Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Jan 1.
Published in final edited form as: J Psychiatr Res. 2020 Dec 3;133:60–66. doi: 10.1016/j.jpsychires.2020.12.001

PTSD Symptoms Among Trauma-Exposed Adults Admitted to Inpatient Psychiatry for Suicide-Related Concerns

Ian H Stanley 1,2, Brian P Marx 1,2, Terence M Keane 1,2, Anka A Vujanovic 3
PMCID: PMC7856162  NIHMSID: NIHMS1651819  PMID: 33310501

Abstract

Individuals admitted to inpatient psychiatry for suicide-related concerns are at increased risk of suicide post-discharge, necessitating an understanding of factors, such as posttraumatic stress disorder (PTSD), that are associated with suicide-related hospitalizations. This study examined if individuals admitted for suicide-related concerns were more likely than those admitted for other reasons to have elevated PTSD symptoms or a probable PTSD diagnosis. We also examined the moderating role of impulsivity. Participants were 188 trauma-exposed adult psychiatric inpatients (M[SD]age=33.6y[11.7y], 63.3% male, 46.3% white). We used the Life Events Checklist for DSM-5, PTSD Checklist for DSM-5, Beck Scale for Suicide Ideation, and Barratt Impulsiveness Scale-11 to assess trauma exposure, PTSD symptoms, suicidal ideation severity, and impulsivity, respectively. We controlled for trauma load, number of psychiatric diagnoses, and comorbid depressive and substance use disorders. Patients admitted for suicide-related concerns (55.3%; n=104), compared with those admitted for other reasons (44.7%; n=84), had more severe PTSD symptoms, corresponding to medium-to-large effect sizes; associations were stronger at higher levels of impulsivity. Additionally, patients admitted for suicide-related concerns were nearly four times more likely than their counterparts to screen positive for a provisional PTSD diagnosis. Among the subset of individuals admitted for suicide-related concerns, greater PTSD symptoms were associated with more severe suicidal ideation. In sum, PTSD symptoms are elevated among psychiatric inpatients admitted for suicide-related concerns, and among this subgroup, greater PTSD symptom severity covaries with suicidal ideation severity. Screening for and treating PTSD, and attending to cooccurring impulsivity, in psychiatric inpatients may reduce suicide risk.

Keywords: trauma, posttraumatic stress disorder, suicide, inpatient psychiatry, acute care

The prevalence of suicide in the United States has steadily increased over the past two decades (CDC, 2020). Addressing the rising suicide rate requires a comprehensive approach that addresses biopsychosocial concomitants, including psychiatric disorders such as posttraumatic stress disorder (PTSD). PTSD is associated with increased risk for suicidal thoughts and behaviors (Panagioti et al., 2009, 2012; Stanley et al., 2019), and is one of the few psychiatric disorders that differentiates individuals who think about suicide and those who engage in suicidal behaviors (Nock et al., 2009). Indeed, individuals with a PTSD diagnosis (Gradus et al., 2010) or elevated PTSD symptoms (Cooper et al., 2020) are at increased risk for suicide mortality.

Whereas the extant literature has focused on the intersection of PTSD and suicidal thoughts and behaviors, other suicide-related outcomes are important to consider, including suicide-related psychiatric admissions. A recent meta-analysis found that in the first month following discharge from inpatient psychiatry, the pooled suicide rate was 2,060 per 100,000 person-years, which is nearly 200 times greater than the global suicide rate (Chung et al., 2019). Rates were even higher for individuals admitted for suicidal thoughts and/or suicidal behaviors (i.e., 6,210 per 100,000 person-years; Chung et al., 2019). The elevated suicide risk for patients admitted for suicide-related concerns persists well beyond the first month (Chung et al., 2017). Given the strong link between PTSD and suicidal thoughts and behaviors, a considerable proportion of individuals admitted to inpatient psychiatry for suicide-related concerns may have a PTSD diagnosis or elevated PTSD symptoms. If so, this suggests that screening for and treating PTSD during hospitalization, and prioritizing PTSD treatment in the post-discharge period, may yield decreases in suicide risk during the high-risk post-discharge period.

In this context, it is crucial to consider the heterogeneity of PTSD. PTSD is defined in part as the development of characteristic symptoms—intrusions, avoidance, negative alterations in cognitions and mood, and hyperarousal—that endure for at least one month following exposure to a traumatic event (American Psychiatric Association, 2013). PTSD symptom presentations are complex, and longitudinal investigations have shown that the hyperarousal symptom cluster, relative to the other symptom clusters, uniquely predicts the development of suicidal ideation (Panagioti et al., 2017) and suicide attempts (Stanley et al., 2019). PTSD-related hyperarousal symptoms more broadly (e.g., agitation, insomnia) may be central to the phenomenology of acute suicidal crises (Joiner & Stanley, 2016), including those that culminate in an inpatient psychiatric hospitalization. Accordingly, PTSD-related hyperarousal symptoms may be especially elevated among patients admitted for suicide-related concerns.

Furthermore, an important feature of both PTSD and suicidality to consider in this context is impulsivity. Impulsivity is a multifaceted construct that reflects “a predisposition toward rapid, unplanned reactions to internal or external stimuli without regard to the negative consequences of these reactions to the impulsive individual or to others” (Moeller et al., 2001, p. 1784). PTSD symptom severity and impulsivity are positively correlated (e.g., Roley et al., 2017), and among individuals with PTSD, there is a robust link between impulsivity and suicide risk (Kotler et al., 2001). It is plausible that individuals with PTSD who are prone to reckless action (cf. high impulsivity) are more likely than individuals with PTSD who are less prone to reckless action (cf. low impulsivity) to be admitted for suicide-related concerns. Indeed, a behavioral repertoire characterized by reckless action is associated with a greater risk for nonfatal suicide attempts (Stanley et al., 2018), which might prompt psychiatric hospitalizations. Thus, impulsivity might moderate the association between PTSD symptoms and psychiatric hospitalization for suicide-related concerns.

