the range of suicide urges (ßPTSD = 0.25, p = .014; ßDep = 1.49, p < .001).

Mediation analyses: Mediation effect of PTSD and Depression

According to the causal mediation analyses, both PTSD and depressive symptoms played significant mediating roles in the relationships between sleep quality and subsequent suicide urges. PTSD symptoms significantly mediated the relationship between prior day sleep quality and max suicidal urges on the following day (ACME = −0.29, p = 0.02) but not the range of suicidal urges on the following day (ACME = −0.11, p = 0.32). Depressive symptoms emerged as significant mediators between sleep indicators and suicide urges. Specifically, depressive symptoms significantly mediated the relationship between prior day sleep quality and suicide urges, both in terms of the maximum of S-VAS on the following day (ACME = −0.74, p = 0.04) and the range of SVAS on the following day (ACME = −0.28, p = 0.04).

Sensitivity to missingness.

Missingness of sleep efficiency (ß = −4.46, p < .001), sleep quality (ß = −5.03, p < .001), and awakening length (ß = −5.03, p < .001) were significantly associated with a lower range of suicide urges in the next day. That is, participants who had more missing data on sleep efficiency, quality, and awakening length experienced more stable levels of suicide urges on the subsequent day. No other significant relationships were observed.

Discussion

In this sample of Marines who were at high risk for suicide, poor sleep quality predicted worsened next day PTSD and depression symptoms which, in turn, predicted worsened suicidal urges. The relationship between self-reported sleep quality on a given night and next-day suicidal urges was mediated by PTSD and depression severity. PTSD and depression severity were not associated with other sleep parameters (i.e., sleep efficiency, sleep duration, number of awakenings, length of nocturnal wakefulness, and nightmare frequency). An analysis of a reversed path, wherein suicide urges was explored as a predictor of sleep quality, was not significant. Finally, consistent with hypotheses, the middle of the night (between 0000 and 0300) was associated with a greater than chance likelihood of Marines endorsing elevations in suicidal urges.

Consistent with hypotheses and qualitative feedback from individuals with lived experience (Littlewood et al., 2016), Marines who self-reported worsened sleep quality on a given night reported elevations in suicidal urges on the subsequent day, though this relationship did not hold after accounting for prior-day suicide urges. Total sleep time and sleep efficiency both predicted next-day suicidal urges when controlling for prior-day suicidal urges. Sleep quality ratings reflect a subjective global appraisal of sleep, which may or may not be impacted by other sleep parameters (i.e., sleep efficiency, sleep duration, number of awakenings, length of nocturnal wakefulness, and nightmare frequency). For example, someone may fall asleep quickly, stay asleep, and wake at a desired time but still not feel rested or that they obtain good sleep quality. Or they may not experience a sense of having had “deep” sleep. It may be that the sleep quality rating is a general measure of distress. These findings for the effect of three different sleep indicators predicting next-day suicidal urges is a compelling finding given the high prevalence of sleep disorder symptoms among active duty military personnel, with some rates as high as 27.3% for sleep apnea and 24.7% for insomnia (Mysliwiec et al., 2013). Numerous prior cross-sectional (Soberay et al., 2019; Tucker et al., 2021) and longitudinal (Bryan et al., 2015) studies have demonstrated significant associations between insomnia severity and suicidal ideation among active duty military personnel. In addition, interventions that improve insomnia are associated with reductions in suicidal ideation severity (Kalmbach et al., 2022; Trockel et al., 2015). When studies dissected which features of insomnia are most strongly associated with negative affect in general, subjective perception of poor sleep quality tends to have the strongest predictive power (Orff et al., 2007). However, one cross-sectional study found that problems with sleep onset had the strongest association with suicidal ideation or suicide attempts, whereas sleep quality was not predictive of either (Batterham et al., 2021). Two studies of civilian samples found that poor sleep quality and short sleep duration both predicted higher next-day suicidal ideation. Still, these studies only evaluated participants for one week (Brüdern et al., 2022; Littlewood et al., 2019). More intensive research is needed on this topic to resolve these discrepant findings. This is one of the first studies of its kind to demonstrate daily-level effects over a longer period (one month) of subjective sleep quality, total sleep time, and sleep efficiency on suicidal urges. This is also the first study in an active-duty military personnel sample to demonstrate the immediate effects of poor sleep quality, total sleep time, and sleep efficiency on a given night and elevated suicidal urges the next day.

