Table 1.

Studies referring to PTSD, suicidality and sleep disturbances (insomnia).

References	Core issue	Sample	Sex (male)	Measurements	Main findings
Betts et al. (46)	Association between the risk of SI in individuals with PTSD symptoms on comorbid sleep disturbance	Participants of the MUSP (n = 2,465) with PTSD-symptoms: 9.4%; proportion of those with PTSD-diagnosis not mentioned	48.0%	Single item (SI), shortened version PSQI (sleep quality), CIDI-Auto (PTSD)	- PTSD symptoms did not directly predict SI when adjusting for MDD symptoms, polyvictimization, and gender - PTSD symptoms had an indirect effect on SI via past-month sleep disturbance
Bishop et al. (49)	Association of sleep disturbance with SI after controlling for age, alcohol dependence, depression, and PTSD	Veterans (n = 654), with PTSD: 24.3%	95,3%	PSS (SI), single item PCL (sleep disturbance), PTSD (PCL)	Sleep disturbance was a predictor of SI, even after controlling for age, alcohol dependence, PTSD, and depression
Bishop et al. (56)	Association among sleep disorders and SA after controlling for several mental disorders, medical comorbidity, and obesity	Veterans (data base review) (n = 60,102, 1:1 case control with no SA), with PTSD: 24.7%	87.1%	Data extraction for SA, sleep disturbance, mental disorders	- Insomnia was associated with SA - Sleep medicine visits were associated with a reduced risk of SA in sleep disordered patients
Britton et al. (57)	Associations among insomnia symptoms, PTSD symptoms and depressive symptoms, IPT variables, and risk for SB	Veterans (n = 392), with PTSD: 41.8%	69.6%	SBQ-R (SB), ISI (insomnia), PCL-M (PTSD)	- Insomnia symptoms may be associated with increased PTSD and depressive symptoms - PTSD emerged as a mediator between insomnia and SI
Bryan et al. (50)	Identifying clinical variables (incl. PTSD) associated with suicidality in military personnel with mTBI	Deployed soldiers (n = 158), with mTBI: 85.4%; proportion of PTSD not mentioned	93%	SBQ-R (suicidality), ISI (insomnia), PCL-M (PTSD)	Suicidality was significantly associated with depression and the interaction of depression with PTSD symptoms
Bryan et al. (58)*	Associations of insomnia severity with SI, and SB	3 samples of active duty military (n = 380); proportion of PTSD not mentioned	76.6%	BSSI (SI), ISI (insomnia), PCL (PTSD)	Association between sleep disturbances and SI (concurrent/prospective) in all samples; this effect was no longer present after adjusting for age, gender, depression, and PTSD
Chakravorty et al. (51)	Association between SI and insomnia symptoms adjusted for socio-demographic, psychiatric and addiction-related variables	Outpatient veterans, misusing alcohol (n = 162), with PTSD: 27%	93%	PSI (SI, SB, SA), single item PSQI (sleep quality), PCL (PTSD)	- After controlling for psychopathology, a significant association between insomnia and SI was no longer present; poor sleep quality remained associated with increased SI - SI was not associated with sleep duration
Davidson et al. (48)	Association between exercise and suicide risk including potential mediators (i.e., sleep disturbance, PTSD symptoms, depression)	Veterans admitted to a residential rehabilitation program for PTSD (n = 346)	81%	8 items of BDI II(suicidality), PSQI (sleep quality), PCL-M without sleep item (PTSD)	- Direct negative association between suicide risk and exercise, indirectly through association with depressive symptoms and sleep quality - PTSD symptoms did not mediate the relation between exercise and suicide risk