The Present Study

In this study, we examined differences in PTSD symptom severity and probable PTSD diagnostic status between patients admitted to inpatient psychiatry for suicide-related concerns (i.e., suicidal ideation, intent, and/or behaviors) and patients admitted for other reasons (e.g., disorientation, substance use without cooccurring suicidality). Specifically, we hypothesized: (1) patients admitted for suicide-related concerns, compared with those admitted for different reasons, would (a) have greater PTSD symptoms overall and for each symptom cluster and (b) be more likely than their counterparts to screen positive for a provisional PTSD diagnosis based on a previously validated clinical cutoff score; (2) the hyperarousal symptom cluster, relative to the other symptom clusters, would predict unique variance in suicide-related psychiatric admissions; (3) greater PTSD symptoms would be associated with more severe current suicidal ideation among the subset of individuals admitted for a suicide-related concern; and (4) greater facets of impulsivity (i.e., attentional, motor, and non-planning) would each potentiate the association between PTSD symptoms and suicide-related hospitalizations. Finally, individuals with multiple prior psychiatric admissions have higher post-discharge suicide rates than those without a history of psychiatric hospitalizations (Chung et al., 2017). Thus, we conducted exploratory analyses to examine if the number of prior same-facility psychiatric hospitalizations was positively associated with suicidal ideation severity, and further, if this association was stronger for those with compared to without a provisional PTSD diagnosis. We predicted that the forgoing would be true even after controlling for trauma load, the number of psychiatric diagnoses (as a proxy of clinical severity), and the presence/absence of diagnoses highly comorbid with PTSD (i.e., depressive and substance use disorders; Gradus, 2017; Wisco et al., 2012).

Methods

Participants and Procedures

Data were obtained from a larger study that examined PTSD and treatment outcomes among psychiatric inpatients (Vujanovic et al., 2017). Participants were 188 trauma-exposed adult psychiatric inpatients in a large metropolitan area in the southern U.S. (Table 1). Inclusion criteria were: (1) aged ≥18 years and (2) history of trauma exposure per the Life Events Checklist for DSM-5 (LEC-5; Weathers, Blake, et al., 2013b). Participants were excluded if they had significant cognitive impairment (Mini-Mental State Examination [MMSE] score ≤20; Folstein et al., 1975). Individuals were screened by clinical staff for a trauma history within 24 hours of admission. Within five days of admission, study personnel approached potentially eligible participants to obtain written informed consent. Participants were administered the MMSE, and eligible individuals completed the battery of questionnaires. Participants did not receive compensation. The relevant institutional review boards approved study procedures.

Table 1.

Sociodemographic Characteristics of Participants

Characteristic N Valid %
Sex
 Female 69 36.7
 Male 119 63.3
Race/Ethnicity
 White/Caucasian 87 46.3
 Black/African American 67 35.6
 Asian 4 2.1
 Native Hawaiian or Other Pacific Islander 1 0.5
 Hispanic or Latino/a 29 15.4
Marital Status
 Single 125 66.5
 Married 19 10.1
 Divorced 20 10.6
 Separated 16 8.5
 Cohabitating 5 2.7
 Widowed 3 1.6
Education
 Did Not Complete High School 48 26.2
 High School Graduate/GED 70 38.3
 Some College 46 25.1
 College Graduate 14 7.7
 Graduate Degree 5 2.7
 Missing 5 --
Open in a new tab

Note. N=188. Participants were on average 33.6 years old (SD=11.7).

Measures

Medical Records Review

Study personnel reviewed participants’ electronic medical records (EMR) to determine if suicidality was the basis for admission (Yes/No). Suicidality as a basis for admission was defined as significant suicidal ideation, including intent or plan, and/or suicidal behavior that necessitated hospitalization for stabilization (i.e., safety concerns were primary). Study personnel also extracted psychiatric diagnoses from participants’ EMR; diagnoses were derived from psychiatrists’ clinical interviews at intake.

Life Events Checklist for DSM-5 (LEC-5; Weathers, Blake, et al., 2013b)

The LEC-5 is a 17-item self-report measure of lifetime exposure to various traumatic events. Three modifications to the LEC-5 were made: (1) three items were removed (i.e., “exposure to toxic substance,” “severe human suffering,” “sudden accidental death”); (2) two items were added (i.e., “childhood physical abuse,” “childhood sexual abuse”); and (3) the answer choices were limited to querying if the event had “happened to you.” These changes were made to reduce the potential for false positives (e.g., in the case of a patient with psychotic symptoms indicating exposure to toxic substances that is part of their delusional content), add a childhood timeframe for the report of sexual and physical abuse, and reduce participant burden. The LEC-5 is based on the LEC; the psychometric properties of the original LEC, including test-retest reliability and convergent validity, are robust (Gray et al., 2004).

PTSD Checklist for DSM-5 (PCL-5; Weathers et al., 2013)

The PCL-5 is a 20-item self-report measure of PTSD symptoms per DSM-5 diagnostic criteria (American Psychiatric Association, 2013). Study personnel instructed participants to complete the PCL-5 about their worst traumatic event identified on the LEC-5. Participants rated how much they were bothered by each symptom over the past month, using a five-point scale (0=Not at all to 4=Extremely). Responses were summed to generate a total score (range: 0–80) as well as symptom cluster scores; higher scores reflect more severe PTSD symptomatology. We ascertained a provisional PTSD diagnosis via a cutoff score of ≥33 (Bovin et al., 2016). The PCL-5 has strong psychometric properties, including internal consistency, test-retest reliability, and convergent and discriminant validity (Bovin et al., 2016; Keane et al., 2014). Within this sample, the PCL-5 total score (α=.95) demonstrated excellent internal consistency; each of the symptom clusters demonstrated acceptable-to-good internal consistency (α=.77-.89).