Naturalistic observation and epidemiological studies demonstrated that sleep deficiencies tend to precede the worsening of PTSD symptoms and depression (Gehrman et al., 2013; Mellman et al., 2007; Riemann & Voderholzer, 2003; Taylor et al., 2003). Given that PTSD and depression diagnoses both include sleep deficiencies as a part of the symptom profile, clinicians often assume that sleep deficiencies are a consequence, rather than a cause, of PTSD or depression exacerbation. In this study, we observed that greater nightmare frequency and poorer perception of sleep quality on a given night predicted worsened PTSD and depression symptoms the next day.

PTSD and depression are well-established predictors of suicidal ideation and suicide attempts (Brown et al., 2016, 2018; Sher et al., 2012). In a large study that evaluated numerous psychiatric diagnoses, PTSD was one of the few diagnoses that predicted the transition from suicidal ideation to attempts (May & Klonsky, 2016). However, this is the first study to our knowledge that demonstrated daily-level associations between suicidal urges and both PTSD and depression. Specifically, we demonstrated that poor sleep quality on a given night predicted PTSD, which then predicted increased suicidal urges. To our knowledge, it is also the only study to demonstrate these effects among active-duty military personnel.

Our findings on the timing of increased suicidal urges are consistent with a report among civilians (Brown, Zhu, et al., 2022) and with suicide mortality studies in the general population (Perlis et al., 2016) and among Veterans (McCarthy et al., 2019), all of which demonstrate that the middle of the night is characterized by higher risk for suicide. Our study adds to this literature by demonstrating that Marines reported their highest suicidal urges occurred between midnight and 0300. These findings suggest that future research should explore the utility of interventions that are made available to service members in the middle of the night to reduce suicide risk.

Several important limitations of this study require consideration. First, this study included a relatively small sample (N = 40) comprising mostly White male Marines. Therefore, these findings may not generalize outside of this sample. Second, as is typical for extended (month-long) EMA assessment studies, missingness was substantial for some assessments. Missingness of sleep assessments predicted more stability in suicidal urges (lower range) but no other effects in our sample. This should be investigated in future research. Third, Marines were enrolled in a clinical trial to intervene to reduce suicide risk as a parent study, which might have altered naturalistic associations among sleep, suicidal urges, PTSD, and depression. Fourth, when participants were completing the suicide urge assessment retrospectively in the morning and nighttime sleep diary and when they were completing past 24-hour ratings of PTSD symptoms and depression, those assessments could have been subjected to recall bias. Fifth, in our EMA assessments we assessed suicidal urges, which are urges to act on suicidal ideation, whereas most prior research assessed suicidal ideation alone. This is a strength of our research in that suicidal urges may reflect a higher acuity state compared to suicidal ideation alone but may not translate to individuals with lower intensity suicidal ideation. Finally, participants in this study were randomized to one of two types of psychotherapy and it is possible that their therapy receipt, which only in the first month of treatment, might have altered the pattern of results.

This study has important clinical implications for understanding the timing of suicide risk among Marines. First, this study revealed the importance of regularly assessing sleep quality, sleep efficiency, and total sleep time among active-duty military personnel. By including frequent assessments of sleep indicators, units may improve their detection of service members transitioning from lower to high-risk states. Second, this study revealed the close associations among sleep nightmares, PTSD, depression, and, ultimately, suicide risk among Marines. Third, this study revealed that the middle of the night is a high-risk time for Marines. Our findings demonstrate the utility of brief assessments in catching Marines transitioning from lower-risk to higher-risk states and provide a call-to-action for brief interventions to interrupt risk during these transition states.

Funding:

This work was in part supported by the Military Suicide Research Consortium (MSRC), an effort supported by the Office of the Assistant Secretary of Defense for Health Affairs under Award No. W81XWH-16- 2-0004 (PI: Brown), and the Department of Defense under Award No. W81XWH-18-2-002 (PI: Bryan). This publication was made possible through core services and support from the Penn Mental Health AIDS Research Center (PMHARC), an NIH-funded program (P30 MH 097488; Brown), and from the National Institute of Mental Health (5P50MH127511 & R01MH13274001A1). Opinions, interpretations, conclusions and recommendations are those of the authors and are not necessarily endorsed by the MSRC or the Department of Defense.

Footnotes

Conflict of Interest Disclosures: None.

