Skip to main content
NCBI home page
VA Author Manuscripts logoLink to VA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jun 24.
Published in final edited form as: J Psychiatr Res. 2022 Jul 31;154:159–166. doi: 10.1016/j.jpsychires.2022.07.047

Encouraging the use of the Veterans Crisis Line among high-risk Veterans: A randomized trial of a Crisis Line Facilitation intervention

Mark A Ilgen 1,2, Amanda M Price 1,2, Lara N Coughlin 1,2, Paul N Pfeiffer 1,2, Haylie J Stewart 1,2, Elizabeth Pope 3, Peter C Britton 4,5
PMCID: PMC13288178  NIHMSID: NIHMS2183143  PMID: 35940001

Abstract

The Veterans Crisis Line (VCL) is a core component of VA’s suicide prevention strategy. Despite the availability and utility of the VCL, many Veterans do not utilize this resource during times of crisis. A brief, psychoeducational behavioral intervention (termed Crisis Line Facilitation [CLF]) was developed to increase utilization of the VCL and reduce suicidal behaviors in high-risk Veterans. The therapist-led session includes educational information regarding the VCL, as well as a chance to discuss the participant’s perceptions of contacting the VCL during periods of crisis. The final component of the session is a practice call placed to the VCL by both the therapist and the participant. The CLF intervention was compared to Enhanced Usual Care (EUC) during a multi-site randomized clinical trial for 307 Veteran participants recently hospitalized for a suicidal crisis who reported no contact with the VCL in the prior 12 months. Initial analyses indicated that participants randomized to the CLF intervention were less likely to report suicidal behaviors, including suicide attempts compared to participants randomized to receive EUC over 12-months of follow-up (χ2=18.48 / p<.0001), however this effect was not sustained when analyses were conducted on an individual level. No significant differences were found between conditions on VCL utilization. Initial evidence suggests a brief CLF intervention has an impact on preventing suicidal behaviors in Veterans treated in inpatient mental health programs; however, it may not change use of the VCL. This brief intervention could be easily adapted into clinical settings to be delivered by standard clinical staff.

Keywords: Suicide, Veterans Crisis Line, Brief Interventions, Behavioral rehearsal, Inpatient treatment

INTRODUCTION

Suicide prevention is a top national priority (Department of Veterans Affairs, 2020; Office of Mental Health and Suicide Prevention, 2018; Office of the Surgeon General, 2021; Rubin, 2019).Veterans are a particularly high-risk group, accounting for 13.7% of all suicides in the United States population in 2019 (Office of Mental Health and Suicide Prevention, 2021). In that year, the age and gender adjusted suicide rate was 52.3% higher in Veterans than non-Veterans, although data from 2018 to 2019 show a7.2% decrease in age- and sex-adjusted suicide mortality rates for Veterans, which was greater than the 1.8% decrease seen among non-Veteran U.S. adults. The Veterans Health Administration’s (VHA) has invested heavily in efforts to identify and support Veterans at high-risk for suicide, including the creation of the Veterans Crisis Line (VCL). The VCL represents a key resource in suicide prevention as the line is a free, confidential 24-hour point of contact available to Veterans in crisis, regardless of their involvement with VHA services (Acosta et al., 2020; Britton et al., 2016; Gould et al., 2007; Karras et al., 2016). Veterans, Service Members, and their families experiencing a crisis can contact the VCL via phone call, text, or chat. During these contacts, specially trained responders will assist in reducing immediate distress, assessing current severity of suicide risk, and determining the best course of action to support the individual in crisis. From 2007 through the end of 2021, the VCL has answered more than 5.9 million calls, initiated emergency dispatch to callers more than 220,000 times, and referred more than 1.1 million Veterans to their local Suicide Prevention Coordinators (SPC) as a resource following a point of contact to the VCL (Veterans Crisis Line, 2021).

The 24-hour availability of crisis lines, including the VCL, is well-suited to the transient nature of suicidal thoughts and plans, allowing Veterans to utilize the resource when needed (Joiner et al., 1999). Although the general risk for suicide likely reflects factors that accumulate over many years, periods of acute suicidal crisis are often temporary. Preventing the transition from suicidal ideation to planning to attempt is exceedingly difficult because, even for individuals who are receiving care from a mental health provider, acute suicidal crises can occur rapidly, before others can intervene. Suicide prevention efforts, then, require crisis services to be easily accessible at any time during high-risk periods and, after the acute crisis, increasing utilization of more comprehensive services to address some of the longer-term symptoms or problems contributing to suicidal thoughts/behaviors (Ammerman et al., 2021; Knox et al., 2012).

Research examining the usages and effectiveness of crisis line services remains limited. A recent systematic review found that current evidence supports the efficacy of crisis lines for short term outcomes; however, evidence supporting long term efficacy, specifically regarding reducing suicidal behaviors among high-risk callers, is lacking (Gould et al., 2016; Hoffberg et al., 2019). Observational data suggest that calls to crisis lines are associated with initial reductions in suicidal ideation including among Veterans (Britton et al., 2017; Britton, Crasta, et al., 2021; Britton, Ilgen, et al., 2021; Britton et al., in press), and these reductions are maintained or continue to decrease, with more than half of suicidal callers reporting no suicidal thoughts or behaviors after 1-2 weeks (Gould et al., 2007; Gould et al., 2016; Johnson et al., 2021; Mishara et al., 2007).

Despite the high potential value of crisis lines, some Veterans, Service Members, or families who could benefit from the VCL may not use this service during a suicidal crisis. This study developed a brief, single session intervention, called Crisis Line Facilitation (CLF), designed to encourage those at high risk to contact the VCL during a future crisis. Briefly, the CLF intervention involved a review of prior suicidal crises, a discussion of factors that could facilitate future help-seeking, and presentation of psychoeducational information about the VCL. The final portion of the session involved the participant and the therapist calling the VCL together to rehearse the process of help-seeking and counter potential negative expectancies about calling. For more information about CLF, see Ilgen et al. (2020).

This present study was a multi-site randomized controlled trial testing the efficacy of the CLF behavioral intervention compared to enhanced usual care (EUC) on utilization of the VCL and suicidal behaviors. Veterans hospitalized following a suicidal crisis at two VHA inpatient facilities who had not used the VCL within the prior year were recruited and randomized to receive a single session CLF intervention or EUC. Participants completed study follow-ups at 3-, 6-, and 12-months post enrollment.

MATERIALS AND METHODS

Recruitment occurred within inpatient mental health units of two Veterans Health Administration (VHA) facilities between March 31, 2016, and August 31, 2018. All protocols and procedures were approved by the local VHA sites Institutional Review Boards (IRBs) and the study was registered on ClinicalTrials.gov (#NCT02459587). This study was conducted in accordance with the latest version of the Declaration of Helsinki and the Consolidated Standards of Reporting Trials (CONSORT) reporting guidelines. The VA Health Services Research & Development Data Safety Monitoring Board reviewed participant safety and enrollment throughout the duration of the study.

Medical records were screened to identify patients reporting significant suicidal thoughts, plans, or attempts associated with the current inpatient admission. Additional inclusion criteria were (1) report of suicidal ideation within the week prior to hospitalization ≥5 on the Beck Scale for Suicidal Ideation, and (2) no reported use of the VCL within the past 12 months. Exclusion criteria were (1) age younger than 18 years, (2) inability to provide informed consent, (3) inability to understand English, (4) significant psychotic symptoms or cognitive impairment, (5) medical instability, (6) receipt of electroconvulsive therapy (ECT), and (7) prisoners.

Eligible and enrolled participants completed a baseline assessment and were randomized to either CLF or EUC. Randomization sequences were generated independently for each study site by the study statistician prior to recruitment. Randomization was conducted using a two-group design with equal allocation per group and was blocked by sex and lifetime history of suicide attempt (no prior attempt vs one or more).

Participants received their assigned condition session prior to leaving the hospital. Participants in both conditions were asked to complete a brief post-session survey to assess the relevance and helpfulness of the therapy intervention (CLF) or informational materials (EUC). Participants were re-assessed at 3-, 6-, and 12-months post enrollment. All assessments included a brief self-report survey and a researcher-administered interview. Participants received compensation up to $125 for study participation.

Intervention and control conditions

Both CLF and EUC were delivered in a single session during an inpatient hospital stay. All participants, regardless of condition assignment, received the same standard of care for hospitalized inpatients, often including the dissemination of VCL information. Relatedly, standard inpatient care within VA also includes the mandated completion of Safety Plans, which provide opportunity to list numbers for outreach during times of crisis that may include the VCL. Because of the high-risk nature of the study sample, all participants were independently assessed for suicide risk by research staff and were provided with information regarding mental health support services, including the number for the VCL. The resources provided by research staff were similar in scope and content to those provided by clinical staff to patients treated for suicidal ideation and were deemed necessary to provide for participant safety during the trial.

The CLF intervention delivered during this study has been described previously (Ilgen et al., 2020). Briefly, CLF was based on the Theory of Planned Behavior, where the likelihood of engaging in the targeted behavior (calling the VCL) was guided by three core concepts–behavioral attitudes, subjective norms, and perceived behavioral control (Ajzen, 2002; Ajzen, 2011). The session began with a discussion focused on past suicidal crises and perceptions of the VCL, targeting behavioral attitudes by highlighting ways in which calling the VCL could lead to positive outcomes in future crises. The participant was then provided educational materials emphasizing the broad use of the VCL by Veterans, methods of contact available, and types of support offered. Perceived barriers or facilitators to using the VCL were discussed, and a list of questions were generated to address concerns the participant may have regarding contacting the VCL. Since behavioral rehearsal is a strong predictor of increased self-efficacy towards behavior change (Ajzen, 1991; Bandura, 1977, 1982, 2000), the session ended with a practice call to the VCL. CLF sessions lasted approximately 30-45 minutes and were conducted by master’s level therapists trained to deliver intervention specific content. With participant permission, sessions were audio recorded for adherence monitoring to delivering session content. Study therapists received ongoing training and group clinical supervision during the trial to ensure intervention content was delivered consistently across sites. As noted previously, general perceptions of helpfulness of CLF were high (Ilgen et al., 2020).

Following study completion, a random sample of 10% of CLF sessions from each site were selected for fidelity rating. Independent raters completed ratings forms which included 17 binary content elements (e.g., “Did the therapist ask the veteran questions in order to understand the crisis?”) pertaining to specific session content, and 4 binary global ratings based on general motivational interviewing skills (e.g., “Did the therapist appropriately use reflections?”). Global motivational interviewing skills were given a “yes” (adequate) rating based on the expectation that the person delivering the intervention would have good “spirit of MI” skills and some basic training but would not necessarily be an advanced practice therapist. Overall, study therapists delivered the session content adequately, as only six sessions out of 31 coded (19.4%) had any inadequate content element. Of these six sessions, the majority (five sessions; 83.3%) had 1 inadequate content element (out of 17) and only one session had two inadequate content elements. All sessions had 100% fidelity to global motivational interviewing skills ratings.

Participants assigned to the EUC condition received standard inpatient care at the recruitment site as usual. In addition, participants received promotional items containing the VCL number and educational information about the VCL from research staff following the baseline assessment. While participants in the EUC condition were encouraged to seek help if suicidal thoughts occurred in the future, no additional discussion related to the VCL occurred. EUC sessions were brief (5-10 minutes) and were not audio recorded.

Primary and Secondary outcomes

The primary outcome for this study was utilization of the VCL. Data on VCL utilization was collected via participant self-report and was operationalized as any contact (e.g., calls, texts, online chats) made to the VCL for any crisis reason (with or without suicidal intent) during the 12-months post enrollment. Because suicide attempts are rare events even for individuals at high-risk, suicidal behaviors were examined as a secondary outcome. Suicidal behaviors included the following: (1) Actual attempts, as defined as deliberate self-injurious behaviors; (2) Interrupted attempts, defined as events in which an outside force stopped the individual before the self-injurious behavior began; and (3) Aborted attempts, defined as events in which the individual stopped themselves before the behavior could begin; all requiring at least some intent to die.

Information regarding specific behaviors was collected using the Columbia Suicide Severity Rating Scale (C-SSRS) (Posner et al., 2011). At baseline the lifetime version of the C-SSRS was administered, with the Since Last Visit version administered at each follow-up. This version allowed for the capture of suicide behavioral data during the entire study period, regardless of whether the participant completed all follow-up assessments.

Data for both primary and secondary outcomes were collected using the Suicidal Crisis Timeline Interview, a measure developed by the study team based on the Timeline Follow-Back (TLFB) method (Linehan et al., 2006; Sobell & Sobell, 1992) . The TLFB has shown strong test-retest reliability when used to collect substance use and other behavioral data over specific time intervals (Carey et al., 2001; Fals-Stewart et al., 2000; Sobell et al., 1979). Following the TLFB method, the Suicidal Crisis Timeline Interview was designed as a semi-structured interview where participants are asked to recall specific information regarding suicidal crises and help seeking behaviors, including contacting the VCL. Using a calendar as a guide, participants were asked to recall events over the previous six months at baseline, and since their last study visit at all subsequent follow-ups. All data related to suicidal behaviors collected on the C-SSRS was compared to data collected during the Suicidal Crisis Timeline Interview to ensure all reported events were captured. Because of the complex nature of suicidal behaviors, all reported suicidal behaviors were reviewed by study team members who were blinded to condition to ensure consistent coding of suicidal behaviors across the trial, using a set of uniform definitions and guidelines created by the study team.

Statistical Analyses

Participants who provided outcome data for at least one follow-up assessment timepoint (n=272) were included for analysis. Longitudinal data models used an intent-to-treat framework and data from all follow-up times points to analyze primary and secondary outcomes. Prior to longitudinal modeling, all outcomes and baseline demographic characteristics were evaluated for distribution and summarized overall and by condition for all enrolled participants. We utilized unadjusted multiple-event survival models to compare both patterns of VCL contacts (primary outcome) and suicidal behaviors (secondary outcome) by condition. The distributions of events over time were examined via Kaplan-Meier analysis to determine the appropriate modeling approach. Continuous-time models were selected to allow for the evaluation of time to event over the entire span of the follow-up period (i.e., 12-months). Appropriateness of the proportional hazards assumption was tested and verified; model fit was assessed via Akaike’s Information Criterion (AIC). For both outcomes, survival analyses used start/stop counting methods to identify time until event and were structured to allow for multiple events per person. A total of 7 participants died during the follow-up interval. No verifiable information was available on cause of death across all participants. In primary analyses, participants were censored at the time of death. All data management and statistical analysis conducted for this paper was generated using SAS 9.4 software. (Institute Inc., Cary, NC, USA).

RESULTS

Across the recruitment sites, a total of 1,718 Veteran patients were admitted during the recruitment window; 508 participants provided written informed consent for the in-depth screening and were assessed for study eligibility. Of those assessed, 315 (62.4%) were eligible for the trial, with 146 (46.3%) eligible from one site, and 169 (53.7%) eligible from the second site. Main reasons for ineligibility included: no report of a suicidal crisis prompting the hospital admission (48.7%) and self-report of a contact to the VCL within the past 12-months (48.7%). A total of 307 (97.5%) participants provided additional informed consent and were enrolled and randomized into the trial. Of those, 157 participants (51.1%) were randomized to receive the CLF intervention, and 150 participants (48.9%) were randomized to EUC. Further recruitment and enrollment information, including enrollment and retention per study site can be found in Figure 1. All participants completed their assigned condition session. Overall, 272 of 307 (88.6%) participants provided data for at least 1 of the follow-up assessments, with 244 participants (83.3%) completing the 3-month follow-up, 244 (85.0%) completing the 6-month follow-up, and 238 (85.0%) completing the 12-month follow-up. There were no significant differences between recruitment site or treatment condition on participant retention. Specific baseline demographic and clinical characteristics for the sample and by treatment condition are provided in Table 1. There were no significant by-group differences in any baseline characteristics. Some significant differences in demographic and clinical characteristics were noted between study sites and are presented in Table 2, however these differences did not impact condition assignment or outcomes (see Table 3).

Figure 1:

Figure 1:

Open in a new tab

CONSORT Diagram

Table 1.

Demographic Characteristics and Baseline Measures of Suicidal Ideation, Suicide Behaviors, and VCL Utilization by Condition

Condition
Total (n=307, 100%) CLF (n=157, 51%) Control (n=150, 49%)
Recruitment Site
A 140 (46%) 71 (45%) 69 (46%)
B 167 (54%) 86 (55%) 81 (54%)
Demographics
Age (yr.) (mean, SD) NS 47.0 (13.1) 47.2 (12.8) 46.8 (13.5)
Male Gender NS 268 (87%) 138 (88%) 130 (87%)
Hispanic ethnicity NS 17 (6%) 11 (7%) 6 (4%)
Non-white race, or more than one race NS 81 (26%) 44 (29%) 36 (24%)
Employed (full- or part-time) NS 66 (21%) 35 (22%) 31 (21%)
Domestic status: Married/cohabitating NS 94 (31%) 42 (27%) 52 (35%)
# Combat tours NS
None 156 (51%) 83 (53%) 73 (49%)
One 93 (30%) 44 (28%) 49 (33%)
Two or more 58 (19%) 30 (19%) 28 (19%)
Clinical Characteristics
Suicidal Ideation, Current (BSS-SR; Mean, SD) NS 19.1 (6.9) 19.4 (7.1) 18.7 (6.6)
Lifetime history of Suicide Attempt(s) NS (CSSRS)
No attempts, ever 87 (28%) 46 (29%) 41 (27%)
One attempt 68 (22%) 31 (20%) 37 (25%)
2+ attempts 152 (49%) 80 (51%) 72 (48%)
VCL utilization
VCL Utilization, Lifetime NS
Never 208 (68%) 104 (66%) 104 (69%)
Once 47 (15%) 25 (16%) 22 (15%)
Two or more times 52 (17%) 28 (18%) 24 (16%)
Open in a new tab
NS

Not significant at 0.05 level of significance.

Table 2.

Demographic Characteristics and Baseline Measures of Suicidal Ideation, Suicide Behaviors, and VCL Utilization by Study Site.

Site
Total (n=307, 100%) Site A (n=140, 46%) Site B (n=167, 54%)
Condition NS
CLF 157 (51%) 71 (51%) 86 (51%)
Control 150 (49%) 69 (49%) 81 (49%)
Demographics
Age (yr.) (mean, SD) 0.005 47.0 (13.1) 44.7 (13.3) 48.9 (12.7)
Male Gender NS 268 (87%) 119 (85%) 149 (89%)
Hispanic ethnicity NS 17 (5%) 7 (5%) 10 (6%)
Non-white race, or more than one race NS 81 (26%) 31 (22%) 50 (30%)
Employed (full- or part-time) 0.006 68 (22%) 41 (29%) 27 (16%)
Domestic status: Married/cohabitating 0.003 94 (31%) 55 (39%) 39 (23%)
# Combat tours 0.02
None 156 (51%) 60 (43%) 96 (57%)
One 93 (30%) 46 (33%) 47 (28%)
Two or more 58 (19%) 34 (24%) 24 (14%)
Clinical Characteristics (BSS-RS)
Level of Suicidal Ideation (mean, SD) NS (BSS-SR; maximum score= 38) 19.1 (6.9) 19.9 (6.7) 18.4 (7.0)
Lifetime history of Suicide Attempt(s) 0.038
No attempts, ever 91 (30%) 33 (24%) 58 (35%)
One attempt 94 (31%) 41 (29%) 53 (32%)
2+ attempts 121 (39%) 65 (47%) 56 (33%)
VCL utilization
VCL Utilization, Lifetime NS
Never 208 (68%) 94 (67%) 114 (68%)
Once 47 (15%) 23 (16%) 24 (14%)
Twice or more times 52 (17%) 23 (16%) 29 (17%)
Open in a new tab
NS

Not significant at 0.05 level of significance.

Table 3.

Demographic Characteristics and Baseline Measures of Suicidal Ideation, Suicide Behaviors, and VCL Utilization by Study Site and Condition

Condition
EUC CLF
Site A Site B Site A Site B P-value Site x trt
Demographics
Age (yr.) (mean, SD) 43.9 (13.6) 49.2 (13.0) 45.4 (12.9) 48.7 (12.6) 0.52
 
Employed (full- or part-time) 20 (29%) 12 (15%) 21 (30%) 15 (17%) 0.77
 
Domestic status: Married/cohabitating 32 (46%) 20 (25%) 23 (32%) 19 (22%) 0.38
# Combat tours 0.17
None 32 (46%) 41 (51%) 28 (39%) 55 (64%)
One 21 (30%) 28 (35%) 25 (35%) 19 (22%)
Two or more 16 (23%) 12 (15%) 18 (25%) 12 (14%)
Clinical Characteristics (BSS-RS)
Lifetime history of Suicide Attempt(s) 0.55
No attempts, ever 13 (19%) 28 (34%) 20 (29%) 30 (35%)
One attempt 23 (33%) 26 (32%) 18 (26%) 27 (31%)
2+ attempts 33 (48%) 27 (33%) 32 (46%) 29 (34%)
Open in a new tab

Veterans Crisis Line Utilization

During the follow-up period, 52 participants (19.1%) contacted the VCL. Of those who contacted the VCL, the majority (n=34; 65.4%) reported only one contact during the 12-month follow-up interval. The average number of contacts was 1.6 (SD 1.1). Overall, the VCL use incidence rate of the sample was 0.064 contacts per 100 person-days during the follow-up interval; 0.064 in CFL and 0.063 in EUC. When allowing for multiple VCL contacts per person, this amounts to 0.088 per 100 person-days, overall; 0.095 per 100 person-days in CFL and 0.079 per 100 person-days in EUC. All reported VCL contacts (i.e., calls, chats, or texts) from the researcher administered Suicidal Crisis Timeline Interview were examined and results are presented in Figure 2. There was no significant difference observed between conditions on multiple VCL contacts over the one-year follow-up period: hazard ratio (CLF) = 1.24 (95%CI: 0.80 – 1.93, p=0.34. AIC=918.21) (Table 4, Model 1).

Figure 2. Survival Curve for any VCL contact:

Figure 2.

Open in a new tab

for Model 1 (unadjusted).

Table 4.

Survival analyses of VCL utilization and of any suicide behavior event, allowing for multiple events per person

Model 1: VCL contacts, adjusted for site † β estimate (se)* χ2/ p-value** Hazard Ratio HR*** 95% CI
Condition = CLF 0.213 (0.222) 0.92 / p=0.34 1.24 0.80 – 1.93
Condition = EUC Referent
Model 2: Any suicide behavior event, adjusted for site † β estimate (se) χ2/ p-value Hazard Ratio HR 95% CI
Condition = CLF −0.680 (0.153) 19.63 / p<.0001 0.51 0.37-0.68
Condition = EUC (referent)
Open in a new tab
*

se=standard error;

**

Likelihood ratio chi-square test statistic with p-value;

***

HR=Hazard Ratio

Suicidal Behaviors

A total of 47 participants (17.3%) had at least one suicide behavior during the follow-up. A total of 183 behaviors were reported at the following frequency: 84 (45.9%) were actual attempts, 89 (48.6%) were aborted attempts, and 10 (5.5%) were interrupted attempts. Among those participants persons with suicide behaviors, the average number of suicide behaviors was 3.9 (SD 7.1). The incidence rate of at least one suicidal behavior was 0.056 per 100 person-days; 0.049 in CFL and 0.066 in EUC. When allowing for multiple suicidal behaviors per person, this amounts to rates per 100 person-days of 0.193 for the full sample; 0.133 in CFL; and 0.260 in EUC. Figure 3 displays the comparison of CLF and EUC conditions on multiple-event suicidal behaviors. Participants in the CLF condition had significantly lower hazard of reporting a suicidal behavior event during the one-year follow-up time period compared to those in the EUC condition, based on multiple events per person: hazard ratio (CLF) = 0.51 (95%CI: 0.37 – 0.68, p<0.0001. AIC=1992.49). (Table 4, Model 2).

Figure 3. Survival Curve for Suicide Behaviors:

Figure 3.

Open in a new tab

for Model 2 (unadjusted).

To understand the effect of the intervention on suicidal behavior, we examined the number of behavior events per individual. Given that n=3 individuals reported 10 or more suicide attempts during the follow-up interval, sensitivity analyses were conducted where the maximum number of per-person suicidal events was limited to 9. With this model, there were no significant differences between groups on suicidal behaviors: hazard ratio (CLF) = 0.75 (95%CI: 0.53 – 1.5, p=0.092. AIC=1461.14), although the direction and general pattern of effects was similar to those described in the primary outcomes.

Additional supplementary analyses were conducted that assumed that all deaths (N=7) were suicides and were added to data on other suicidal behaviors. The initial model was generally consistent with the results reported in Table 4 (Model 2), with HR for CLF = 0.54 (HR 95CI: 0.40-0.72); however, the test for proportional hazards for this model failed (chi-square=4.213, p=0.041). When we ran the same analyses but capped the number of post-baseline suicide attempts at 9, the proportional hazards assumption was not violated (chi-square=0.0101; p=0.92) and there were no significant differences between groups: hazard ratio (CLF) = 0.79 (95% CI: 0.56 – 1.10, p=0.163, AIC=1529.971).

DISCUSSION

Individuals receiving inpatient mental health care are at clear elevated risk for suicidal behaviors following discharge (Brenner & Barnes, 2012; Chung et al., 2019; Forte et al., 2019; Hunt et al., 2009; Luxton et al., 2013; Meehan et al., 2006; Ruengorn et al., 2011; Valenstein et al., 2009), including Veterans (Britton et al., 2017; Britton, Crasta, et al., 2021; Britton, Ilgen, et al., 2021). New and effective strategies are needed to intervene and reduce suicide risk in these individuals. The present study tested the impact of a single-session intervention, referred to as CLF, designed to encourage usage of the VCL among Veterans psychiatrically hospitalized for a recent suicidal crisis. The results indicate that those randomized to CLF were significantly less likely to report suicide behaviors in the year following discharge, however subsequent analyses suggest this effect may be driven primarily by outliers with a large number of attempts. CLF did not significantly impact use of the VCL, the primary hypothesized mechanism of action of CLF.

Suicide prevention is a national priority but the evidence base for suicide prevention is still in the early stages (Office of the Surgeon General, 2021). More intensive interventions offer better evidence for the efficacy of suicide prevention measures (Brown et al., 2016; Brown et al., 2005; Rudd et al., 2015), however, delivery of these interventions requires highly specialized training. Even brief versions of these interventions last between 4-8 sessions. This creates significant barriers for wide-scale implementation of these types of interventions in health systems.

The present CLF intervention is appealing because it is only a single session and intended to require less training for the clinician. Other studies of brief suicide prevention interventions show promise such as a large-scale cohort comparison study of Safety Planning in VHA Emergency Departments (ED), which found evidence for the potential beneficial impact of this single-session study on risk of suicide attempt over 6-months (Stanley et al., 2018). Consistent with the VHA’s widespread adoption of Safety Planning as part of standard clinical care, this intervention was provided to both the CLF and EUC conditions as part of usual care within the study sites for the present trial. Thus, the present results show the impact CLF has above and beyond the potential benefits of Safety Planning. However, the present study did not monitor or attempt to shape the quality or content of the Safety Plan, so it is likely that the quality and potential impact of Safety Plans in this study varied. A separate line of research has shown the benefits of “caring contacts”, such as letters or texts following an inpatient stay or other high-risk contact (Landes et al., 2019; Luxton et al., 2020; Motto & Bostrom, 2001; Reger et al., 2019). It is possible that this type of continued contact could have enhanced the benefits of the CLF intervention and/or reinforced the learning that occurred during the CLF sessions. Alternatively, it is possible that the CLF intervention is perceived to be a “caring contact” by inpatients, and that efforts to show that health care providers genuinely care comprise a systemic intervention. In terms of future work, it will be important to determine what could be done to better integrate Safety Planning and VLC approaches during the initial secession as well as whether additional post-baseline calls could help maintain initial positive responses to CLF and support greater utilization of this service following inpatient discharge. Depending on the strength of additional research on the topic, efforts to support implementation could build around a combination of services delivered by the inpatient treatment teak with potential for outreach calls from either the VCL or the local suicide prevention coordinators.

Somewhat surprisingly, the impact of the CLF intervention was not seen on use of the VCL, despite the focus of the intervention on encouraging and practicing VCL use and the previously-described findings that randomization to the CLF condition was associated with significantly greater increases in self-efficacy to use the VCL compared to EUC (Ilgen et al., 2020). Given the study’s exclusion criteria, no participants had used the VCL in the past 12-months, but over one in five participants in the total sample used the VCL during the follow-up interval. It is possible that the simple overview of the study and material shared in the EUC condition were sufficient to increase VCL use. It is also possible that the general discussion of help-seeking, combined with the practice of a practical strategy for reaching help, increased general willingness to disclose suicidal thoughts to others or helped to avoid engaging in suicidal behaviors in the future. More broadly, the present work underscores how little is known about patterns and predictors of Crisis Line use under real world circumstances in high-risk patients. The importance of this work is underscored by the fact that, even with structured encouragement, close to 80% of participants did not utilize the VCL during the year follow-up interval. Thus, a clear opportunity exists to develop and test additional strategies to encourage and support VCL utilization in high-risk individuals.

It is important to note several additional limitations. First, the study is based within the VA healthcare system that has a robust system for suicide risk assessment and management. Similarly, the sample was predominantly male Veterans with an average age of 47 years. The findings may not generalize to other healthcare systems or populations, including Veterans not enrolled in VHA services. In addition, the high level of treatment engagement in the VA may have limited the ability to detect the impact on VCL utilization. Specifically, supplementary analyses (not shown) indicate that approximately 80% of all participants attended an outpatient mental health appointment during the follow-up period. Although follow-up rates were reasonably high for the population, missing data could have influenced the results. The primary (VCL contacts) and secondary (suicidal behaviors) outcomes were measured via a researcher-administered interviews which may not have captured all events of interest. Studying suicidal behaviors is difficult due to the rarity of the outcome. Relatively few participants reported suicide attempts within the one-year follow-up time period (n=34 or 12.5% of the sample with usable follow-up data with an actual attempt; n=45 or 16.5% reporting any suicidal behavior as reported in Table 2, model 2), which was notably lower than the rates of suicide attempts reported within the literature in similar populations (Brown et al., 2005; Rudd et al., 2015). Therefore, the study utilized an analytic approach that examined multiple suicide behaviors as the main outcome. Similar findings were not seen when restricting the number of suicides examined per person, which may reflect a lack of power or that the intervention may have more impact on repeated attempts, which may be more common in those who are impulsive, than single attempts. Finally, seven participants (2.3%) died during the follow-up interval and no validated cause of death was available, which made examining suicide mortality as an outcome impractical.

CONCLUSIONS

The present study provides initial evidence for the impact of a brief intervention to increase use of crisis lines and prevent suicidal behaviors in suicidal Veterans treated in inpatient mental health programs. As others have noted, inpatient mental health settings see some of the most high-risk patients in health systems but usually do not provide evidence-based suicide-prevention services (Ghahramanlou-Holloway et al., 2020). To be widely adopted, suicide prevention services for inpatient psychiatry settings need to be relatively short and deliverable by existing clinical staff. The single-session CLF intervention meets these criteria, and, if the present results are replicated, the CLF intervention has the potential to reduce suicidal behaviors following mental health discharge.

References

  1. Acosta JD, Cerully JL, Wong EC, Sayers EL, Zaydman M, Meredith LS, Blum I, Nanda N, Tanielian T, & Ross R (2020). Veterans Crisis Line: Findings and Recommendations. [Google Scholar]
  2. Ajzen I (1991). The Theory of Planned Behavior. Organizational Behavior and Human Decision Processes, 50, 179–211. [Google Scholar]
  3. Ajzen I (2002). Perceived behavioral control, self-efficacy, locus of control, and the theory of planned behavior. Journal of Applied Social Psychology, 32(4), 665–683. 10.1111/j.1559-1816.2002.tb00236.x [DOI] [Google Scholar]
  4. Ajzen I (2011). The theory of planned behaviour: reactions and reflections. Psychol Health, 26(9), 1113–1127. 10.1080/08870446.2011.613995 [DOI] [PubMed] [Google Scholar]
  5. Ammerman BA, Carter SP, Walker K, Gebhardt HM, Rojas SM, Duman C, Reger GM, & Reger MA (2021). Characterizing Veterans crisis line callers and regional follow-up efforts. Military Psychology, 1–8. 10.1080/08995605.2021.1963631 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bandura A (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84(2), 191–215. 10.1037/0033-295X.84.2.191 [DOI] [PubMed] [Google Scholar]
  7. Bandura A (1982). Self-efficacy mechanism in human agency. American Psychologist, 37(2), 122–147. 10.1037/0003-066X.37.2.122 [DOI] [Google Scholar]
  8. Bandura A (2000). Self-efficacy: The foundation of agency. In Perrig WJ & Grob A (Eds.), Control of human behavior, mental processes, and consciousness: Essays in honor of the 60th birthday of August Flammer. (pp. 17–33). Lawrence Erlbaum Associates Publishers. http://proxy.lib.umich.edu/login?url=http://search.ebscohost.com/login.aspx?direct=true&db=psyh&AN=2000-08381-002&site=ehost-live&scope=site [Google Scholar]
  9. Brenner LA, & Barnes SM (2012). Facilitating Treatment Engagement During High-Risk Transition Periods: A Potential Suicide Prevention Strategy. American Journal of Public Health, 102(S1), S12–S14. 10.2105/ajph.2011.300581 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Britton PC, Bohnert KM, Ilgen MA, Kane C, Stephens B, & Pigeon WR (2017). Suicide mortality among male veterans discharged from Veterans Health Administration acute psychiatric units from 2005 to 2010. Social psychiatry and psychiatric epidemiology, 52(9), 1081–1087. [DOI] [PubMed] [Google Scholar]
  11. Britton PC, Crasta D, Bohnert KM, Kane C, Klein J, & Pigeon WR (2021). Shorter and longer-term risk for non-fatal suicide attempts among male US military veterans after discharge from psychiatric hospitalization. J Psychiatr Res, 143, 9–15. [DOI] [PubMed] [Google Scholar]
  12. Britton PC, Ilgen MA, Bohnert KM, Ashrafioun L, Kane C, Klein J, & Pigeon WR (2021). Shorter-and longer-term risk for suicide among male US military veterans in the year after discharge from psychiatric hospitalization. J Clin Psychiatry, 82(2), 27962. [DOI] [PubMed] [Google Scholar]
  13. Britton PC, Karras E, Stecker T, Klein J, Crasta D, Brenner LA, & Pigeon WR (in press). Veterans Crisis Line call outcomes: Distress, suicidal ideation, and suicidal urgency. Am J Prev Med. [DOI] [PubMed] [Google Scholar]
  14. Britton PC, Kopacz MS, Stephens B, & Bossarte RM (2016). Veterans Crisis Line Callers With and Without Prior VHA Service Use. Arch Suicide Res, 20(3), 314–322. 10.1080/13811118.2015.1017681 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Brown GK, Karlin BE, Trockel M, Gordienko M, Yesavage J, & Taylor CB (2016). Effectiveness of cognitive behavioral therapy for veterans with depression and suicidal ideation. Archives of Suicide Research, 20(4), 677–682. 10.1080/13811118.2016.1162238 [DOI] [PubMed] [Google Scholar]
  16. Brown GK, Ten Have T, Henriques GR, Xie SX, Hollander JE, & Beck AT (2005). Cognitive Therapy for the Prevention of Suicide AttemptsA Randomized Controlled Trial. Jama, 294(5), 563–570. 10.1001/jama.294.5.563%J JAMA [DOI] [PubMed] [Google Scholar]
  17. Carey MP, Carey KB, Maisto SA, Gordon CM, & Weinhardt LS (2001). Assessing sexual risk behaviour with the Timeline Followback (TLFB) approach: Continued development and psychometric evaluation with psychiatric outpatients. International Journal of STD & AIDS, 12(6), 365–375. 10.1258/0956462011923309 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. 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]
  19. Department of Veterans Affairs. (2020). PREVENTS: The President’s Roadmap to Empower Veterans and End a National Tragedy of Suicide. [Google Scholar]
  20. Fals-Stewart W, O’Farrell TJ, Freitas TT, McFarlin SK, & Rutigliano P (2000). The Timeline Followback reports of psychoactive substance use by drug-abusing patients: Psychometric properties. Journal of Consulting and Clinical Psychology, 68(1), 134–144. 10.1037/0022-006X.68.1.134 [DOI] [PubMed] [Google Scholar]
  21. Forte A, Buscajoni A, Fiorillo A, Pompili M, & Baldessarini RJ (2019). Suicidal Risk Following Hospital Discharge: A Review. Harvard review of psychiatry, 27(4), 209–216. 10.1097/hrp.0000000000000222 [DOI] [PubMed] [Google Scholar]
  22. Ghahramanlou-Holloway M, LaCroix JM, Perera KU, Neely L, Grammer G, Weaver J, Novak LA, Colborn V, & Lee-Tauler SY (2020). Inpatient psychiatric care following a suicide-related hospitalization: A pilot trial of Post-Admission Cognitive Therapy in a military medical center. Gen Hosp Psychiatry, 63, 46–53. 10.1016/j.genhosppsych.2018.11.006 [DOI] [PubMed] [Google Scholar]
  23. Gould MS, Kalafat J, HarrisMunfakh JL, & Kleinman M (2007). An Evaluation of Crisis Hotline Outcomes Part 2: Suicidal Callers. 37(3), 338–352. 10.1521/suli.2007.37.3.338 [DOI] [PubMed] [Google Scholar]
  24. Gould MS, Lake AM, Munfakh JL, Galfalvy H, Kleinman M, Williams C, Glass A, & McKeon R (2016). Helping callers to the National Suicide Prevention Lifeline who are at imminent risk of suicide: Evaluation of caller risk profiles and interventions implemented. Suicide and Life-Threatening Behavior, 46(2), 172–190. 10.1111/sltb.12182 [DOI] [PubMed] [Google Scholar]
  25. Hoffberg AS, Stearns-Yoder KA, & Brenner LA (2019). The Effectiveness of Crisis Line Services: A Systematic Review. Front Public Health, 7, 399. 10.3389/fpubh.2019.00399 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Hunt IM, Kapur N, Webb R, Robinson J, Burns J, Shaw J, & Appleby L (2009). Suicide in recently discharged psychiatric patients: a case-control study. Psychol Med, 39(3), 443–449. 10.1017/s0033291708003644 [DOI] [PubMed] [Google Scholar]
  27. Ilgen MA, Stewart HJ, Lhermitte SL, Pfeiffer PN, Britton PC, & Pope EB (2020). Developing and testing crisis line facilitation to encourage help seeking in adults receiving inpatient treatment for a suicidal crisis. Cognitive and Behavioral Practice. 10.1016/j.cbpra.2020.05.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Johnson LL, Muehler T, & Stacy MA (2021). Veterans’ satisfaction and perspectives on helpfulness of the Veterans Crisis Line. Suicide and Life-Threatening Behavior, 51(2), 263–273. [DOI] [PubMed] [Google Scholar]
  29. Joiner TE Jr., Walker RL, Rudd MD, & Jobes DA (1999). Scientizing and routinizing the assessment of suicidality in outpatient practice. Professional Psychology: Research and Practice, 30(5), 447–453. 10.1037/0735-7028.30.5.447 (Special Section: Assessing and Handling Suicidal Risk in Clinical Practice.) [DOI] [Google Scholar]
  30. Karras E, Lu N, Zuo G, Tu XM, Stephens B, Draper J, Thompson C, & Bossarte RM (2016). Measuring Associations of the Department of Veterans Affairs’ Suicide Prevention Campaign on the Use of Crisis Support Services. Suicide Life Threat Behav, 46(4), 447–456. 10.1111/sltb.12231 [DOI] [PubMed] [Google Scholar]
  31. Knox KL, Kemp J, McKeon R, & Katz IR (2012). Implementation and early utilization of a suicide hotline for veterans. American Journal of Public Health, 102(Suppl 1), S29–S32. 10.2105/AJPH.2011.300301 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Landes SJ, Kirchner JE, Areno JP, Reger MA, Abraham TH, Pitcock JA, Bollinger MJ, & Comtois KA (2019). Adapting and implementing Caring Contacts in a Department of Veterans Affairs emergency department: a pilot study protocol. Pilot Feasibility Stud, 5, 115. 10.1186/s40814-019-0503-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Linehan MM, Comtois KA, Brown MZ, Heard HL, & Wagner A (2006). Suicide Attempt Self-Injury Interview (SASII): Development, reliability, and validity of a scale to assess suicide attempts and intentional self-injury. Psychological Assessment, 18(3), 303–312. 10.1037/1040-3590.18.3.303 [DOI] [PubMed] [Google Scholar]
  34. Luxton DD, June JD, & Comtois KA (2013). Can postdischarge follow-up contacts prevent suicide and suicidal behavior? A review of the evidence. Crisis: The Journal of Crisis Intervention and Suicide Prevention, 34(1), 32–41. 10.1027/0227-5910/a000158 [DOI] [PubMed] [Google Scholar]
  35. Luxton DD, Smolenski DJ, Reger MA, Relova RMV, & Skopp NA (2020). Caring E-mails for Military and Veteran Suicide Prevention: A Randomized Controlled Trial. 50(1), 300–314. 10.1111/sltb.12589 [DOI] [PubMed] [Google Scholar]
  36. Meehan J, Kapur N, Hunt IM, Turnbull P, Robinson J, Bickley H, Parsons R, Flynn S, Burns J, Amos T, Shaw J, & Appleby L (2006). Suicide in mental health in-patients and within 3 months of discharge: National clinical survey. British Journal of Psychiatry, 188(2), 129–134. 10.1192/bjp.188.2.129 [DOI] [PubMed] [Google Scholar]
  37. Mishara BL, Chagnon F. o., Daigle M, Balan B, Raymond S, Marcoux I, Bardon C, Campbell JK, & Berman A (2007). Which Helper Behaviors and Intervention Styles are Related to Better Short-Term Outcomes in Telephone Crisis Intervention? Results from a Silent Monitoring Study of Calls to the U.S. 1–800-SUICIDE Network. Suicide and Life-Threatening Behavior, 37(3), 308–321. 10.1521/suli.2007.37.3.308 [DOI] [PubMed] [Google Scholar]
  38. Motto JA, & Bostrom AG (2001). A randomized controlled trial of postcrisis suicide prevention. Psychiatr Serv, 52(6), 828–833. 10.1176/appi.ps.52.6.828 [DOI] [PubMed] [Google Scholar]
  39. Office of Mental Health and Suicide Prevention. (2018). The National Strategy for the Prevention of Suicide in Veterans 2018-2028. [Google Scholar]
  40. Office of Mental Health and Suicide Prevention. (2021). National Veteran Suicide Prevention Annual Report. [Google Scholar]
  41. Office of the Surgeon General. (2021). The Surgeon General’s Call to Action to Implement the National Strategy for Suicide Prevention. [Google Scholar]
  42. Posner K, Brown GK, Stanley B, Brent DA, Yershova KV, Oquendo MA, Currier GW, Melvin GA, Greenhill L, Shen S, & Mann JJ (2011). The Columbia-Suicide Severity Rating Scale: Initial validity and internal consistency findings from three multisite studies with adolescents and adults. Am J Psychiatry, 168(12), 1266–1277. 10.1176/appi.ajp.2011.10111704 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Reger MA, Gebhardt HM, Lee JM, Ammerman BA, Tucker RP, Matarazzo BB, Wood AE, & Ruskin DA (2019). Veteran Preferences for the Caring Contacts Suicide Prevention Intervention. Suicide Life Threat Behav, 49(5), 1439–1451. 10.1111/sltb.12528 [DOI] [PubMed] [Google Scholar]
  44. Rubin R (2019). Task Force to Prevent Veteran Suicides. Jama, 322(4), 295–295. 10.1001/jama.2019.9854 [DOI] [PubMed] [Google Scholar]
  45. Rudd MD, Bryan CJ, Wertenberger EG, Peterson AL, Young-McCaughan S, Mintz J, Williams SR, Arne KA, Breitbach J, Delano K, Wilkinson E, & Bruce TO (2015). Brief cognitive-behavioral therapy effects on post-treatment suicide attempts in a military sample: results of a randomized clinical trial with 2-year follow-up. Am J Psychiatry, 172(5), 441–449. 10.1176/appi.ajp.2014.14070843 [DOI] [PubMed] [Google Scholar]
  46. Ruengorn C, Sanichwankul K, Niwatananun W, Mahatnirunkul S, Pumpaisalchai W, & Patumanond J (2011). Incidence and risk factors of suicide reattempts within 1 year after psychiatric hospital discharge in mood disorder patients. Clin Epidemiol, 3, 305–313. 10.2147/clep.S25444 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Sobell L, & Sobell M (1992). Timeline follow-back: A technique for assessing self-reported alcohol consumption. Measuring Alcohol Consumption: Psychosocial and Biochemical Methods, 41–72. [Google Scholar]
  48. Sobell LC, Maisto SA, Sobell MB, & Cooper AM (1979). Reliability of alcohol abusers’ self-reports of drinking behavior. Behaviour Research and Therapy, 17(2), 157–160. 10.1016/0005-7967(79)90025-1 [DOI] [PubMed] [Google Scholar]
  49. Stanley B, Brown GK, Brenner LA, Galfalvy HC, Currier GW, Knox KL, Chaudhury SR, Bush AL, & Green KL (2018). Comparison of the Safety Planning Intervention With Follow-up vs Usual Care of Suicidal Patients Treated in the Emergency Department. JAMA Psychiatry, 75(9), 894–900. 10.1001/jamapsychiatry.2018.1776 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Valenstein M, Kim HM, Ganoczy D, McCarthy JF, Zivin K, Austin KL, Hoggatt K, Eisenberg D, Piette JD, Blow FC, & Olfson M (2009). Higher-risk periods for suicide among VA patients receiving depression treatment: Prioritizing suicide prevention efforts. Journal of Affective Disorders, 112(1), 50–58. 10.1016/j.jad.2008.08.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Veterans Crisis Line. (2021). Veterans Crisis Line: What It Is. https://www.veteranscrisisline.net/about/what-is-vcl [Google Scholar]

RESOURCES