Beck Scale for Suicide Ideation (BSS; Beck & Steer, 1991)

The BSS is a 21-item self-report measure of the severity of past-week suicidal symptoms. We used the first five screening items of the BSS to generate a total score (i.e., BSS-5), consistent with past research (Koldsland et al., 2012; Pachkowski et al., 2019; Shahnaz et al., 2018). Participants rated each BSS-5 item on a 3-point scale ranging from 0 to 2; higher scores indicate more severe suicidal ideation. The BSS has strong psychometric properties, including internal consistency, test-retest reliability, and concurrent validity (Batterham et al., 2015). Within this sample, the BSS-5 evinced good internal consistency (α=.87).

Barratt Impulsiveness Scale-11 (BIS-11; Patton et al., 1995)

The BIS-11 is a 30-item self-report measure of three facets of impulsivity: attentional (e.g., “I don’t ‘pay attention’”), motor (e.g., “I do things without thinking”), and non-planning (e.g., “I say things without thinking”). Patients rated items on a 4-point scale (1=Rarely/Never to 4=Almost Always/Always). Responses are reverse-scored as appropriate and summed to generate subscale scores ranging from 8–32, 11–44, and 11–44, respectively; higher scores indicate greater impulsivity. Given the multifaceted nature of impulsivity, examining BIS-11 subscale scores is recommended (Stanford et al., 2009). The BIS-11 has strong psychometric properties (Stanford et al., 2009), and demonstrated acceptable internal consistency in the present sample (α=.79).

Data Analytic Approach

We first screened variables for departures from normality; all variables were within acceptable ranges (i.e., skewness and kurtosis ≤2). To test hypotheses 1a and 1b, we used one-way analysis of covariance (ANCOVA) and logistic regression, respectively. To test hypothesis 2, we used logistic regression. To test hypothesis 3, we conducted moderation analyses using the PROCESS macro within SPSS (Hayes, 2013). Consistent with recommendations (Stanford et al., 2009), each BIS-11 subscale was examined separately; predictor variables were centered around their means. To test hypothesis 4, we used linear regression. Across models, we adjusted for trauma load (LEC-5 total score), number of psychiatric diagnoses (to account for clinical severity), and the presence/absence of both a depressive disorder (e.g., major depressive disorder) and substance use disorder (e.g., cocaine abuse) diagnosis. In exploratory analyses, we conducted moderation analyses using the same process as described for hypothesis 3. Missing data were generally minimal (1.0% for the BSS-5). However, a considerable proportion of participants did not have BIS-11 data (22.3%). Results of the Little’s missing completely at random (MCAR) test indicated that the data were MCAR (χ2 [42]=34.00, p=.806), suggesting that the patients who completed the BIS-11 represent a random subsample of the larger sample and thus the missing data do not introduce bias. Corroborating this finding, there were no significant differences in the number of psychiatric diagnoses, presence of a depressive or substance use disorder diagnosis, LEC-5 trauma load, BSS-5 suicidal ideation severity, and PCL-5 total score between those who completed the BIS-11 and those who did not (ps>.05). Thus, we used listwise deletion for the analyses that used the BIS-11 (only one of our four hypotheses used the BIS-11). We conducted all analyses using SPSS version 23.

Results

Per study inclusion criteria, all (100%; n=188) patients reported a non-zero score on the LEC-5, suggesting at least some degree of lifetime exposure to traumatic events (M[SD]=5.9[3.5]; Table 2). The average level of total PCL-5 PTSD symptoms was 29.0 (SD=19.0; Table 3). As expected, PTSD symptom cluster scores were highly correlated (rs=.65-.85, ps <.001; Table 3). Using the PCL-5 cutoff of ≥33, 41.5% (n=78) of patients met for a provisional PTSD diagnosis. Only 3.2% (n=6) of patients in this trauma-exposed sample had a PTSD diagnosis per EMR review.1 In terms of other clinical characteristics, participants had on average 2.1 (SD=1.3) total psychiatric diagnoses. A total of 29.8% (n=56) of patients had a depressive disorder diagnosis and 51.6% (n=97) had a substance use disorder diagnosis; additional diagnostic data are presented in Table 2.

Table 2.

Clinical Characteristics of Participants

Characteristic N Valid %
Psychiatric Diagnosis Category†
 Substance Use Disorders 97 51.6
 Bipolar and Related Disorders 86 45.7
 Depressive Disorders 56 29.8
 Psychotic Spectrum Disorders 42 22.3
 Personality Disorders 19 10.1
 Posttraumatic Stress Disorder 6 3.2
 Anxiety Disorders 5 2.7
LEC-5 Trauma Exposure Types†,‡
 Transportation Accident 117 62.2
 Physical Assault 114 60.6
 Natural Disaster 110 58.5
 Childhood Physical Abuse 93 49.5
 Serious Accident 85 45.2
 Assault with a Weapon 70 37.2
 Witnessed Sudden Violent Death 63 33.5
 Sexual Assault 62 33.0
 Life-Threatening Illness or Injury 57 30.3
 Childhood Sexual Abuse 53 28.2
 Fire or Explosion 52 27.7
 Other Unwanted Sexual Experiences 51 27.1
 Causing Serious Injury to Someone Else 38 20.2
 Captivity 34 18.1
 Combat War Zone Exposure 27 14.4
 Other Stressful Event or Experience 76 40.4
Number of Days of Current Hospitalization, M(SD) -- 8.4(4.2)
Number of Prior Same-Facility Psychiatric Hospitalizations, M(SD) -- 1.4(2.7)
Open in a new tab

Note. N=188. LEC-5=Life Events Checklist for DSM-5 (Weathers, Blake, et al., 2013).