Disclaimer: The views expressed in this presentation are those of the authors and do not necessarily reflect the official policy or position of the Department of Defense nor the U.S. Government. This work was prepared as part of her official duties. Title 17 U.S.C. 105 provides that “Copyright protection under this title is not available for any work of the United States Government.” Title 17 U.S.C. 101 defines a United States Government work as a work prepared by a military service member or employee of the United States Government as part of that person’s official duties.

References

  1. Allan NP, Conner KR, Pigeon WR, Gros DF, Salami TK, & Stecker T (2017). Insomnia and suicidal ideation and behaviors in former and current U.S. service members: Does depression mediate the relations? Psychiatry Research, 252, 296–302. 10.1016/j.psychres.2017.03.009 [DOI] [PubMed] [Google Scholar]
  2. ATUS home. (n.d.). Bureau of Labor Statistics. Retrieved August 6, 2024, from https://www.bls.gov/tus/
  3. Bates D (2016). lme4: Linear mixed-effects models using Eigen and S4. R Package Version, 1, 1. [Google Scholar]
  4. Batterham PJ, Werner-Seidler A, Calear AL, McCallum S, & Gulliver A (2021). Specific aspects of sleep disturbance associated with suicidal thoughts and attempts. Journal of Affective Disorders, 282, 574–579. 10.1016/j.jad.2020.12.150 [DOI] [PubMed] [Google Scholar]
  5. Bernert RA, Kim JS, Iwata NG, & Perlis ML (2015). Sleep Disturbances as an Evidence-Based Suicide Risk Factor. Current Psychiatry Reports, 17(3), 15. 10.1007/s11920-015-0554-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Blais RK, & Geiser C (2019). Depression and PTSD-related anhedonia mediate the association of military sexual trauma and suicidal ideation in female service members/veterans. Psychiatry Research, 279, 148–154. 10.1016/j.psychres.2018.12.148 [DOI] [PubMed] [Google Scholar]
  7. Blanchard EB, Jones-Alexander J, Buckley TC, & Forneris CA (1996). Psychometric properties of the PTSD checklist (PCL). Behaviour Research and Therapy, 34(8), 669–673. 10.1016/0005-7967(96)00033-2 [DOI] [PubMed] [Google Scholar]
  8. Bovin MJ, Marx BP, Weathers FW, Gallagher MW, Rodriguez P, Schnurr PP, & Keane TM (2016). Psychometric properties of the PTSD checklist for diagnostic and statistical manual of mental disorders–fifth edition (PCL-5) in veterans. Psychological Assessment, 28(11), 1379. [DOI] [PubMed] [Google Scholar]
  9. Brown LA, Armey MA, Sejourne C, Miller IW, & Weinstock LM (2016). Trauma history is associated with prior suicide attempt history in hospitalized patients with major depressive disorder. Psychiatry Research, 243, 191–197. 10.1016/j.psychres.2016.06.046 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Brown LA, Chen S, Narine K, Contractor AA, & Oslin D (2020). DSM-5 PTSD symptom clusters and suicidal ideation in veterans. Psychiatry Research, 288, 112942. 10.1016/j.psychres.2020.112942 [DOI] [PubMed] [Google Scholar]
  11. Brown LA, Fernandez CA, Kohn R, Saldivia S, & Vicente B (2018). Pre-disaster PTSD as a moderator of the relationship between natural disaster and suicidal ideation over time. Journal of Affective Disorders, 230, 7–14. 10.1016/j.jad.2017.12.096 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Brown LA, Taylor DJ, Bryan C, Wiley JF, Pruiksma K, Khazem L, Baker JC, Young J, & O’Leary K (2022). Digital phenotyping to improve prediction of suicidal urges in treatment: Study protocol. Aggression and Violent Behavior, 66, 101733. 10.1016/j.avb.2022.101733 [DOI] [Google Scholar]
  13. Brown LA, Zhu Y, Narine K, Simon S, & Akpoduado E (2022). Timing of Suicidal Ideation Among Persons Living With HIV. The Behavior Therapist, 45(8), 298–306. [Google Scholar]