Dell'Osso et al. (47)	Impact of alterations in rhythmicity and vegetative function (incl. sleep disturbances) as correlates of suicidality	Civilian inpatients/outpatients with PTSD (n = 65); patients with current (n = 20) or former depression (n = 14) were excluded	50.8%	MOOD-SR (SI/SA/sleep disturbances)	All MOODS-SR sub-domains (rhythmicity, sleep, appetite/weight, physical symptoms) were associated with an increased likelihood of SI; changes in appetite/weight were associated with greater OR of SA
Don Richardson et al. (52)	Association between SI and sleep after controlling for probable PTSD, MDD, GAD, AUD	Veterans (n = 324)/active duty military (n = 80), with PTSD: 72.8%	92.4%	Single item PHQ-9 (SI), quantitative single item (sleep disturbances resp. nightmares), PCL-M (PTSD)	- Sleep disturbances did not predict SI - Probable MDD emerged as a significant predictor
Don Richardson et al. (59)	Mediating role of depression in the relationship between: 1) sleep disturbances and SI, and 2) trauma-related nightmares and SI after controlling for PTSD-, anxiety- and alcohol-use-severity	CAF personnel (17.6%)/veterans (82.4%), (n = 663), with PTSD: 72.6%	91%	Single item PHQ-9 (SI), quantitative single item (sleep disturbances resp. nightmares), PCL-M (PTSD)	Sleep disturbances were associated with SI as a function of depressive symptoms
Fisher et al. (60)	The moderating role of agitation within the relationship between insomnia and current SI	U.S. military personnel (n = 937); proportion of PTSD is not mentioned	75.3%	BSSI/DSI-SS (SI), ISI (insomnia), PCL-M (PTSD)	- Significant association between insomnia and SI only at high levels of agitation - PTSD symptoms, depressive symptoms, and lifetime number of SA were each associated with greater levels of agitation, insomnia, and current SI
Kachadourian et al. (61)	Association between individual symptoms of PTSD and measures of functioning, quality of life, and SI	U. S. military veterans (n = 1,484), with PTSD: 10.9%, trauma-exposed: 85,4%	89.8%	Single item on the PHQ-9 (SI), PCL-5 (sleep difficulties/ nightmares/PTSD)	- Sleep difficulties explained problems in physical functioning/quality of life after adjustment for severity of PTSD/depressive symptoms - No association between SI and sleep difficulties
Kim et al. (62)	The mediating role of AUD and insomnia in the relationship between PTSD symptoms and SI	Korean firefighters (n = 7,190), with PTSD: 3.6%	90%	PHQ-9 suicide item (SI), AIS (insomnia), PCL (PTSD)	AUD and insomnia mediated the relationship between PTSD symptoms and SI
Luxton et al. (63)	Prevalence and impact of short sleep duration in redeployed OIF soldiers	Redeployed OIF soldiers (n = 2,738), with PTSD: 15.4%	96%	HRA II (SA), 2 items (sleep duration/sleep quality), PC-PTSD (PTSD)	- SSD was a significant predictor of suicide risk - SSD was the strongest predictor of PTSD
McClure et al. (64)	To determine the prevalence of factors that may serve as warnings of acute suicidality risk	Veterans attending an urgent care psychiatric clinic (n = 473), with PTSD: 49%	89%	SWS survey (SI/insomnia/ hypersomnia), PC-PTSD (PTSD)	- Past week SI and sleep disturbances were among others a highly prevalent warning sign - 97% endorsed at least one warning sign, participants with MDD and/or PTSD endorsed the largest number of warning signs
Morgan et al. (65)	The relationship between sleep issues, mental health (perceived stress, PTSD symptoms, and depressive symptoms), and SI	Military service members (n = 891), with PTSD: 13%	95.5%	Single item (SI), PROMIS (sleep disturbances), PCL-C (PTSD)	PTSD, perceived stress, and depressive symptoms mediated the relationship between sleep issues and SI; after accounting for mental health symptoms, sleep no longer had a direct effect on SI