†

Response options were not mutually exclusive.

‡

Reflects the number and percentage of participants answering “happened to me” to this question.

Patients admitted for suicide-related concerns, compared with those admitted for non-suicide-related reasons, were significantly more likely to have a depressive disorder diagnosis (40.4% [42/104] vs. 16.7% [14/84], χ2 [1]=12.50, p < .001) and significantly less likely to have a psychotic-spectrum disorder diagnosis (13.5% [14/104] vs. 33.3% [28/84], χ2 [1]=10.58, p=.001). There were no significant between-group differences regarding a diagnosis of a substance use disorder (54.8% [57/104] vs. 47.6% [40/84], χ2 [1]=0.96, p=.327), anxiety disorder (3.8% [4/104] vs. 1.2% [1/84], χ2 [1]=1.27, p=.261), bipolar-spectrum disorder (44.2% [46/104] vs. 47.6% [40/84], χ2 [1]=0.22, p=.643), or personality disorder (12.5% [13/104] vs. 7.1% [6/84], χ2 [1]=1.47, p=.226).

Table 3.

Descriptive Statistics and Bivariate Correlations for Study Variables

Variable M / % Yes SD 1 2 3 4 5 6 7 8
1. Suicide-Related Psychiatric Admission 55.3% - —
2. BSS-5 Suicidal Ideation 1.6 2.4 .29** —
3. PCL-5 Intrusions 7.3 5.5 .24** .34** —
4. PCL-5 Avoidance 2.9 2.3 .24** .32** .71** —
5. PCL-5 NACM 10.1 7.2 .26** .47** .76** .69** —
6. PCL-5 Hyperarousal 8.7 5.9 .26** .43** .72** .65** .85** —
7. PCL-5 Total Score 29.0 19.0 .28** .45** .89** .79** .95** .92** —
8. Provisional PTSD Dx† 41.5% - .26** .38** .76** .65** .79** .78** .84** —
9. BIS-11 Attentional Impulsivity 18.75 4.83 .21* .35** .47** .38** .46** .51** .52** .42** —
10. BIS-11 Motor Impulsivity 26.20 5.66 .11 .36** .35** .29** .38** .44** .42** .31** .61** —
11. BIS-11 Non-Planning Impulsivity 26.90 5.76 23** .21* .23** .17 .25** .32** .28** .24** .46** .22* —
Open in a new tab

Note. BIS-11=Barratt Impulsiveness Scale-11. BSS-5=Beck Scale for Suicide Ideation-5. PCL-5=PTSD Checklist for DSM-5. Dx=Diagnosis. NACM=Negative alterations in cognitions and mood. PTSD=posttraumatic stress disorder.

†

PCL-5 total score ≥ 33 (Bovin et al., 2016).

*

p < .05.

**

p < .01.

Overall, 55.3% (n=104) of patients were admitted for suicide-related concerns. The mean level of BSS-5 suicidal ideation was 1.6 (SD=2.4) for the full sample; BSS-5 scores were missing for three participants. BSS-5 scores were significantly higher for the subsample of individuals admitted for suicide-related concern (M[SD]=2.2[2.7] vs. 0.8[1.5]; F[1, 183]=16.24, p<.001, pη2=.08).

Primary Analyses

Patients admitted for suicide-related concern had significantly greater overall PTSD symptoms compared with patients admitted for reasons not suicide-related (M[SD]=33.7[20.0] vs. 23.2[15.8]; F[1, 182]=22.34, p < .001, pη2=.11), adjusting for trauma load, number of psychiatric diagnoses, as well as the presence of depressive disorder and substance use disorder diagnoses. In adjusted analyses, the elevated levels of PTSD symptoms persisted for each PTSD symptom cluster (see Table 4). When examining each symptom cluster simultaneously (in the same model) in the cross-sectional prediction of a suicide-related psychiatric admission, none of the individual PTSD symptom clusters predicted unique variance in the outcome (ps>.05).

Table 4.

Differences in PTSD Symptoms as a Function of Suicide-Related Psychiatric Admission, Adjusted for Covariates

PTSD Variable Suicide-Related Admission (n=104) Non-Suicide-Related Admission (n=84) F(1, 182) P ηp2
M SD M SD
PCL-5 Total Score 33.7 20.0 23.2 15.8 22.34 < .001 .11
PCL-5 Criterion B: Intrusions 8.5 5.9 5.8 4.6 14.93 < .001 .08
PCL-5 Criterion C: Avoidance 3.4 2.4 2.3 2.0 13.90 < .001 .07
PCL-5 Criterion D: Negation Cognitions and Mood 11.7 6.2 8.0 6.4 19.26 < .001 .10
PCL-5 Criterion E: Hyperarousal 10.1 6.2 7.0 5.1 19.49 <. 001 .10
Open in a new tab

Note. N=188. PCL-5=PTSD Checklist for DSM-5 (Weathers, Litz, et al., 2013). PTSD=posttraumatic stress disorder. Covariates include: overall trauma load (total score on the Life Events Checklist for DSM-5; Weathers, Blake, et al., 2013), number of psychiatric diagnoses, depressive disorder diagnosis (Yes/No), and substance use disorder diagnosis (Yes/No).

Patients admitted for suicide-related concerns were significantly more likely than those admitted for other reasons to have a provisional PTSD diagnosis per the PCL-5 clinical cutoff of ≥33 (52.9% [55/104] vs. 27.4% [23/84]; aOR=3.90, 95% CI=1.94–7.83, p<.001), while controlling for trauma load, number of psychiatric diagnoses, depressive disorder diagnoses, and substance use disorder diagnoses. Controlling for these same clinical variables, among the subsample of individuals admitted for suicide-related concerns (n=104), greater overall PTSD symptoms were associated with more severe suicidal ideation (β=.45, p<.001).

Regarding moderation analyses, first, there was a statistically significant interaction between PTSD symptoms and attentional impulsivity (B=.01, SE<.01, p=.010). The effect of PTSD symptoms on suicide-related hospitalizations was significant for those at mean (B=.06, SE=.02, p<.001) and high (+1 SD; B=.09, SE=.02, p<.001) levels of attentional impulsivity, but not for those at low levels (−1 SD; B=.02, SE=.02, p=.352). Second, there was a statistically significant interaction between PTSD symptoms and motor impulsivity (B=.01, SE<.01, p=.037). The effect of PTSD symptoms on suicide-related hospitalizations was significant for those at mean (B=.06, SE=.02, p<.001) and high (+1 SD; B=.08, SE=.02, p<.001) levels of motor impulsivity, but not for those at low levels (−1 SD; B=.03, SE=.02, p=.102). Third, there was no significant interaction between PTSD symptoms and non-planning impulsivity in the association with suicide-related hospitalizations (B<.01, SE<.01, p=.524).

Exploratory Analyses

Patients had an average of 1.4 (SD=2.7) prior same-facility psychiatric hospitalizations. There was a significant interaction between the number of prior psychiatric hospitalizations and the presence of a provisional PTSD diagnosis in the cross-sectional prediction of suicidal ideation severity (B=.25, SE=.12, p=.037). The form of the interaction indicated that the positive association between the number of prior inpatient psychiatric hospitalizations and current suicidal ideation severity was significant for those with (B=.26, SE=.08, p=.002) but not for those without (B=.01, SE=.09, p=.911) a provisional PTSD diagnosis.

Discussion

This study examined differences in PTSD symptom severity and the prevalence of a probable PTSD diagnosis between patients hospitalized for suicide-related concerns and patients admitted for other reasons. Patients admitted for suicide-related reasons, relative to their counterparts, had significantly greater PTSD symptom levels and were nearly four times more likely to have a probable PTSD diagnosis. Higher levels of impulsivity accentuated the relation between PTSD symptoms and suicide-related psychiatric hospitalizations. Moreover, among the subsample of patients admitted for suicide-related concerns, more severe PTSD symptoms were associated with greater current suicidal ideation severity. The forgoing associations persisted even after controlling for clinically- and theoretically-relevant covariates. Pending replication utilizing a multimodal assessment of constructs, results suggest that PTSD, specifically, rather than distress, broadly, might be driving the association with suicide-related psychiatric hospitalizations. Effect sizes were medium-to-large in magnitude, underscoring clinical significance.

Multiple studies have found that PTSD is associated with increased risk for suicidal thoughts and behaviors (Cooper et al., 2020; Gradus et al., 2010; Nock et al., 2009; Panagioti et al., 2009, 2012; Stanley et al., 2019). Our findings suggest that PTSD is also associated with inpatient psychiatric hospitalization for suicide-related concerns. Our findings further suggest that the externalizing subtype of PTSD, characterized in part by higher levels of impulsivity (Wolf et al., 2012), might be at greatest risk of suicide-related psychiatric hospitalizations. This dovetails with past work showing that, in the context of PTSD, impulsivity is associated with engagement in risky behaviors (Weiss et al., 2015), including suicidal behaviors (Kotler et al., 2001). Understanding factors related to suicide-related hospitalizations is critical to inform suicide prevention efforts during hospitalization and the post-discharge period.

Concerningly, risk for suicide mortality is markedly elevated in the period following discharge from inpatient psychiatry, especially for patients admitted for suicide-related concerns (Chung et al., 2017, 2019). Targeting PTSD—through rigorous assessment, treatment, and post-discharge referral/follow-up—might be one conduit to preventing post-discharge suicide. Encouragingly, treating PTSD appears to reduce suicide risk (Brown et al., 2019; Bryan et al., 2016; Cox et al., 2016; Gradus et al., 2013). A trial is currently underway examining the efficacy of written exposure therapy (Sloan et al., 2018), combined with crisis response planning (Bryan et al., 2017), in reducing suicide risk among high-risk, suicidal military service members with PTSD and admitted to inpatient psychiatry (ClinicalTrials.gov: NCT04225130).

Other study findings are noteworthy. Post-discharge suicide rates are higher for individuals with a history of psychiatric admissions than individuals admitted for the first time (Chung et al., 2017). We found that the number of previous same-facility inpatient psychiatric hospitalizations was positively associated with current suicidal ideation severity only for individuals with a provisional PTSD diagnosis. This suggests that PTSD may be one mechanism linking re-hospitalization and suicide risk, corroborating the assertion that treating PTSD among psychiatric inpatients holds promise for suicide prevention.

Regarding the clinical severity of our sample, 41.5% of patients had a probable PTSD diagnosis per the PCL-5 cutoff score. Strikingly, however, only 3.2% of patients were diagnosed with PTSD by their psychiatrist. One possibility for this discrepancy is that the PCL-5 overestimated probable PTSD in this sample. Notably, psychiatrists’ diagnoses for this study were not based on structured or semi-structured clinical interviews; diagnostic accuracy is improved when structured or semi-structured assessment instruments are utilized (Miller et al., 2001). All patients in our sample had a history of trauma exposure, and although the vast majority of individuals exposed to traumas do not develop PTSD (Yehuda et al., 2015), given the clinical severity of individuals who present to inpatient psychiatry, we would expect a prevalence of PTSD greater than 3.2% among trauma-exposed adults. Based on the PCL-5 responses, then, we suspect that PTSD was under-diagnosed in our sample. An under-diagnosis of PTSD among psychiatric inpatients may equate to missed opportunities for the prevention of suicide. Considering past work demonstrating that the treatment of PTSD reduces suicide risk, if PTSD is missed in psychiatric inpatients, then proportionate clinical actions (e.g., evidence-based PTSD treatment) either during the hospitalization or post-discharge would not be provisioned, thereby maintaining a patients’ risk for suicide.

Finally, contrary to hypotheses, no PTSD symptom cluster predicted unique variance in reasons for admittance to inpatient psychiatry. Prior work linking PTSD-related hyperarousal symptoms to suicidal ideation (Panagioti et al., 2017) and attempts (Stanley et al., 2019) has been longitudinal. By contrast, cross-sectional findings—like those in this study—have yielded inconsistent results regarding the intersection of PTSD facets and suicide-related outcomes (Boffa et al., 2017). Longitudinal research, ideally with multiple timepoints, is crucial to establish temporality. It is possible that exacerbations in PTSD-related hyperarousal symptoms (e.g., increasing anger) prompt a suicide-related psychiatric admission (cf. Joiner & Stanley, 2016), regardless of a patient’s PTSD-related avoidance and anhedonia, for instance.

Limitations

Several study limitations are notable. First, data were cross-sectional; thus, it was thus indeterminable if PTSD and/or PTSD-related exacerbations prospectively prompted a suicide-related psychiatric hospitalization. Second, data were collected via self-report. Although the PCL-5 clinical cutoff score is well-validated (Bovin et al., 2016), a structured clinical interview is the gold standard for determining PTSD status (Wisco et al., 2012). Third, the psychiatric disorders obtained from the EMR were not derived using validated structured or semi-structured clinical interviews; this likely resulted in under- or misdiagnosis of PTSD and other disorders in some cases (Miller et al., 2001). Fourth, we did not have access to the nuanced circumstances surrounding individuals’ suicide-related hospitalization, such as whether admission was voluntary versus involuntary and whether they experiencing suicidal ideation versus attempt; these reasons may differentially correlate with PTSD symptoms. Finally, data on prior psychiatric hospitalizations were limited to those that occurred at the same hospital; some patients may have received past inpatient psychiatric care at another hospital.

Conclusions

Trauma-exposed adults admitted to inpatient psychiatry for suicide-related concerns, compared with patients admitted for other reasons, had significantly greater PTSD symptom levels and were also approximately four times more likely to have a provisional PTSD diagnosis. The association between PTSD symptom severity and suicide-related hospitalizations was even greater in the context of elevated impulsivity. PTSD symptoms were also significantly associated with current suicidal ideation severity among the subset of individuals admitted for suicide-related concerns. Interestingly, a greater number of prior psychiatric hospitalizations was significantly associated with more severe suicidal ideation, but only for those with a provisional PTSD diagnosis. Together, study findings suggest that the treatment of PTSD, and attending to cooccurring impulsivity, might be one avenue by which to prevent suicide among high-risk inpatients.

Acknowledgments

This study was supported in part by grants from the National Institute for Mental Health (T32MH019836-17) and the Hogg Foundation for Mental Health (JRG-263).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflict of Interest

We have no known conflict of interests to disclose.

1

Due to the large discrepancy between the proportion of patients who met for a provisional PTSD diagnosis via the validated PCL-5 clinical cutoff score and the psychiatrists’ clinical interviews (a point to which we return in the discussion), we conducted additional analyses to derive an even more conservative estimate of the prevalence of a probable PTSD diagnosis based on the PCL-5. To do this, we estimated the proportion who exceeded both (1) the clinical cutoff and (2) an algorithm-based approach concordant with DSM-5 diagnostic criteria (i.e., one intrusion item, one avoidance item, two negative alterations in cognitions and mood items, and two hyperarousal items rated as “2=Moderately” or above; Weathers, Litz, et al., 2013). Overall, 36.7% (n=69) of patients met for a provisional PTSD diagnosis via these more stringent criteria (i.e., meeting for a probable diagnosis via both methods).