  14. Brüdern J, Hallensleben N, Höller I, Spangenberg L, Forkmann T, Rath D, Strauß M, Kersting A, & Glaesmer H (2022). Sleep disturbances predict active suicidal ideation the next day: An ecological momentary assessment study. BMC Psychiatry, 22(1), 65. 10.1186/s12888-022-03716-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Bryan CJ (2019). A Preliminary Validation Study of Two Ultra-Brief Measures of Suicide Risk: The Suicide and Perceived Burdensomeness Visual Analog Scales. Suicide & Life-Threatening Behavior, 49(2), 343–352. 10.1111/sltb.12447 [DOI] [PubMed] [Google Scholar]
  16. Bryan CJ, Gonzales J, Rudd MD, Bryan AO, Clemans TA, Ray-Sannerud B, Wertenberger E, Leeson B, Heron EA, Morrow CE, & Etienne N (2015). Depression Mediates the Relation of Insomnia Severity with Suicide Risk in Three Clinical Samples of U.s. Military Personnel. Depression and Anxiety, 32(9), 647–655. 10.1002/da.22383 [DOI] [PubMed] [Google Scholar]
  17. Department of Defense. (2022). Annual report on suicide in the military 2021. https://www.dspo.mil/Portals/113/Documents/2021%20ASR/Annual%20Report%20on%20Suicide%20in%20the%20Military%20CY%202021%20with%20CY21%20DoDSER.pdf
  18. Fawcett J, Scheftner WA, Fogg L, Clark DC, Young MA, Hedeker D, & Gibbons R (1990). Time-related predictors of suicide in major affective disorder. The American Journal of Psychiatry, 147(9), 1189–1194. 10.1176/ajp.147.9.1189 [DOI] [PubMed] [Google Scholar]
  19. Franklin JC, Ribeiro JD, Fox KR, Bentley KH, Kleiman EM, Huang X, Musacchio KM, Jaroszewski AC, Chang BP, & Nock MK (2017). Risk factors for suicidal thoughts and behaviors: A meta-analysis of 50 years of research. Psychological Bulletin, 143(2), 187–232. 10.1037/bul0000084 [DOI] [PubMed] [Google Scholar]
  20. Gehrman P, Seelig AD, Jacobson IG, Boyko EJ, Hooper TI, Gackstetter GD, Ulmer CS, & Smith TC (2013). Predeployment Sleep Duration and Insomnia Symptoms as Risk Factors for New-Onset Mental Health Disorders Following Military Deployment. Sleep, 36(7), 1009–1018. 10.5665/sleep.2798 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Geoffroy PA, Borand R, Ambar Akkaoui M, Yung S, Atoui Y, Fontenoy E, Maruani J, & Lejoyeux M (2022). Bad Dreams and Nightmares Preceding Suicidal Behaviors. The Journal of Clinical Psychiatry, 84(1), 22m14448. 10.4088/JCP.22m14448 [DOI] [PubMed] [Google Scholar]
  22. Kalmbach DA, Cheng P, Ahmedani BK, Peterson EL, Reffi AN, Sagong C, Seymour GM, Ruprich MK, & Drake CL (2022). Cognitive-behavioral therapy for insomnia prevents and alleviates suicidal ideation: Insomnia remission is a suicidolytic mechanism. Sleep, 45(12), zsac251. 10.1093/sleep/zsac251 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Khazem LR, Rozek DC, Baker JC, & Bryan CJ (2021). The Marine Suicide Prevention and Intervention REsearch (M-SPIRE) study: A randomized clinical trial investigating potential treatment mechanisms for reducing suicidal behaviors among military personnel. Contemporary Clinical Trials Communications, 21, 100731. 10.1016/j.conctc.2021.100731 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kleiman EM, Turner BJ, Fedor S, Beale EE, Huffman JC, & Nock MK (2017). Examination of real-time fluctuations in suicidal ideation and its risk factors: Results from two ecological momentary assessment studies. Journal of Abnormal Psychology, 126(6), 726. [DOI] [PubMed] [Google Scholar]
  25. Kroenke K, Spitzer RL, & Williams JBW (2003). The Patient Health Questionnaire-2: Validity of a two-item depression screener. Medical Care, 41(11), 1284–1292. 10.1097/01.MLR.0000093487.78664.3C [DOI] [PubMed] [Google Scholar]
