Table 2.
SI among military personnel and veterans with PTSD (n = 28).
| Study | Population/sample | Measurement of PTSD | Measurement of SI | Prevalence of SI among those with PTSD | Relationship of PTSD to SI | Notes |
|---|---|---|---|---|---|---|
| Barnes et al. (2012) | 92 male OEF/OIF veterans with combat-related PTSD presenting for VA outpatient PTSD treatment. | Clinical interview (including CAPS for DSM-IV) | Interview question: “Have you had thoughts about death or about killing yourself?” (Yes/No; lifetime) | 16 (17.39%) endorsed lifetime SI.* | Did not analyze relationship. | Exclusion criteria of TBI-related LOC > 30 min or PTA > 24 h. |
| Blakey et al. (2018) | 667 OEF/OIF/OND veterans (81.0% male), active duty personnel, and National Guard and Reserve members who had served since September 11, 2001 and endorsed chronic pain. | SCID-IV | BSS and/or BDI-II item 9 | Not reported. | PTSD was associated with SI on a bivariate level, P = 0.17, p <0.01. After accounting for demographics, chronic pain intensity and interference, and comorbid MH conditions, PTSD remained a significant correlate (OR = 2.25, 95% CI: 1.18–4.24, p = 0.01). | Disagreement in SI scoring on BSS and BDI-II item 9 was present for 65 (51.18%) of SI data.* |
| Braden et al. (2015) | 110 veterans (90.9% male) with a primary mood disorder diagnosis. | SCID-IV | BSS | Not reported. | Accounting for demographics, PTSD was not associated with SI, p > 0.05; however, poorer physical health, prior psychiatric hospitalization, and past SA were associated in this sample. | Participants were excluded for a diagnosis of bipolar disorder, schizophrenia, dementia, organic brain damage, and intellectual disability. |
| Carroll et al. (2017) | 217 male Iraq/Afghanistan-era veterans with combat-related PTSD entering a PTSD residential program. | “Intake procedures” (measure not reported) | Interview question: “Have you ever had serious thoughts of committing suicide?” (Yes/No; lifetime) | 156 (71.89%) endorsed lifetime SI. | Did not analyze relationship. | Inclusion was based on presence of combat-related PTSD diagnosis. Participants were excluded based on presence of psychotic symptoms, unwillingness to stop using drugs/alcohol, and medical conditions impeding engagement in treatment. |
| Cohen et al. (2015) | 186,460 OEF/OIF/OND veterans (90.6% male) with PTSD and an initial VHA visit from January 1, 2007 to September 30, 2011. | VA medical record | VA medical record | 5,988 (3.21%) with PTSD had documented SI.* | Did not analyze relationship. | Veterans with a diagnosis of bipolar disorder or schizophrenia were excluded. |
| Cornelius et al. (2012) | 101 veterans (89.1%) using VHA outpatient behavioral health services. | SCID-IV | BDI item 9 | 2 (1.98%) of those with PTSD endorsed SI. | PTSD was not associated with SI at a bivariate level, χ2= 0.08, p = 0.773. | |
| Corson et al. (2013) | 1,340 veterans (89.4% male) using VHA care with a positive depression screen (PHQ-2). | VA medical record | PHQ-9 item 9 and VA Pocket Card Risk Assessment | 235 (34.81%) of those with PTSD endorsed SI.* | Bivariate associations between PTSD and SI were not significant, p = 0.06. After controlling for age, sex, marital status, branch, PHQ-2 scores, SUD, and when SI was assessed, PTSD was not predictive of SI (AOR = 1.18, 95% CI: 0.93–1.50, p = 0.186). Rather, the only significant MH diagnoses predictive of SI were depression and bipolar/schizophrenia. | |
| Cox et al. (2016) | 289 veterans (89.0% male) receiving Prolonged Exposure Therapy from a VHA clinic. | CAPS for DSM-IV or PSS-I | BDI-II item 9 | 127 (43.94%) endorsed SI. | Did not analyze relationship. | Inclusion criterion of at least 2 timepoints of SI measurement. |
| Denneson et al. (2014) | 465 OEF/OIF veterans (87.5% male) new to VA care who screened positive for depression. | VA medical record | PHQ-9 item 9 and/or VA Pocket Card Risk Assessment | 80 (35.71%) with documented PTSD screened positive for SI.* | PTSD diagnosis was not associated with SI on a bivariate, p = 0.07, or multivariate level (AOR = 1.10; 95% CI: 0.72-1.69, p = 0.67). Multivariate results found depression to be a significant correlate adjusting for sex, age, race/ethnicity, rurality, marital status, SI assessment by MH clinician, and MH comorbidity. | |
| Elbogen et al. (2018) | 2,543 Iraq/Afghanistan-era veterans, active duty personnel, and reserve forces (80.2% male). | SCID-IV | BSS | 205 (28.55%) with PTSD endorsed SI.* | Accounting for demographics and resiliency, childhood abuse, pain, and depression, but not PTSD (OR = 1.34, 95% CI: 0.94–1.90, p = 0.106), were associated with SI. | Most participants were registered at a VA medical facility. |
| Finley et al. (2015) | 211,652 veterans (86.4% male) receiving VA care during FY 2009-2011. | VA medical record | VA medical record | Not reported. | Accounting for demographics and MH comorbidity, PTSD was associated with SI (OR = 2.3, 95% CI: 2.0–2.6, p <0.01), with risk increasing in the presence of MH comorbidity (e.g., depression, SUD). | |
| Ganocy et al. (2016) | 418 Army National Guard personnel (88.0% male) participating in a longitudinal cohort study. | CAPS for DSM-IV (lifetime) | C-SSRS | Not reported. | Bivariate associations between lifetime PTSD and SI were significant, r = 0.24, p <0.01. | The number of participants endorsing lifetime PTSD (28; 6.8%), as well as SI (32; 7.7%), were small. |
| Green et al. (2010) | 497 deployed OEF/OIF veterans (83.1% male). | SCID-IV | BSS | 41 (21.69%) with PTSD had past-week SI. | Those diagnosed with PTSD reported higher BSS scores at a bivariate level, t = −4.98, p <0.001. | Veterans without lifetime trauma exposure were excluded. |
| Kachadourian et al. (2018) | 93 veterans (93.5% male) with comorbid PTSD and alcohol dependence participating in an RCT. | SCID-IV | C-SSRS (lifetime) | 60 (63.83%) with PTSD reported lifetime SI. | Did not analyze relationship. | Exclusion criteria of pregnancy, schizophrenia, schizophrenia-type disorders, bipolar disorder, active SI or homicidal ideation, use of medication likely to influence alcohol consumption, and Prazosin contraindication. |
| Kimbrel et al. (2017) | 3,233 Iraq/Afghanistan-era veterans (79.7% male). | SCID-IV (lifetime) | BSS | Not reported. | Adjusting for sex, lifetime depression, lifetime alcohol use disorder, lifetime non-cannabis drug use disorder, lifetime cannabis use disorder, childhood sexual abuse, and combat exposure, lifetime PTSD diagnosis (OR = 2.08, 95% CI: 1.50–2.89, p <0.001) as well as depression, childhood sexual abuse, combat exposure, and cannabis use disorder were associated with SI. | The majority of veterans were enrolled in VHA care. |
| Kimbrel et al. (2018a) | 1,143 veterans (95.9% male) seeking PTSD treatment. | SCID-IV | BDI-II Item 9 | Not reported. | PTSD was associated with presence of SI (OR = 2.13; 95% CI: 1.43–3.18, p <0.05), accounting for gender, age, combat exposure, and NSSI. | |
| Kittel et al. (2016) | 130 (85.3% male) Iraq and Afghanistan-era veterans enrolled in VHA care. | CAPS for DSM-IV | BSS | Not reported. | A bivariate, positive relationship between current PTSD diagnosis and past week SI severity was significant, p = 0.002. | Participants were excluded if endorsed SI, intent or plan warranting crisis intervention; none excluded based on this criterion. |
| Kopacz et al. (2016) | 472 (94.1% male) veterans admitted for treatment at one of two VA PTSD Residential Rehabilitation Programs | Clinical interview and unreported assessment measures | Affirmative response to: “Have you ever had serious thoughts of committing suicide?” (lifetime) | 336 (71.19%) reported a lifetime history of SI* | Not reported. | |
| Legarreta et al. (2015) | 95 veterans (77.9% male) who reported at least one lifetime DSM−5 Criterion A traumatic event. | Checklist created using SCID-IV, Trauma Symptom Index, HAM-D, Profile of Mood States, and methodical query | C-SSRS (lifetime) | 47 (49.47%) with PTSD endorsed lifetime SI.* | PTSD was associated with lifetime SI at a bivariate level, χ2 = 5.05, p = 0.03. | Exclusion criteria of major sensorimotor handicaps, IQ <80, and psychosis. Unvalidated measure of PTSD. |
| Lemaire and Graham (2011) | 1,740 OEF/OIF veterans (84.1% male) participating in routine VA mental health screening. | VA medical record | Clinical interview | 64 (12.88%) with PTSD reported SI.* | PTSD was associated with documented SI at the bivariate level (OR = 10.02, 95% CI: 4.02–24.97, p <0.001). PTSD was not significant and thus not included in the final forward stepwise multivariate regression predicting SI. Rather, depressive disorders, social support, gender, and prior SA were noted correlates in this model. | |
| Magruder et al. (2012) | 816 veterans (83.9% male) randomly selected from 4 Southeast VA hospitals. | CAPS for DSM-IV | MINI Suicidality module | 44 (44.90%) with PTSD reported current SI.* | PTSD was positively associated with current SI at the bivariate level, p <0.01. In multivariate analyses, PTSD was associated with SI in the presence of other comorbidities (depression, anxiety disorders; OR = 4.02, 95% CI: 1.95–8.29, p <0.05). | Women oversampled. |
| Maguen et al. (2012) | 259 male Vietnam veterans participating in a national survey. | SCID for DSM-III-R | Affirmative response: “Have you ever felt so low that you thought of committing suicide?” (lifetime) | Not reported. | After adjusting for demographic variables, killing experiences, depression, SUD, and PTSD (OR = 3.42, 95% CI = 1.09–10.73, p <0.05) were associated with SI. | Weighted results to population of 1.3 million veterans. |
| Matthews et al. (2012) | 26 male combat-exposed OEF/OIF veterans evaluated at an outpatient mood or psychiatric emergency clinic. | Semi-structured clinical interview | Comprehensive Suicide Risk Assessment (lifetime) | 13 (54.16%) with PTSD endorsed lifetime SI. | Did not analyze relationship. | Exclusion criteria of alcohol/substance dependence in past 30 days; lifetime ADHD; psychotic, bipolar, chronic pain disorders; active medical problems; or claustrophobia. |
| Monteith et al. (2016) | 354 (87.6% male) veterans accessing VHA care. | VA medical record | BSS | Not reported. | PTSD was associated with SI presence, r = 0.12, p <0.05, and severity, r = 0.23, p <0.01. When age, gender, combat exposure, depressive disorders, PTSD diagnosis, negative affect, past SA, military sexual trauma, and an interactive term of military sexual trauma x gender were included in a regression model, lifetime PTSD diagnosis was no longer associated with SI severity, B = 0.08, 95% CI: 0.06–2.06, p = 0.80. | |
| Ribeiro et al. (2012) | 311 military personnel (82.0% male) referred for suicide-focused treatment. | DIS | MSSI | Not reported. | PTSD was associated with SI at the bivariate level, r = 0.20, p <0.01. After accounting for baseline hopelessness, depression, anxiety, drug abuse, alcohol abuse, and insomnia symptoms, PTSD was not associated with SI, p > 0.05. | Results did not account for sociodemographic variables (e.g., gender). |
| Smith et al. (2015) | 832 veterans (gender breakdown not reported) entering a VHA or non-profit PTSD residential program. | Prior primary diagnosis of PTSD (measure not reported) | Interview question: “Have you ever had serious thoughts of committing suicide?” (Yes/No; lifetime) | 194 (71.06%) reported a lifetime history of SI.* | Did not analyze relationship. | Article reported a total sample of 832; however, suicide-related data were only reported for 273 veterans. Exclusion criteria of active psychosis, unwillingness to discontinue substance misuse, and medical conditions that would hinder/prevent engagement in treatment. |
| Suris et al. (2011) | 128 veterans (11% male) currently diagnosed with military sexual assault-related PTSD participating in a PTSD RCT. | CAPS for DSM-IV | BDI-II Item 9 | 59 (46.09%) of sample endorsed SI.* | Did not analyze relationship. | Exclusion criterion of “active suicidality.” |
| Wisco et al. (2014) | 1,649 deployed OEF/OIF/OND veterans (49.9% male). | SCID for DSM-IV (lifetime) | MINI Suicidality module | 375 (30.00%) of veterans with a lifetime diagnosis of PTSD reported current SI.* | Those with PTSD reported SI more frequently at a bivariate level, V = 0.19, p <0.001. After accounting for age, gender, ethnicity, race, combat and postbattle experiences, postdeployment support, depressive symptoms, alcohol problems, and TBI, lifetime PTSD was associated with SI: RR = 2.16, 95% CI: 1.29-3.61, p <0.05. | Participants deemed to be at “high suicide risk” were excluded. Oversampled veterans with probable PTSD. Results were significant for both males and females for gender-stratified. |
*
Calculated using data reported in manuscript.
BDI-II, Beck Depression Inventory-II; BSS, Beck Scale for Suicide Ideation; C-SSRS, Columbia-Suicide Severity Rating Scale; CAPS, Clinician Administered PTSD Scale; DIS, Diagnostic Interview Schedule; DSM-IV, Diagnostic and Statistical Manual of Mental Disorders-IV; FY, fiscal year; HAM-D, Hamilton Rating Scale for Depression; IQ, intelligence quotient; LOC, loss of consciousness; MH, mental health; MINI, Mini International.
Neuropsychiatric Interview; MSSI, Modified Scale for Suicidal Ideation; NSSI, non-suicidal self-injury; OR, odds ratio; PHQ-2: Patient Health Questionnaire-2; PHQ-9: Patient Health Questionnaire-9; PTA, posttraumatic amnesia; PSS-I, Posttraumatic Symptom Interview; PTSD, posttraumatic stress disorder; RCT, randomized clinical trial; RR, relative risk; SCID for DSM-III-R, Structured Clinical Interview for DSM-III-R; SCID-IV, Structured Clinical Interview for DSM-IV; SUD, substance use disorder; TBI, traumatic brain injury; VA, Department of Veterans Affairs; VHA, Veterans Health Administration.