References

  1. American Psychiatric Association. (2013). Diagnostic and Statistical Manual of Mental Disorders, 5th Edition: DSM-5. American Psychiatric Association. [Google Scholar]
  2. Batterham PJ, Ftanou M, Pirkis J, Brewer JL, Mackinnon AJ, Beautrais A, Fairweather-Schmidt AK, & Christensen H (2015). A systematic review and evaluation of measures for suicidal ideation and behaviors in population-based research. Psychological Assessment, 27(2), 501–512. 10.1037/pas0000053 [DOI] [PubMed] [Google Scholar]
  3. Beck AT, & Steer R (1991). Manual for the Beck Scale for Suicide Ideation. Psychological Corporation. [Google Scholar]
  4. Boffa JW, Stanley IH, Hom MA, Norr AM, Joiner TE, & Schmidt NB (2017). PTSD symptoms and suicidal thoughts and behaviors among firefighters. Journal of Psychiatric Research, 84, 277–283. 10.1016/j.jpsychires.2016.10.014 [DOI] [PubMed] [Google Scholar]
  5. 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), 1379–1391. 10.1037/pas0000254 [DOI] [PubMed] [Google Scholar]
  6. Brown LA, McLean CP, Zang Y, Zandberg L, Mintz J, Yarvis JS, Litz BT, Peterson AL, Bryan CJ, Fina B, Petersen J, Dondanville KA, Roache JD, Young-McCaughan S, & Foa EB (2019). Does prolonged exposure increase suicide risk? Results from an active duty military sample. Behaviour Research and Therapy, 118, 87–93. 10.1016/j.brat.2019.04.003 [DOI] [PubMed] [Google Scholar]
  7. Bryan CJ, Clemans TA, Hernandez AM, Mintz J, Peterson AL, Yarvis JS, & Resick PA (2016). Evaluating potential iatrogenic suicide risk in trauma-focused group cognitive behavioral therapy for the treatment of PTSD in active duty military personnel. Depression and Anxiety, 33(6), 549–557. 10.1002/da.22456 [DOI] [PubMed] [Google Scholar]
  8. Bryan CJ, Mintz J, Clemans TA, Leeson B, Burch TS, Williams SR, Maney E, & Rudd MD (2017). Effect of crisis response planning vs. contracts for safety on suicide risk in U.S. Army Soldiers: A randomized clinical trial. Journal of Affective Disorders, 212, 64–72. 10.1016/j.jad.2017.01.028 [DOI] [PubMed] [Google Scholar]
  9. Centers for Disease Control and Prevention [CDC]. (2020). WISQARS: Web-Based Injury Statistics Query and Reporting System. http://www.cdc.gov/injury/wisqars/index.html
  10. Chung D, Hadzi-Pavlovic D, Wang M, Swaraj S, Olfson M, & Large M (2019). Meta-analysis of suicide rates in the first week and the first month after psychiatric hospitalisation. BMJ Open, 9(3), e023883 10.1136/bmjopen-2018-023883 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Chung D, Ryan CJ, Hadzi-Pavlovic D, Singh SP, Stanton C, & Large MM (2017). Suicide rates after discharge from psychiatric facilities. JAMA Psychiatry. 10.1001/jamapsychiatry.2017.1044 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cooper SA, Szymanski BR, Bohnert KM, Sripada RK, & McCarthy JF (2020). Association between positive results on the Primary Care–Posttraumatic Stress Disorder Screen and suicide mortality among US Veterans. JAMA Network Open, 3(9), e2015707 10.1001/jamanetworkopen.2020.15707 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Cox KS, Mouilso ER, Venners MR, Defever ME, Duvivier L, Rauch SAM, Strom TQ, Joiner TE, & Tuerk PW (2016). Reducing suicidal ideation through evidence-based treatment for posttraumatic stress disorder. Journal of Psychiatric Research, 80, 59–63. 10.1016/j.jpsychires.2016.05.011 [DOI] [PubMed] [Google Scholar]
  14. Folstein MF, Folstein SE, & McHugh PR (1975). “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12(3), 189–198. [DOI] [PubMed] [Google Scholar]
  15. Gradus JL (2017). PTSD and Death from Suicide. PTSD Research Quarterly, 28(4), 1–8. [Google Scholar]
  16. Gradus JL, Qin P, Lincoln AK, Miller M, Lawler E, Sorensen HT, & Lash TL (2010). Posttraumatic stress disorder and completed suicide. American Journal of Epidemiology, 171(6), 721–727. 10.1093/aje/kwp456 [DOI] [PubMed] [Google Scholar]
  17. Gradus JL, Suvak MK, Wisco BE, Marx BP, & Resick PA (2013). Treatment of posttraumatic stress disorder reduces suicidal ideation. Depression and Anxiety, 30(10), 1046–1053. 10.1002/da.22117 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Gray MJ, Litz BT, Hsu JL, & Lombardo TW (2004). Psychometric properties of the Life Events Checklist. Assessment, 11(4), 330–341. 10.1177/1073191104269954 [DOI] [PubMed] [Google Scholar]
  19. Hayes AF (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. Guilford Press. [Google Scholar]
  20. Joiner TE, & Stanley IH (2016). Can the phenomenology of a suicidal crisis be usefully understood as a suite of antipredator defensive reactions? Psychiatry: Interpersonal and Biological Processes, 79(2), 107–119. 10.1080/00332747.2016.1142800 [DOI] [PubMed] [Google Scholar]
  21. Keane TM, Rubin A, Lachowicz M, Brief D, Enggasser JL, Roy M, Hermos J, Helmuth E, & Rosenbloom D (2014). Temporal stability of DSM–5 posttraumatic stress disorder criteria in a problem-drinking sample. Psychological Assessment, 26(4), 1138–1145. 10.1037/a0037133 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Koldsland B, Mehlum L, Mellesdal L, Walby FA, & Diep LM (2012). The suicide assessment scale: Psychometric properties of a Norwegian language version. BMC Research Notes, 5(1), 417 10.1186/1756-0500-5-417 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kotler M, Iancu I, Efroni R, & Amir M (2001). Anger, impulsivity, social support, and suicide risk in patients with posttraumatic stress disorder. The Journal of Nervous and Mental Disease, 189(3), 162–167. 10.1097/00005053-200103000-00004 [DOI] [PubMed] [Google Scholar]