  26. LeardMann CA, Matsuno RK, Boyko EJ, Powell TM, Reger MA, Hoge CW, & Millennium Cohort Study. (2021). Association of Combat Experiences With Suicide Attempts Among Active-Duty US Service Members. JAMA Network Open, 4(2), e2036065. 10.1001/jamanetworkopen.2020.36065 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Levis B, Sun Y, He C, Wu Y, Krishnan A, Bhandari PM, Neupane D, Imran M, Brehaut E, & Negeri Z (2020). Accuracy of the PHQ-2 alone and in combination with the PHQ-9 for screening to detect major depression: Systematic review and meta-analysis. Jama, 323(22), 2290–2300. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Littlewood DL, Gooding P, Kyle SD, Pratt D, & Peters S (2016). Understanding the role of sleep in suicide risk: Qualitative interview study. BMJ Open, 6(8). 10.1136/bmjopen-2016-012113 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Littlewood DL, Kyle SD, Carter L-A, Peters S, Pratt D, & Gooding P (2019). Short sleep duration and poor sleep quality predict next-day suicidal ideation: An ecological momentary assessment study. Psychological Medicine, 49(3), 403–411. 10.1017/S0033291718001009 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. May AM, & Klonsky ED (2016). What distinguishes suicide attempters from suicide ideators? A meta-analysis of potential factors. Clinical Psychology: Science and Practice, 23(1), 5–20. 10.1037/h0101735 [DOI] [Google Scholar]
  31. McCall WV, Blocker JN, D’Agostino R, Kimball J, Boggs N, Lasater B, & Rosenquist PB (2010). Insomnia severity is an indicator of suicidal ideation during a depression clinical trial. Sleep Medicine, 11(9), 822–827. 10.1016/j.sleep.2010.04.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. McCarthy MS, Hoffmire C, Brenner LA, & Nazem S (2019). Sleep and timing of death by suicide among U.S. Veterans 2006–2015: Analysis of the American Time Use Survey and the National Violent Death Reporting System. Sleep, 42(8), zsz094. 10.1093/sleep/zsz094 [DOI] [PubMed] [Google Scholar]
  33. Mellman TA, Pigeon WR, Nowell PD, & Nolan B (2007). Relationships between REM sleep findings and PTSD symptoms during the early aftermath of trauma. Journal of Traumatic Stress, 20(5), 893–901. 10.1002/jts.20246 [DOI] [PubMed] [Google Scholar]
  34. Morgan J, Hourani L, Tueller S, Strange L, Lane M, & Lewis G (2017). Effects of Sleep Issues on Suicidal Ideation in a Military Sample: The Mediating Role of Mental Health. Military Behavioral Health, 6, 00–00. 10.1080/21635781.2017.1406415 [DOI] [Google Scholar]
  35. Mysliwiec V, McGraw L, Pierce R, Smith P, Trapp B, & Roth BJ (2013). Sleep Disorders and Associated Medical Comorbidities in Active Duty Military Personnel. Sleep, 36(2), 167–174. 10.5665/sleep.2364 [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Nadorff MR, Fiske A, Sperry JA, Petts R, & Gregg JJ (2013). Insomnia symptoms, nightmares, and suicidal ideation in older adults. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 68(2), 145–152. 10.1093/geronb/gbs061 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Nadorff MR, Nazem S, & Fiske A (2013). Insomnia Symptoms, Nightmares and Suicide Risk: Duration of Sleep Disturbance Matters. Suicide & Life-Threatening Behavior, 43(2), 139–149. 10.1111/sltb.12003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Orff HJ, Drummond SPA, Nowakowski S, & Perlis ML (2007). Discrepancy Between Subjective Symptomatology and Objective Neuropsychological Performance in Insomnia. Sleep, 30(9), 1205–1211. 10.1093/sleep/30.9.1205 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Paxton Willing MM, Pickett TC, Tate LL, Sours Rhodes C, Riggs DS, & DeGraba TJ (2021). Understanding the role of sleep on suicidal ideation in active-duty service members: Implications for clinical practice. Practice Innovations, 6(2), 67–76. 10.1037/pri0000146 [DOI] [Google Scholar]
  40. Perlis ML, Grandner MA, Brown GK, Basner M, Chakravorty S, Morales KH, Gehrman PR, Chaudhary NS, Thase ME, & Dinges DF (2016). Nocturnal Wakefulness as a Previously Unrecognized Risk Factor for Suicide. The Journal of Clinical Psychiatry, 77(6), e726–733. 10.4088/JCP.15m10131 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Pigeon WR, Bishop TM, & Titus CE (2016). The Relationship Between Sleep Disturbance, Suicidal Ideation, Suicide Attempts, and Suicide Among Adults: A Systematic Review. Psychiatric Annals, 46(3), 177–186. 10.3928/00485713-20160128-01 [DOI] [Google Scholar]