Pigeon et al. (53)*	Role of sleep disturbance in time to suicide since the last treatment visit among veterans receiving VHA services	Suicide decedents (n = 423) Visit of the VHA (n = 381); proportion of PTSD is not mentioned	99.7%	Chart review for number of days between last visit and death, sleep disturbances, and psychiatric symptoms	Veterans with sleep disturbance died sooner after their last visit than did those without sleep disturbance, after adjusting for the presence of mental health or substance use symptoms, age, and region
Pigeon et al. (66)*	bCBTi delivered to veterans endorsing SI with a diagnosis of MDD and/or PTSD	Veterans (n = 54, RCT 1:1 TAU vs. TAU plus bCBTi); proportion of PTSD not mentioned	80%	C-SSRS (SI), ISI (insomnia), PCL-M (PTSD)	- No significant effect of bCBTi on SI intensity - Effects were large on insomnia and depression with no effect on PTSD
Ribeiro et al. (54)*	Relationship between insomnia symptoms and SI/SB after controlling for depressive symptom severity, hopelessness, PTSD diagnosis, anxiety symptoms, drug and alcohol abuse	Military personnel (n = 311), with PTSD: about 20%	82%	MSSI (SI), insomnia symptom index (sleeplessness), diagnosis (PTSD)	- Insomnia symptoms were cross-sectionally associated with SI - Insomnia symptoms were unique predictors of SA longitudinally after controlling for baseline self-insomnia resp. depressive symptoms and hopelessness
Ribeiro et al. (67)	Association between PTSD status and functional impairment (sleep quality, alcohol use, social problem-solving, work and social adjustment) among suicidal military inpatients	Suicidal military psychiatric inpatients and a lifetime history of at least one SA (n = 166), with PTSD: 38%	65%	C-SSRS (SI/SB), PSQI (sleep quality), MINI (PTSD)	- Patients with PTSD reported disturbed sleep and reduced social and work adjustment, association was no longer significant after adjusting for gender and psychiatric comorbidity - Those with a greater number of psychiatric comorbidities demonstrated higher likelihood of meeting PTSD criteria
Richardson et al. (68)	The relationship between insomnia, SI, and past-year mental health status	Canadian Regular Forces personnel (n = 6,700), with PTSD: 5.3%	86.1%	Single item (past-year SI/insomnia), WHO-CIDI (PTSD)	- Both insomnia and number of mental health conditions incrementally increased the risk of SI - Insomnia significantly increased the odds of SI, but only among individuals with no or one mental health condition
Selaman et al. (69)	To determine specific DSM-IV symptoms of PTSD that are independently associated with SA	Data from wave 2 of the NESARC (n = 34,653), with PTSD: n = 2,322	27.9%	Single item (nightmares/sleep disturbances/SA), DSM-IV-criteria (PTSD)	- Increasing numbers of re-experiencing and avoidance symptoms were correlated with SA - No association between SA and sleep disturbances (A)OR about 0,6
Swinkels et al. (55)	Association of sleep duration and sleep quality with mental health and SI	U.S. Afghanistan/Iraq era veterans (n = 1,640), with PTSD: 31%	80%	BSSI (SI), PSQI-A (sleep quality/ duration), SCIDI/P (PTSD)	- Very SSD (≤ 5 h of sleep) and LSD (≥ 9 h) were each (after adjusting for diverse covariates) associated with increased odds of current PTSD, MDD, and smoking - Poor sleep quality was associated with PTSD, PD, MDD, SI, and risky drinking
Wang et al. (70)*	Association of pre-deployment insomnia with post-deployment PTSD and SI	U. S. Army soldiers (n = 8,558, cross-sectional, n = 4,645, longitudinal), with PTSD 11.9%	94.7%	C-SSRS (suicidality), items of the Brief Insomnia Questionnaire (insomnia), PCL (PTSD)	Pre-deployment insomnia was associated with increased risk of post-deployment PTSD and SI even after adjusting for socio-demographic characteristics and prior deployment history

^*Denotes studies with longitudinal designs; all others are cross-sectional studies.

Cursive references denotes studies on civilian samples.