  24. Miller PR, Dasher R, Collins R, Griffiths P, & Brown F (2001). Inpatient diagnostic assessments: 1. Accuracy of structured vs. unstructured interviews. Psychiatry Research, 105(3), 255–264. 10.1016/S0165-1781(01)00317-1 [DOI] [PubMed] [Google Scholar]
  25. Moeller FG, Barratt ES, Dougherty DM, Schmitz JM, & Swann AC (2001). Psychiatric aspects of impulsivity. American Journal of Psychiatry, 158(11), 1783–1793. 10.1176/appi.ajp.158.11.1783 [DOI] [PubMed] [Google Scholar]
  26. Nock MK, Hwang I, Sampson N, Kessler RC, Angermeyer M, Beautrais A, Borges G, Bromet E, Bruffaerts R, De Girolamo G, De Graaf R, Florescu S, Gureje O, Haro JM, Hu C, Huang Y, Karam EG, Kawakami N, Kovess V, … Williams DR (2009). Cross-national analysis of the associations among mental disorders and suicidal behavior: Findings from the WHO World Mental Health Surveys. PLoS Medicine, 6(8). 10.1371/journal.pmed.1000123 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Pachkowski MC, May AM, Tsai M, & Klonsky ED (2019). A brief measure of unbearable psychache. Suicide and Life-Threatening Behavior, 49(6), 1721–1734. 10.1111/sltb.12556 [DOI] [PubMed] [Google Scholar]
  28. Panagioti M, Angelakis I, Tarrier N, & Gooding P (2017). A prospective investigation of the impact of distinct posttraumatic (PTSD) symptom clusters on suicidal ideation. Cognitive Therapy and Research, 41(4), 645–653. 10.1007/s10608-016-9829-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Panagioti M, Gooding PA, & Tarrier N (2012). A meta-analysis of the association between posttraumatic stress disorder and suicidality: The role of comorbid depression. Comprehensive Psychiatry, 53(7), 915–930. 10.1016/j.comppsych.2012.02.009 [DOI] [PubMed] [Google Scholar]
  30. Panagioti M, Gooding P, & Tarrier N (2009). Post-traumatic stress disorder and suicidal behavior: A narrative review. Clinical Psychology Review, 29(6), 471–482. 10.1016/j.cpr.2009.05.001 [DOI] [PubMed] [Google Scholar]
  31. Patton JH, Stanford MS, & Barratt ES (1995). Factor structure of the Barratt Impulsiveness Scale. Journal of Clinical Psychology, 51(6), 768–774. [DOI] [PubMed] [Google Scholar]
  32. Roley ME, Contractor AA, Weiss NH, Armour C, & Elhai JD (2017). Impulsivity facets’ predictive relations with DSM–5 PTSD symptom clusters. Psychological Trauma: Theory, Research, Practice, and Policy, 9(1), 76–79. 10.1037/tra0000146 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Shahnaz A, Saffer BY, & Klonsky ED (2018). The relationship of perfectionism to suicide ideation and attempts in a large online sample. Personality and Individual Differences, 130, 117–121. 10.1016/j.paid.2018.04.002 [DOI] [Google Scholar]
  34. Sloan DM, Marx BP, Lee DJ, & Resick PA (2018). A brief exposure-based treatment vs cognitive processing therapy for posttraumatic stress disorder: A randomized noninferiority clinical trial. JAMA Psychiatry, 75(3), 233 10.1001/jamapsychiatry.2017.4249 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Stanford MS, Mathias CW, Dougherty DM, Lake SL, Anderson NE, & Patton JH (2009). Fifty years of the Barratt Impulsiveness Scale: An update and review. Personality and Individual Differences, 47(5), 385–395. 10.1016/j.paid.2009.04.008 [DOI] [Google Scholar]
  36. Stanley IH, Rogers ML, Hanson JE, Gutierrez PM, & Joiner TE (2019). PTSD symptom clusters and suicide attempts among high-risk military service members: A three-month prospective investigation. Journal of Consulting and Clinical Psychology, 87(1), 67–78. 10.1037/ccp0000350 [DOI] [PubMed] [Google Scholar]
  37. Stanley IH, Yancey JR, Patrick CJ, & Joiner TE (2018). A distinct configuration of MMPI-2-RF scales RCd and RC9/ACT is associated with suicide attempt risk among suicide ideators in a psychiatric outpatient sample. Psychological Assessment, 30(9), 1249–1254. 10.1037/pas0000588 [DOI] [PubMed] [Google Scholar]
  38. Vujanovic AA, Dutcher CD, & Berenz EC (2017). Multimodal examination of distress tolerance and posttraumatic stress disorder symptoms in acute-care psychiatric inpatients. Journal of Anxiety Disorders, 48, 45–53. 10.1016/j.janxdis.2016.08.005 [DOI] [PubMed] [Google Scholar]
  39. Weathers FW, Blake DD, Schnurr PP, Kaloupek DG, Marx BP, & Keane TM (2013). The Life Events Checklist for DSM-5 (LEC-5). National Center for PTSD; www.ptsd.va.gov [Google Scholar]
  40. Weathers FW, Litz BT, Keane TM, Palmieri PA, Marx BP, & Schnurr PP (2013). The PTSD Checklist for DSM-5 (PCL-5). National Center for PTSD; http://www.ptsd.va.gov [Google Scholar]
  41. Weiss NH, Tull MT, Sullivan TP, Dixon-Gordon KL, & Gratz KL (2015). Posttraumatic stress disorder symptoms and risky behaviors among trauma-exposed inpatients with substance dependence: The influence of negative and positive urgency. Drug and Alcohol Dependence, 155, 147–153. 10.1016/j.drugalcdep.2015.07.679 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Wisco BE, Marx BP, & Keane TM (2012). Screening, diagnosis, and treatment of posttraumatic stress disorder. Military Medicine, 177(8S), 7–13. 10.7205/MILMED-D-12-00111 [DOI] [PubMed] [Google Scholar]
  43. Wolf EJ, Miller MW, Harrington KM, & Reardon A (2012). Personality-based latent classes of posttraumatic psychopathology: Personality disorders and the internalizing/externalizing model. Journal of Abnormal Psychology, 121(1), 256–262. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Yehuda R, Hoge CW, McFarlane AC, Vermetten E, Lanius RA, Nievergelt CM, Hobfoll SE, Koenen KC, Neylan TC, & Hyman SE (2015). Post-traumatic stress disorder. Nature Reviews Disease Primers, 1, 1–22. 10.1038/nrdp.2015.57 [DOI] [PubMed] [Google Scholar]

RESOURCES