  42. Pruitt LD, Smolenski DJ, Bush NE, Tucker J, Issa F, Hoyt TV, & Reger MA (2019). Suicide in the Military: Understanding Rates and Risk Factors Across the United States’ Armed Forces. Military Medicine, 184(Suppl 1), 432–437. 10.1093/milmed/usy296 [DOI] [PubMed] [Google Scholar]
  43. Riemann D, & Voderholzer U (2003). Primary insomnia: A risk factor to develop depression? Journal of Affective Disorders, 76(1–3), 255–259. 10.1016/s0165-0327(02)00072-1 [DOI] [PubMed] [Google Scholar]
  44. Ringer FB, Soberay KA, Rogers ML, Hagan CR, Chu C, Schneider M, Podlogar MC, Witte T, Holm-Denoma J, Plant EA, Gutierrez PM, & Joiner TE (2018). Initial validation of brief measures of suicide risk factors: Common data elements used by the Military Suicide Research Consortium. Psychological Assessment, 30(6), 767–778. 10.1037/pas0000519 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Rogers ML, & Bozzay ML (2024). Daily and Cumulative Sleep Duration as Predictors of Suicidal Desire and Intent: An Ecological Momentary Assessment Study. The Journal of Clinical Psychiatry, 85(2), 23m15164. 10.4088/JCP.23m15164 [DOI] [PubMed] [Google Scholar]
  46. Rubio A, Oyanedel JC, Bilbao M, Mendiburo-Seguel A, López V, & Páez D (2020). Suicidal Ideation Mediates the Relationship Between Affect and Suicide Attempt in Adolescents. Frontiers in Psychology, 11. https://www.frontiersin.org/articles/10.3389/fpsyg.2020.524848 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Rugo KF, Tabares JV, Crowell SE, Baucom BR, Rudd MD, & Bryan CJ (2020). The role of depression and suicidal cognitions as contributors to suicide risk among active duty soldiers reporting symptoms of posttraumatic stress disorder. Journal of Affective Disorders, 265, 333–341. 10.1016/j.jad.2020.01.095 [DOI] [PubMed] [Google Scholar]
  48. Schmied EA, Jun H-J, Glassman LH, Pippard N, & Walter KH (2023). Investigating the effects of suicide exposure among a clinical sample of active duty service members. Journal of Traumatic Stress, 36(2), 310–324. 10.1002/jts.22909 [DOI] [PubMed] [Google Scholar]
  49. Sher L, Braquehais MD, & Casas M (2012). Posttraumatic stress disorder, depression, and suicide in veterans. Cleveland Clinic Journal of Medicine, 79(2), 92–97. 10.3949/ccjm.79a.11069 [DOI] [PubMed] [Google Scholar]
  50. Soberay KA, Hanson JE, Dwyer M, Plant EA, & Gutierrez PM (2019). The Relationship between Suicidal Responses and Traumatic Brain Injury and Severe Insomnia in Active Duty, Veteran, and Civilian Populations. Archives of Suicide Research, 23(3), 391–410. 10.1080/13811118.2018.1479322 [DOI] [PubMed] [Google Scholar]
  51. Tanskanen A, Tuomilehto J, Viinamäki H, Vartiainen E, Lehtonen J, & Puska P (2001). Nightmares as predictors of suicide. Sleep, 24(7), 844–847. [PubMed] [Google Scholar]
  52. Taylor DJ, Lichstein KL, & Durrence HH (2003). Insomnia as a health risk factor. Behavioral Sleep Medicine, 1(4), 227–247. 10.1207/S15402010BSM0104_5 [DOI] [PubMed] [Google Scholar]
  53. Tingley D, Yamamoto T, Hirose K, Keele L, & Imai K (2014). Mediation: R package for causal mediation analysis. [Google Scholar]
  54. Trockel M, Karlin BE, Taylor CB, Brown GK, & Manber R (2015). Effects of Cognitive Behavioral Therapy for Insomnia on Suicidal Ideation in Veterans. Sleep, 38(2), 259–265. 10.5665/sleep.4410 [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Tucker RP, Cramer RJ, Langhinrichsen-Rohling J, Rodriguez-Cue R, Rasmussen S, Oakey-Frost N, Franks CM, & Cunningham CAPT. C. A. (2021). Insomnia and suicide risk: A multi-study replication and extension among military and high-risk college student samples. Sleep Medicine, 85, 94–104. 10.1016/j.sleep.2021.06.032 [DOI] [PubMed] [Google Scholar]
  56. Wang X, Cheng S, & Xu H (2019). Systematic review and meta-analysis of the relationship between sleep disorders and suicidal behaviour in patients with depression. BMC Psychiatry, 19(1), 303. 10.1186/s12888-019-2302-5 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES