Nature and determinants of suicidal ideation among U.S. veterans: Results from the national health and resilience in veterans study

Author’s Accepted Manuscript NATURE AND DETERMINANTS OF SUICIDAL IDEATION AMONG U.S. VETERANS: RESULTS FROM THE NATIONAL HEALTH AND RESILIENCE IN VETERANS STUDY Noelle B. Smith, Natalie Mota, Jack Tsai, Lindsey Monteith, Ilan Harpaz-Rotem, Steven M. Southwick, Robert H. Pietrzak

PII: DOI: Reference:

www.elsevier.com/locate/jad

S0165-0327(15)31491-9 http://dx.doi.org/10.1016/j.jad.2016.02.069 JAD8087

To appear in: Journal of Affective Disorders Received date: 31 December 2015 Revised date: 25 February 2016 Accepted date: 28 February 2016 Cite this article as: Noelle B. Smith, Natalie Mota, Jack Tsai, Lindsey Monteith, Ilan Harpaz-Rotem, Steven M. Southwick and Robert H. Pietrzak, NATURE AND DETERMINANTS OF SUICIDAL IDEATION AMONG U.S. VETERANS: RESULTS FROM THE NATIONAL HEALTH AND RESILIENCE IN VETERANS STUDY, Journal of Affective Disorders, http://dx.doi.org/10.1016/j.jad.2016.02.069 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 galley proof before it is published in its final citable 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.

NATURE AND DETERMINANTS OF SUICIDAL IDEATION AMONG U.S. VETERANS: RESULTS FROM THE NATIONAL HEALTH AND RESILIENCE IN VETERANS STUDY Noelle B. Smith1,2*, Natalie Mota1,2, Jack Tsai2,3, Lindsey Monteith4,5, Ilan Harpaz-Rotem1,2,6, Steven M. Southwick1,2, Robert H. Pietrzak1,2 1

U.S. Department of Veterans Affairs National Center for PTSD, VA Connecticut Healthcare System, West Haven, CT, USA 2 Department of Psychiatry, Yale School of Medicine, New Haven, CT, USA 3 New England Mental Illness, Research, Education, and Clinical Center, VA Connecticut Healthcare System, West Haven, CT, USA 4 Denver VA Medical Center, Rocky Mountain MIRECC, Denver, CO, USA 5 Department of Psychiatry, University of Colorado Anschutz Medical Campus, Aurora, CO, USA 6 VA Northeast Program Evaluation Center, West Haven, CT, USA

*

Correspondence concerning this article should be addressed to: Noelle B. Smith, Ph.D., National Center for Posttraumatic Stress Disorder, VA Connecticut Healthcare system, Department of Psychiatry, Yale School of Medicine, 950 Campbell Avenue/116B, West Haven, CT 06516 ([email protected]).



The National Health and Resilience in Veterans Study was supported by the U.S. Department of Veterans Affairs National Center for Posttraumatic Stress Disorder. Preparation of this manuscript was supported in part by the Office of Academic Affiliations, Advanced Fellowship Program in Mental Illness Research and Treatment Department of Veterans Affairs. Dr. Pietrzak is a scientific consultant to Cogstate and Mitsubishi Tanabe Pharma America, Inc. for work that is unrelated to this manuscript. The other authors have no disclosures. The authors assert that all procedures contributing to this work comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

2

Abstract Background Suicidal thoughts and behaviors among U.S. military veterans are a major public health concern. To date, however, scarce data are available regarding the nature and correlates of suicidal ideation (SI) among U.S. veterans. This study evaluated the prevalence and correlates of suicidal ideation in a contemporary, nationally representative, 2-year prospective cohort study. Method Data were analyzed from a total of 2,157 U.S. veterans who participated in the National Health and Resilience Veterans Study (NHRVS; Wave 1 conducted in 2011; Wave 2 in 2013). Veterans completed measures assessing SI, sociodemographic characteristics, and potential risk and protective correlates. Results The majority of veterans (86.3%) denied SI at either time point, 5.0% had SI onset (no SI at Wave 1, SI at Wave 2), 4.9% chronic SI (SI at Waves 1 and 2), and 3.8% had remitted SI (SI at Wave 1, no SI Wave 2). Greater Wave 1 psychiatric distress was associated with increased likelihood of chronic SI (relative risk ratio [RRR]=3.72), remitted SI (RRR=3.38), SI onset (RRR=1.48); greater Wave 1 physical health difficulties were additionally associated with chronic SI (RRR=1.64) and SI onset (RRR=1.47); and Wave 1 substance abuse history was associated with chronic SI (RRR 1.57). Greater protective psychosocial characteristics (e.g., resilience, gratitude) at Wave 1 were negatively related to SI onset (RRR=0.57); and greater social connectedness at Wave 1, specifically perceived social support and secure attachment style, was negatively associated with SI onset (RRR=0.75) and remitted SI (RRR=0.44), respectively. Limitations

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

3

Suicidal ideation was assessed using a past two-week timeframe, and the limited duration of follow-up precludes conclusions regarding more dynamic changes in SI over time.. Conclusions These results indicate that a significant minority (13.7%) of U.S. veterans has chronic, new-onset, or remitted SI. Prevention and treatment efforts designed to mitigate psychiatric and physical health difficulties, and bolster social connectedness and protective psychosocial characteristics may help mitigate risk for SI.

Keywords: suicidal ideation; longitudinal time points; nationally representative veteran sample; protective and risk factors

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

4

Introduction

Suicide among U.S. military veterans is a significant public health concern, with veterans accounting for roughly 22% of suicides nationally (Kemp & Bossarte, 2012). Compared to civilians, veterans in the general population are twice as likely to die by suicide (Kaplan et al., 2007; Gibbons et al., 2012), although the magnitude of these differences is debated and appears to vary across age groups, with the most striking difference being in the relative risk for 17-24 year old male veterans, in which the relative risk is 3.8 times higher than same-age men without military service (Gibbons et al., 2012). Suicidal ideation (SI) is often a precursor to a suicide attempt or suicidal death (e.g., Dobscha et al., 2014; Miranda et al., 2014). Given the high prevalence of SI and associated with suicidal behavior, identifying correlates of SI could result in more targeted outreach and intervention efforts that can aid in suicide prevention.

The prevalence of SI among veterans varies depending on the timeframe and population studied. Among veterans and active members of the military, the lifetime prevalence of SI is quite high (e.g., 13.9%; Nock et al., 2014; 33.4% Bryan & Bryan, 2014). However, the prevalence is lower when examining shorter timeframes, with estimates ranging from 3.3% in the past year among adults with history of military service (Blosnich et al., 2014) to 21.6% among treatment-seeking Iraq and Afghanistan veterans (Pietrzak et al., 2011). Other studies with service members (e.g., currently in the military), as well as veterans have also obtained estimates within this range (e.g., Pietrzak et al., 2010; Bryan & Bryan, 2014; Ramsawh et al., 2014). With such a wide range in the prevalence of SI among veterans (e.g., 3.3 to 33.4%), greater understanding into the nature of SI over time is necessary. Further, to date, much of the extant literature has focused on specific samples of veterans (e.g., veterans utilizing Veterans Health Administration care, treatment

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

5

seeking samples, college students) and data are lacking regarding the prevalence and presentation of SI in nationally representative samples of U.S. veterans. Such data are important, as they can inform the burden and changes in SI over time in the entire U.S. veteran population.

A large body of research has attempted to identify factors associated with suicidal thoughts, plans, and attempts. In a large cross-national investigation, sociodemographic risk factors for SI included younger age, female sex, unmarried marital status, and less education (Nock et al., 2008). Similar factors are associated with SI in veteran populations, although results are mixed, such that Native American race, younger and older ages, and male sex are linked to increased risk of SI (e.g., Allen et al., 2005; Desai et al., 2005; Desai et al., 2008; Jakupcak et al., 2010; Lemaire & Graham, 2011; Bryan & Bryan, 2014). In a longitudinal population-based study, anxiety disorders (e.g., social phobia, simple phobia, generalized anxiety disorder, panic disorder, agoraphobia, and obsessivecompulsive disorder) were significantly associated with suicidal ideation and suicide attempts (Sareen et al., 2005). Other studies have found that ast year substance abuse and mood and anxiety disorders were associated with recurrent SI (Dugas, Low, O’Loughlin, & O’Loughlin, 2015) and anxiety symptoms were associated with the onset of SI (Baek et al., 2015). Psychiatric risk factors commonly associated with SI and suicide attempts in veteran populations include major depressive disorder, PTSD, anxiety disorders, and substance use disorders (e.g., Pietrzak et al., 2010; Lemaire & Graham, 2011; Fanning & Pietrzak, 2013; Nock et al., 2013; Conner et al., 2014; Dobscha et al., 2014; Thompson et al., 2014; Wisco et al., 2014; Ejdesgaard et al., 2015; Zivin et al., 2015). Physical health problems, such as heart attack/stroke, high blood pressure, arthritis, chronic pain, respiratory problems, and gastrointestinal disorders, are also associated with increased rates of SI (Fanning & Pietrzak, 2010; Scott et al., 2010; Thompson et al., 2014). Collectively, results of these

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

6

studies suggest that a broad range of sociodemographic, psychiatric, and physical health factors are correlated with SI in both civilian and veteran populations.

While a large body of research has examined risk factors for SI, relatively fewer studies have examined protective factors. Available data suggest that characteristics such as an optimistic or positive attribution style and increased agency (e.g., one’s sense that they are in control) may be negatively related to SI (for a review, see Johnson et al., 2011). Among Veterans, sense of purpose and control (Pietrzak et al., 2010), acceptance of change (Pietrzak et al., 2011; Youssef et al., 2013), value importance and success (Bahraini et al., 2013), and greater meaning in life (Braden et al., 2015) have been associated negatively with SI. Further, greater social connectedness (Pietrzak et al., 2010; Pietrzak et al., 2011; Fanning & Pietrzak, 2013), social support (Lehmaire & Graham, 2011; Kleiman & Liu, 2013; Ejdesgaard et al., 2015), and secure relationships (Youssef et al., 2013) have also been found to be protective. While these studies help inform models of suicide risk, most are based on cross-sectional data; thus, it is unclear whether these factors can influence presentations of SI over time. Consequently, is unclear whether these factors are in fact protective or whether distressed individuals tend to score lower on measures of protective factors. Longitudinal studies are needed to address this question.

Better characterization of SI over time can be helpful in understanding suicide risk, as risk factors may have a delayed impact and better identification of high-risk individuals could lead to enhanced clinical care (Gutierrez, 2014). Similarly, better understanding of the nature and prevalence of SI over time among veterans, which may be characterized by remitting, onset, and

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

7

chronic SI, can provide insight into the predominant patterns and determinants of SI, and inform targets for prevention and treatment. For example, among Danish soldiers, risk for SI differed based on different PTSD symptom trajectories over a three-year period (Madsen et al., 2014). In Israeli prisoner of war veterans, SI increased over time and was related to PTSD symptoms (Zerach et al., 2014). Among a sample of acutely suicidal U.S. Army soldiers, chronic stressors were associated with more persistent SI, whereas low-to-average chronic stress was associated with the remittance of SI at the six-month follow-up (Bryan et al., 2015). Thus, it is likely that risk and protective correlates for different presentations of SI may vary. To date, however, predominant presentations of SI across time have not been examined in a nationally representative sample of U.S. military veterans.

To address the aforementioned gaps in the literature, the aims of this study were to analyze data from a large, contemporary, and nationally-representative sample of U.S. veterans and evaluate: (1) the prevalence of predominant, population-based presentations of SI (i.e., chronic, onset, remitted SI) over a two-year period; and (2) how a comprehensive range of sociodemographic, risk, and protective variables assessed at baseline relates to predominant presentations of SI.

Method

Participants A nationally representative sample of U.S. military veterans who completed two survey waves (Waves 1 and 2; described below) of the National Health and Resilience Veterans Study (NHRVS), a prospective cohort study of U.S. veterans ages 18 and older, participated in this study.

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

8

This sample was ascertained from a larger, nationally representative sample of more than 50,000 U.S. households who are part of KnowledgePanel, a survey research panel maintained by GfK Knowledge Networks, Inc. (Menlo Park, CA). KnowledgePanel uses probability-based sampling of addresses from the U.S. Postal Service’s Delivery Sequence File (DSF). The key advantage of this methodology is that it allows sampling of almost all U.S. households. Regardless of household telephone status, all households can be reached and contacted through postal mail, including households without a home telephone number or Internet access (potential participants without computer and/or Internet access are provided with them). Wave 1 data were collected between October and December 2011. Wave 2 data were collected two years later between October and December 2013. Participants completed a 45-60 minute anonymous online survey in both waves. Regarding sample ascertainment, of the 4,750 veterans sampled for the NHRVS, 3,408 (71.7%) responded to an invitation to participate and completed a screening question to confirm their study eligibility (current or past active military status). Of these respondents, 3,188 (93.5%) confirmed their current or past active military status and, of these, 3,157 (92.6%) completed a confidential, 60-minute online survey. Of these individuals, 2,157 (68.3%) completed Waves 1 and 2, and 2,107 (66.7% of Wave 1) completed measures of SI at both Waves 1 and 2. Post-stratification weights based on the demographic composition of veterans in KnowledgePanel were calibrated against the demographic composition of the U.S. adult population using data from the U.S. Census Bureau Current Population Survey (U.S. Census Bureau, 2012). Post-stratification weights were computed by statisticians at GfK Knowledge Networks, Inc., and applied in inferential analyses in order to permit generalizability to the U.S. veteran population. The study was approved by the Human Subjects Subcommittee of the VA Connecticut Healthcare System, and all participants provided informed consent.

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

9

Assessments Sociodemographic Variables. Sex (male, female), age (continuous), marital status (unmarried, married/living with partner), race (non-Caucasian, Caucasian), education (up to high school diploma, more than high school), income (less than $60,000, $60,000 or more), and combat status (non-combat, combat) were assessed. Veterans also answered a question regarding whether they had ever received mental health treatment (no/yes).

Suicidal Ideation. Suicidal ideation (SI) was assessed at both survey waves using a modified item from the Patient Health Questionnaire-9 (Kroenke et al., 2002), which assessed passive and active SI. The original question, which asks individuals to rate how often in the past 2 weeks they had thoughts that they would be better off dead or of hurting themselves, was separated into two items that assessed how often over the past two weeks an individual had thoughts: a) that they would be better off dead (passive SI); and b) of hurting themselves in some way (active SI). Response options are: 0=not at all; 1=several days; 2=more than half the days; 3=nearly every day. The presence of current SI was operationalized as a score of “1” or higher on either question. Presence of SI was determined based on responses to SI questions at Waves 1 and 2. No SI was defined as the absence of SI at both Waves 1 and 2 (i.e., endorsement of “0” at both waves). Remitted SI was defined as the presence of SI at Wave 1 and the absence of SI at Wave 2. SI onset was defined as the absence of SI at Wave 1 and the presence of SI at Wave 2. Chronic SI was defined as the presence of SI at both Waves 1 and 2.

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

10

Risk Correlates (Assessed at Wave 1). As described elsewhere (Pietrzak & Cook, 2013), exploratory factor analyses (EFAs) were conducted to reduce thematically and highly correlated variables into factors, which helps to prevent problems related to multicollinearity in multivariable analyses. Factors assessing psychiatric distress, substance use history, and physical health difficulties were included as potential risk factors. Lifetime history of suicide attempt reported at Wave 1 was also included as a potential risk factor. Table 1 provides a detailed description of component measures included in each factor, as well as the instruments used to assess them.

Protective Correlates (Assessed at Wave 1). EFAs were also conducted to reduce thematically and highly correlated protective correlates into factors. Factors assessing protective psychosocial characteristics (i.e., resilience, purpose in life, dispositional gratitude, community integration, and dispositional optimism), social connectedness, and active lifestyle were included as potential protective correlates of SI. Table 1 provides a detailed description of each of these factors and the instruments used to assess them.

Data Analysis

Descriptive statistics were conducted to identify means, standard deviations, and ranges for the study variables. Univariate analyses of variance (ANOVAs) and chi-square analyses were then conducted to identify associations between sociodemographic and military variables, and the outcome variable (four-level variable of SI presentation over the 2-year study period: No SI, Remitted SI, SI onset, and Chronic SI). Sociodemographic variables that were associated with SI at the p < .05 level in bivariate analyses were included in a subsequent multinomial logistic

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

11

regression analysis, which examined the relation between sociodemographic and potential risk (i.e., psychiatric, physical health, and substance use factors, previous suicide attempt) and resilience variables (i.e., protective psychosocial, social connectedness, and active lifestyle factors) at Wave 1 and SI groups (i.e., No SI, Remitted SI, SI onset, and Chronic SI); the No SI group served as the reference group. Table 1 includes detailed description of all the risk and protective factors analyzed in this study. Additional analyses examined Wave 1 predictors of Remitted SI with Chronic SI as the reference group, in order to better understand factors associated with remission of SI. Post-hoc analyses were conducted to identify aspects of psychiatric distress (e.g., PTSD, depressive symptoms) and protective psychosocial variables (e.g., resilience, gratitude) related to SI; to control for Type I error, alpha <.01 was considered statistically significant in these analyses.

Results Sample characteristics

The final sample included 2,107 veterans who completed Waves 1 and 2 of the NHRVS and answered questions about SI at both survey waves; of the 2,157 who completed Wave 1 and 2, 50 were excluded because they did not answer SI questions at Wave 1 and/or Wave 2. The average age of participants was 60.3 (SD = 15.0; range 21-96); the majority were male (weighted 91.7%, n=1,919), Caucasian (77.8%, n=1,787), had some college or higher education (68.1%, n=1,803), were married/living with a partner (74.5%, n=1,668), were non-combat veterans (67.8%, n=1,380), and had a household income less than $60,000 (53.9%, n=980).

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

12

Of the veterans who completed both waves, the prevalence of the SI groups were as follows: No SI (n=1,859; 86.3%), SI onset (n=100; 5.0%), Chronic SI (n=84; 4.9%), and Remitted SI (n=64; 3.8%). Among veterans who completed measures of SI at Waves 1 and 2 (N=2,107), the prevalence of SI varied. Combined active and passive SI had the highest prevalence at both waves (Wave 1: 8.7%; Wave 2: 9.9%) with passive SI endorsed at a greater frequency (Wave 1: 8.4%, Wave 2: 9.7%) than active SI (Wave 1: 3.5%, Wave 2: 4.4%).

Predictors of Suicidal Ideation Over Time (Remitted, Onset, Chronic) Age, marital status, race, education, and combat status differed significantly (p < .05) by SI group and were thus included as covariates in subsequent multivariable analyses. Table 2 includes SI group comparisons and information regarding sample characteristics. Among suicidal ideators, rates of previous suicide attempts ranged from 11.3% in the SI onset group to 28.4% in the remitted SI group, as did the proportion reporting mental health treatment (SI onset: 34.7% to chronic SI: 74.7%). Table 3 includes results from multinomial regression analyses predicting SI patterns, outlined in the text below.

Wave 1 Predictors of Remitted SI. Compared to the No SI group, older age, greater education (college or higher), and psychiatric distress were associated with Remitted SI, whereas lifetime suicide attempt and higher scores on the social connectedness factor were negatively associated with Remitted SI (see Table 3). Post-hoc analyses revealed that secure attachment style (relative risk ratio [RRR]=4.37, 95% confidence interval [CI]=2.03, 9.45, p<.001), as well as depression (RRR=2.44, 95% CI[1.94, 3.08], p<.001) and PTSD (RRR=1.04, 95% CI[1.01, 1.07], p<.01) symptoms were associated with greater likelihood of remitted suicidal ideation compared to No SI.

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

13

Wave 1 Predictors of SI Onset. Compared to the No SI group, higher psychiatric distress and physical health difficulties factor scores were associated with SI onset, while higher protective psychosocial characteristics and social connectedness factor scores were negatively associated with SI Onset (see Table 3). Post-hoc analyses revealed that greater perceived social support (RRR=0.91, 95% CI[0.86, 0.96], p=.001) was negatively associated with SI onset, whereas disability in independent activities of daily living was associated with increased risk of SI onset (RRR=1.46, 95% CI[1.26, 1.70], p <.001). None of the specific component variables comprising the protective psychosocial and psychiatric distress factors were significant in post-hoc analyses.

Wave 1 Predictors of Chronic SI. Relative to the No SI group, marital status, and greater psychiatric distress, physical health difficulties, and substance abuse history factors were associated with Chronic SI (see Table 3). Post-hoc analyses revealed that depressive symptoms (RRR=3.15, 95% CI[2.49, 3.99], p<.01) and disability in independent activities of daily living (RRR= 1.61, 95% CI[1.46, 1.81], p=.001) was positively associated with chronic SI.

Wave 1 Predictors of Remitted SI Compared to Chronic SI. Compared to the Chronic SI group, older age and higher education were associated with remitted SI, while lifetime suicide attempt, physical health difficulties, substance use, and social connectedness were negatively associated with Remitted SI (see Table 3). Post-hoc analyses revealed that disability in independent activities of daily living (RRR= .61, 95% CI[.46, .81], p=.001) and alcohol use disorder (RRR=0.26, 95% CI[0.10, 0.64], p<.01) was negatively associated with Remitted SI, whereas a secure attachment style (RRR=11.20, 95% CI[3.93, 31.90], p<.001) positively predicted Remitted SI.

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

14

Discussion

This is the first study of which we are aware to examine the nature and determinants of predominant, population-based patterns of suicidal ideation (SI) in a contemporary, nationally representative cohort of veterans. Results revealed that 4.9% of veterans had a chronic SI over the two-year assessment period, with an additional 5.0% and 3.8% experiencing SI onset and remission, respectively. At Wave 2, roughly ten percent (9.9%) of veterans reported SI in the past two weeks, which is slightly lower than other selected samples of Iraq and Afghanistan veterans (e.g., Pietrzak et al., 2011), yet slightly higher than the cross-national lifetime prevalence (which was not specific to veterans) reported by Nock and colleagues (9.2%; Nock et al., 2008). With one in ten veterans reporting SI in the past two weeks at Wave 2, the population-based burden of suicidal ideation in this population is clear. Our results further highlight the dynamic nature of SI, as evidenced by meaningful proportions of U.S. veterans reporting changes in their SI over time. These changes underscore the importance of periodic monitoring of SI, irrespective of sociodemographic characteristics, as these factors were not consistently predictive of SI. Understanding the nature of SI over time is critical to more refined and useful assessment to determine the most at-risk groups (Gutierrez, 2014).

This study also identified risk and protective factors associated with predominant, populationbased patterns of SI over a 2-year period. These results shed light on potential distal factors associated with different patterns of SI over an intermediate period of time. When comparing the three presentations of SI compared with No SI, only Wave 1 psychiatric distress consistently

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

15

exhibited significant associations with all presentations of SI, including chronic, onset, and remitted SI. Greater severity of depressive symptoms was independently related to chronic SI, while greater severity of both depressive and PTSD symptoms were associated with remitted SI. In our study, correlations between SI and depression were high, but our results indicate that suicidal ideation is distinct from depression. The robust association between psychiatric distress and SI is consistent with previous work that has highlighted the relation between mood disorders with suicidal thoughts and behaviors (e.g., Nock et al., 2008; Thomson et al., 2014), and depression specifically (e.g., Pietrzak et al., 2010; Lemaire & Graham, 2011; Fanning & Pietrzak, 2013; Dobscha et al., 2014; Thompson et al., 2014; Zivin et al., 2015). Notably, low rates of mental health treatment utilization were observed among some groups of suicide ideators, with 65% of the SI onset group having never engaged in any mental health treatment. This finding suggests that enhancement of outreach efforts to veterans at increased risk for suicidality may be useful. For example, conducting SI screenings in primary care could be helpful, particularly in light of data indicating that over 50% of veterans engaging in a non-fatal suicide attempt were last seen in primary care prior to that event (Kemp & Bossarte, 2012).

Results regarding risk factors for SI were consistent with prior work suggesting the importance of comorbidities in assessing patterns of SI. For example, while only greater psychiatric distress was associated with remitted SI, both psychiatric distress and physical health difficulties predicted SI onset; and psychiatric distress, physical health difficulties, and substance use history predicted chronic SI. Given that physical health difficulties and substance use history differentiated chronic SI and remitted SI groups, assessment and monitoring of these symptoms may be useful in gauging ongoing SI risk in veterans. Prior work has similarly highlighted the importance of physical health

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

16

and substance use in addition to psychiatric correlates in relation to SI (e.g., Scott et al., 2010; Fanning & Pietrzak, 2013; Chan et al., 2014; Thompson et al., 2014). It is likely that physical health and substance use risk factors contribute to, or are associated with, the chronicity of psychiatric distress, which in turn maintains SI (Bryan et al., 2015). Targeting these risk factors could thus be helpful in mitigating risk for chronic SI.

Among protective psychosocial factors, social connectedness emerged as a protective factor for both remitted SI and SI onset, and also predicted remission of SI related to the chronic SI group. These findings are consistent with previous work identifying an association between greater social support and lower rates of SI (e.g., Pietrzak et al., 2010; Lemaire & Graham, 2011; Pietrzak et al., 2011; Fanning & Pietrzak, 2013; Kleiman & Liu, 2013; Youssef et al., 2013; Ejdesgaard et al., 2015). Our results extend these findings to suggest that bolstering social connectedness could be particularly helpful in reducing risk for SI and possibly promoting remission from SI in the general population of U.S veterans. Social connectedness could be targeted through programs seeking to promote veteran support networks (e.g.,peer support program; Nelson, Abraham, Walters, Pfeiffer, & Valenstein, 2014) and present modifiable targets of treatment that could be particularly helpful for individuals with more intermittent SI (e.g., remitted SI, SI onset).

Of note, our results suggest that greater psychiatric distress, physical health difficulties, and substance use problems are linked to chronic SI, and that social connectedness and protective psychosocial characteristics do not buffer this increased risk over and above the effects of these risk factors. Although social connectedness was associated with remitted SI and SI onset, this was not the case for chronic SI. As such, targeting of psychiatric, physical health, and substance use

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

17

problems might be more helpful in mitigating chronic SI in this subset of veterans. Although SI is often related to suicide attempts and death by suicide, not everyone that endorses SI goes on to attempt or die by suicide (e.g., Gvion, Horesh, Levi-Belz, & Apter, 2015). Specifically, 9.2% of individuals report suicidal ideation during their lifetime, only 2.7% attempt suicide (Nock et al., 2008) suggesting that a focus on the short-term prediction of suicidal behavior remains important (Glenn & Nock, 2014).

Limitations. The present study has some limitations. First, questions assessing SI specifically asked about SI in the preceding two weeks, thereby limiting our understanding of SI presentations to the two relatively narrow timeframes assessed. Although suicidal ideation is thought to be relatively stable across time (Williams, Crane, Barnhofer, Van der Does, & Segal, 2006) and has been shown to be highly correlated across time points (assessed 12 months apart; r=.6; Taylor et al., 2011), there is, as we also highlight, fluctuation across time. Suicidal ideation likely fluctuated between the two assessments and we did not capture those variations in this study. Second, roughly 1/3 of the sample from Wave 1 did not complete the Wave 2 survey due to attrition; while application of post-stratification weights helps to ensure that results remain representative of the U.S. veteran population, it is possible that more symptomatic veterans were more likely to drop out of the study, resulting in underestimates of the prevalence of SI. Third, the small number of suicide attempts at Wave 2 reported precluded examination of prospective predictors of suicide attempts. Fourth, further details regarding suicidal ideation (e.g., intensity and seriousness) were not examined in the present study; such details may improve prediction of suicidal behaviors (Miranda et al., 2014). Fifth, there are likely other factors that were not assessed in the present study that may have contributed to the SI risk, such as financial and interpersonal stress (Wang et

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

18

al., 2015). Additionally, our sample was a nationally representative of U.S. Veterans, reflective of the aging Veteran population and inclusive of a particularly at-risk group, as a majority of Veterans who die by suicide are age 50 and older (Veterans Health Administration, 2012). It is possible that patterns of SI in subpopulations of veterans (e.g., younger veterans, female veterans, combat veterans).

Despite these limitations, this study provides contemporary, nationally representative data regarding the predominant, population-based patterns of SI in U.S. veterans. Results suggest that a significant minority of veterans (13.7%) endorsed SI at either Wave 1 or 2. They further suggest that psychiatric distress and physical health problems are key risk factors for both chronic and new-onset SI; and that bolstering of social connectedness, particularly by promoting secure attachment style and perceived social support, may help mitigate risk for SI. Further research is needed to examine the generalizability of the results of this study to subpopulations of veterans (e.g., female veterans, combat veterans); to determine predominant population-based trajectories of SI over longer periods of time and to characterize both baseline and time-varying covariates associated with them; and to evaluate the efficacy of prevention and treatment efforts designed to mitigate key risk factors and to enhance key protective factors associated with suicide risk in veterans and other populations at risk for suicide.

SUICIDAL IDEATION IN U.S. MILITARY VETERANS Contributors Noelle B. Smith Natalie Mota Jack Tsai Lindsey Monteith Ilan Harpaz-Rotem Steven Southwick Robert Pietrzak

19

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

20

References Allen, J. P., G. Cross, Swanner, J., 2005. Suicide in the Army: a review of current information. Military Medicine, 170, 580-584. Baek, J.H., Heo, J.Y., Fava, M., Mischoulon, D., Nierenberg, A., Hong, J.P., Roh, S., Jeon, H.J. (2015). Anxiety symptoms are linked to new-onset suicidal ideation after six months of follow-up in outpatients with major depressive disorder. Journal of Affective Disorders, 187, 183-197. Bahraini, N. H., Devore, M. D., Monteith, L. L., Forster, J. E., Bensen, S., & Brenner, L. A., 2013. The role of value importance and success in understanding suicidal ideation among veterans. Journal of Contextual Behavioral Science, 2(1), 31-38. Blosnich, J. R., Gordon, A. J., & Bossarte, R. M., 2014. Suicidal ideation and mental distress among adults with military service history: results from 5 US states, 2010. American journal of public health, 104, S595-S602. Braden, A., Overholser, J., Fisher, L., & Ridley, J., 2015. Life meaning is associated with suicidal ideation among depressed veterans. Death studies, 39, 24-29. Bryan, C. J., & Bryan, A. O., 2014. Sociodemographic correlates of suicidal thoughts and behaviors among college student service members/veterans. Journal of American College Health. Bryan, C. J., Clemans, T. A., Leeson, B., & Rudd, M. D., 2015. Acute vs. chronic stressors, multiple suicide attempts, and persistent suicide ideation in US soldiers. The Journal of nervous and mental disease, 203(1), 48-53.

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

21

Campbell-Sills, L., & Stein, M. B., 2007. Psychometric analysis and refinement of the ConnorDavidson Resilience Scale (CD-RISC): validation of a 10-item measure of resilience. Journal of traumatic stress, 20, 1019. Chan, L. F., Shamsul, A. S., & Maniam, T., 2014. Are predictors of future suicide attempts and the transition from suicidal ideation to suicide attempts shared or distinct: A 12-month prospective study among patients with depressive disorders. Psychiatry research, 220, 867873. Conner, K. R., Bossarte, R. M., He, H., Arora, J., Lu, N., Tu, X. M., & Katz, I. R., 2014. Posttraumatic stress disorder and suicide in 5.9 million individuals receiving care in the veterans health administration health system. Journal of affective Disorders, 166, 1-5. Desai, M. M., Rosenheck, R. A., & Desai, R. A., 2008. Time trends and predictors of suicide among mental health outpatients in the Department of Veterans Affairs. The journal of behavioral health services & research, 35(1), 115-124. Desai, R.A., Dausey, D.J., & Rosenheck, R.A., 2005. Mental health service delivery and suicide risk: The role of individual patient and facility factors. The American journal of psychiatry, 162(2), 311-318. Dobscha, S. K., Denneson, L. M., Kovas, A. E., Teo, A., Forsberg, C. W., Kaplan, M. S., ... & McFarland, B. H., 2014. Correlates of suicide among veterans treated in primary care: Case–control study of a nationally representative sample. Journal of general internal medicine, 29, 853-860. Dugas, E.N., Low, N.C., O’Loughlin, E.K., & O’Loughlin, J.L. (2015). Recurrent suicidal ideation in young adults. Canadian Journal of Public Health, 106, e303-307.

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

22

Ejdesgaard, B. A., Zøllner, L., Jensen, B. F., Jørgensen, H. O., & Kähler, H., 2015. Risk and protective factors for suicidal ideation and suicide attempts among deployed Danish soldiers from 1990 to 2009. Military medicine, 180, 61-67. Fanning, J. R., & Pietrzak, R. H., 2013. Suicidality among older male veterans in the United States: Results from the National Health and Resilience in Veterans Study. Journal of psychiatric research, 47(, 1766-1775. Gibbons, R. D., Brown, C.H., & Hur, K., 2012. Is the rate of suicide among veterans elevated? Am J Public Health, 102 Suppl 1, S17-19. Gutierrez, P. M., 2014. Viewing suicide risk through a new lens: the benefits of examining symptom trajectories. Journal of clinical psychiatry, 75, e945-e946. Hardy, S. E., & Gill, T. M., 2004. Recovery from disability among community-dwelling older persons. Jama, 291, 1596-1602. Hazan, C., & Shaver, P. R., 1990. Love and work: An attachment-theoretical perspective. Journal of Personality and social Psychology, 59(2), 270. Heatherton, T. F., Kozlowski, L. T., Frecker, R. C., & Fagerstrom, K.O., 1991. The Fagerström test for nicotine dependence: a revision of the Fagerstrom Tolerance Questionnaire. British journal of addiction, 86, 1119-1127. Jakupcak, M., Vannoy, S., Imel, Z., Cook, J. W., Fontana, A., Rosenheck, R., & McFall, M., 2010. Does PTSD moderate the relationship between social support and suicide risk in Iraq and Afghanistan War Veterans seeking mental health treatment?. Depression and anxiety, 27, 1001-1005. Johnson, J., Wood, A. M., Gooding, P., Taylor, P. J., & Tarrier, N., 2011. Resilience to suicidality: The buffering hypothesis. Clinical psychology review, 31, 563-591.

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

23

Kaplan, M. S., Huguet, N., McFarland, B.H., & Newsom, J.T., 2007. Suicide among male veterans: a prospective population-based study. J Epidemiol Community Health. 61, 619624. Kashdan, T. B., Gallagher, M. W., Silvia, P. J., Winterstein, B. P., Breen, W. E., Terhar, D., & Steger, M. F., 2009. The curiosity and exploration inventory-II: Development, factor structure, and psychometrics. Journal of research in personality, 43, 987-998. Kemp, J. & Bossarte, R. for Department of Veterans Affairs, Mental Health Services, and Suicide Prevention Program, 2012. Suicide data report, 2012. Accessed from: http://www.va.gov/opa/docs/suicide-data-report-2012-final.pdf Kleiman, E. M., & Liu, R. T., 2013. Social support as a protective factor in suicide: Findings from two nationally representative samples. Journal of affective disorders, 150, 540-545. Kroenke, K., & Spitzer, R. L., 2002. The PHQ-9: a new depression diagnostic and severity measure. Psychiatr Ann, 32, 1-7. Kroenke, K., Spitzer, R. L., Williams, J. B., & Löwe, B., 2009. An ultra-brief screening scale for anxiety and depression: the PHQ–4. Psychosomatics, 50(6), 613-621. Lemaire, C. M., & Graham, D. P., 2011. Factors associated with suicidal ideation in OEF/OIF veterans. Journal of affective disorders, 130, 231-238. Madsen, T., Karstoft, K. I., Bertelsen, M., & Andersen, S. B., 2014. Postdeployment suicidal ideations and trajectories of posttraumatic stress disorder in Danish soldiers: a 3-year follow-up of the USPER study. The Journal of clinical psychiatry, 75(9), 994-1000. McCullough, M. E., Emmons, R. A., & Tsang, J. A., 2002. The grateful disposition: a conceptual and empirical topography. Journal of personality and social psychology, 82, 112.

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

24

Miranda, R., Ortin, A., Scott, M., & Shaffer, D., 2014. Characteristics of suicidal ideation that predict the transition to future suicide attempts in adolescents. Journal of child psychology and psychiatry, 55, 1288-1296. Montross, L. P., Depp, C., Daly, J., Reichstadt, J., Golshan, S., Moore, D., ... & Jeste, D. V., 2006. Correlates of self-rated successful aging among community-dwelling older adults. The American Journal of Geriatric Psychiatry, 14, 43-51. Nelson, C. B., Abraham, K. M., Walters, H., Pfeiffer, P. N., & Valenstein, M. (2014). Integration of peer support and computer-based CBT for veterans with depression. Computers in Human Behavior, 31, 57-64. Nock, M. K., Borges, G., Bromet, E. J., Alonso, J., Angermeyer, M., Beautrais, A., ... & Williams, D., 2008. Cross-national prevalence and risk factors for suicidal ideation, plans and attempts. The British Journal of Psychiatry, 192, 98-105. Nock, M. K., Deming, C. A., Fullerton, C. S., Gilman, S. E., Goldenberg, M., Kessler, R. C., ... & Ursano, R. J., 2013. Suicide among soldiers: a review of psychosocial risk and protective factors. Psychiatry, 76, 97-125. Nock, M. K., Stein, M. B., Heeringa, S. G., Ursano, R. J., Colpe, L. J., Fullerton, C. S., ... & Kessler, R. C., 2014. Prevalence and correlates of suicidal behavior among soldiers: Results from the Army Study to Assess Risk and Resilience in Servicemembers (Army STARRS). JAMA psychiatry, 71, 514-522. Pietrzak, R. H., & Cook, J. M., 2013. Psychological resilience in older US veterans: results from the national health and resilience in veterans study. Depression and anxiety, 30, 432-443. Pietrzak, R. H., Goldstein, M. B., Malley, J. C., Rivers, A. J., Johnson, D. C., & Southwick, S. M., 2010. Risk and protective factors associated with suicidal ideation in veterans of

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

25

Operations Enduring Freedom and Iraqi Freedom. Journal of Affective Disorders, 123(1), 102-107. Pietrzak, R. H., Russo, A. R., Ling, Q., & Southwick, S. M., 2011. Suicidal ideation in treatmentseeking Veterans of Operations Enduring Freedom and Iraqi Freedom: The role of coping strategies, resilience, and social support. Journal of psychiatric research, 45(6), 720-726. Ramsawh, H. J., Fullerton, C. S., Mash, H. B. H., Ng, T. H. H., Kessler, R. C., Stein, M. B., & Ursano, R. J., 2014. Risk for suicidal behaviors associated with PTSD, depression, and their comorbidity in the US Army. Journal of affective disorders, 161, 116-122. Sareen, J., Cox, B. J., Afifi, T. O., de Graaf, R., Asmundson, G. J., ten Have, M., & Stein, M. B. (2005). Anxiety disorders and risk for suicidal ideation and suicide attempts: a populationbased longitudinal study of adults. Archives of general psychiatry, 62(11), 1249-1257. Scheier, M. F., Carver, C. S., & Bridges, M. W., 1994. Distinguishing optimism from neuroticism (and trait anxiety, self-mastery, and self-esteem): a reevaluation of the Life Orientation Test. Journal of personality and social psychology, 67(6), 1063. Schulenberg, S. E., Schnetzer, L. W., & Buchanan, E. M., 2011. The purpose in life test-short form: Development and psychometric support. Journal of Happiness Studies, 12, 861-876. Scott, K. M., Hwang, I., Chiu, W. T., Kessler, R. C., Sampson, N. A., Angermeyer, M., ... & Nock, M. K., 2010. Chronic physical conditions and their association with first onset of suicidal behavior in the world mental health surveys. Psychosomatic Medicine, 72(7), 712-719. Sheehan, D. V., Lecrubier, Y., Sheehan, K. H., Amorim, P., Janavs, J., Weiller, E., & Lépine, J. P., 1998. The validity of Mini International Neuropsychiatric Interview (MINI). The development and validation of a structured diagnostic interview for DSM-IV and ICD-10. Journal of Clinical Psychiatry, 59(suppl 20), 211-32.

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

26

Sherbourne, C. D., & Stewart, A. L., 1991. The MOS social support survey. Social science & medicine, 32(6), 705-714. Taylor, P.J., Gooding, P.A., Wood, A.M., Johnson, J., & Tarrier, N. (2011). Prospective predictors of suicidality: defeat and entrapment lead to changes in suicidal ideation over time. Suicide & Life-threatening Behavior, 41, 297-306. Thompson, J. M., Zamorski, M. A., Sweet, J., VanTil, L., Sareen, J., Pietrzak, R. H., ... & Pedlar, D., 2014. Roles of physical and mental health in suicidal ideation in Canadian Armed Forces Regular Force veterans. Can J Public Health, 105, e109-e115. U.S. Census Bureau; American Community Survey., 2012. American Community survey 1-year estimates: Sex by age by veteran status for the civilian population 18 years and older. Table B21001; generated using American FactFinder 2012; Retrieved September 27, 2014 from http://factfinder2.census.gov/faces/tableservices/jsf/pages/productview.xhtml?src=bkmk; (27 September 2014) Veterans Health Administration, 2012. 2012 Suicide Data Report. VHA Response and Executive Summary Report. Retrieved from (http://www.va.gov/opa/docs/Response-and-ExecSumSuicide-Data-Report-2012-final.pdf). Wang, Y., Sareen, J., Afifi, T. O., Bolton, S. L., Johnson, E. A., & Bolton, J. M., 2015. A population-based longitudinal study of recent stressful life events as risk factors for suicidal behavior in major depressive disorder. Archives of suicide research, 19, 202-217. Williams, J.M., Crane, C., Barnhofer, T., Van der Does A.J., & Segal, Z.V. (2006). Recurrent of suicidal ideation across depressive episodes. Journal of Affective Disorders, 91, 189-194.

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

27

Wisco, B. E., Marx, B. P., Wolf, E. J., Miller, M. W., Southwick, S. M., & Pietrzak, R. H., 2014. Posttraumatic stress disorder in the US veteran population: results from the National Health and Resilience in Veterans Study. The Journal of clinical psychiatry, 75(12), 1338-1346. Youssef, N. A., Green, K. T., Beckham, J. C., & Elbogen, E. B., 2013. A 3-year longitudinal study examining the effect of resilience on suicidality in veterans. Annals of clinical psychiatry: official journal of the American Academy of Clinical Psychiatrists, 25(1), 59. Zerach, G., Levi-Belz, Y., & Solomon, Z., 2014. Trajectories of suicidal ideation and posttraumatic stress symptoms among former prisoners of war: A 17-year longitudinal study. Journal of psychiatric research, 49, 83-89. Zivin, K., Yosef, M., Miller, E. M., Valenstein, M., Duffy, S., Kales, H. C., ... & Kim, H. M., 2015. Associations between depression and all-cause and cause-specific risk of death: A retrospective cohort study in the Veterans Health Administration. Journal of psychosomatic research, 78, 324-331.

Table 1. Sociodemographic and Potential Risk and Protective Factors Examined in Relation to Suicidal Ideation in U.S. Veterans Sociodemographic characteristics

Risk Correlates Psychiatric Distress Factor

Substance Use History Factor

The following characteristics were examined: sex (dichotomous: male, female), age (continuous), marital status (dichotomous: unmarried, married/living with partner), race (dichotomous: non-Caucasian, Caucasian), education (dichotomous: up to high school diploma, more than high school), income (dichotomous: < $60,000, $60,000 or more), and combat status (dichotomous: no combat exposure , combat). Depressive symptoms - Participant responses on the two depressive symptoms of the PHQ-4 occurring in the past two weeks (PHQ-4; Kroenke, Spitzer, Williams, & Lowe, 2009) PTSD symptoms - Assessed with the Posttraumatic Stress Disorder Checklist (Weathers et al., 1993). This is a 17-item measure that examines the severity of experiencing DSM-IV PTSD symptoms in the past month. Generalized anxiety disorder symptoms - Participant responses on the two generalized anxiety items of the PHQ-4 occurring in the past two weeks (PHQ-4; Kroenke et al., 2009). Nicotine dependence – Assessed with the Fagerström Test for Nicotine Dependence (Heatherton, 1991). This measure inquires about lifetime smoking status and then a series of follow up questions about smoking habits. e.g., “Did you smoke if you were so ill that you were in bed most of the day?”

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

Physical Health Difficulties Factor

Lifetime Suicide Attempt (assessed at Wave 1) Protective Correlates Protective Psychosocial Characteristics Factor

Social Connectedness Factor

Active lifestyle Factor

28

Lifetime alcohol abuse and dependence – Assessed with the MINI Neuropsychiatric Interview (Sheehan et al., 1998). This structured measure assesses symptoms of alcohol use disorder based on DSM-IV diagnostic criteria. Lifetime substance abuse and dependence – Assessed with the MINI Neuropsychiatric Interview (Sheehan et al., 1998). This structured measure assesses symptoms of substance use disorder based on DSM-IV diagnostic criteria. Number of physical health conditions - Aggregate number of medical conditions as assessed by self-reported questions about the lifetime presence of physician or other health professional-diagnosed medical conditions.. The following conditions were included: arthritis, asthma, cancer, chronic pain, liver disease, diabetes, heart attack, heart disease, high cholesterol, high blood pressure, kidney disease, sleep disorder, migraine, multiple sclerosis, osteoporosis, rheumatoid arthritis, stroke, traumatic brain injury, HIV/AIDs, and other. Sum of disability of activities of daily living (Hardy & Gill, 2004) – e.g., Requiring assistance from another individual to get in and out of a chair or to get dressed. Sum of disability in instrumental activities of daily living (Hardy & Gill, 2004) – E.g., Requiring assistance from another individual to pay bills or in meal preparation. One item: Have you ever tried to kill yourself (no/yes)?

Resilience (Connor-Davidson Resilience Scale, Campbell-Sills & Stein, 2007). This measure consists of 10 items, with higher scores reflecting greater resilience. E.g., “I am able to adapt when changes occur.” 5-point Likert scale: 1=not at all true to 5=true nearly all the time. Purpose in life (Purpose in Life Test-Short Form, Schulenberg et al., 2011). This measure consists of 4 items, with higher scores reflecting greater perceived purpose in life. e.g., “In life, I have:” 7-point Likert scale: 1=no goals or desires to 7=very clear goals and desires. Dispositional gratitude (single-item of the Gratitude Questionnaire, McCullough et al., 2002). This construct was assessed with one item: “I have so much in life to be thankful for.” 7-point Likert scale: 1=strongly disagree to 7=strongly agree. Perceived level of community integration – This construct was assessed with one item: “I feel well integrated in my community (e.g., regularly participate in community activities).” 7-point Likert scale: 1= disagree to 7=strongly agree. Dispositional optimism – This construct was assessed with one item from the Life Orientation Test-Revised (Scheier et al., 1994): “In uncertain times, I usually expect the best.” 7-point Likert scale: 1 = strongly disagree to 7 =strongly agree. Curiosity/exploration – This construct was assessed with one item from the Curiosity and Exploration Inventory-II (Kashdan et al., 2009): “I frequently find myself looking for new opportunities to grow as a person (e.g., information, people, resources)”. 7-point Likert scale: 1 = strongly disagree to 7 = strongly agree. Number of close friends/relatives - Response to question: “How many close friends and relatives do you have? People you feel at ease with and can talk to about what is on your mind”. Participant reports a numeric answer to this question. Secure attachment style - Question assessing feeling and attitudes in relationships (Hazan & Shaver, 1990). Secure attachment is the endorsement of response a) from the following options: a) finding it relatively easy to get close to others and being comfortable with depending on others and vice versa; b) finding that others are reluctant to get close; and c) feeling relatively uncomfortable with getting close to others. Perceived social support – Assessed with an abbreviated version of the Medical Outcomes Study Social Support Scale -5 (Sherbourne & Stewart, 1991). This is a 5-item measure where participants are asked how often each type of support is available when needed. E.g., “Someone to confide in or talk to about your problems.” 5-point Likert scale: 1=none of the time to 5=all of the time. Higher scores reflect greater perceived social support. A factor comprised of responses related to the number of days per week typically spent doing the following four activities: sports or exercise, reading, writing, or using the

SUICIDAL IDEATION IN U.S. MILITARY VETERANS computer (Montross et al., 2006).

29

Table 2. Sample characteristics.

Age

No SI Remitted SI SI Onset Chronic SI (n=1,859; (n=64; (n=100; (n=84; 86.3%) 3.8%) 5.0%) 4.9%) 62.27(13.90) 58.80(14.76) 61.14(13.95) 50.90(14.02)

Omnibus Group Difference F(3,1938) =20.87; p < .001 Χ2(3) = 7.66; p = .05 Χ2(3) = 41.86; p < .001 2 Χ (3) = 22.43; p < .001

Sex 1700; 92.1% 58; 92.0% 90; 91.8% 71; 84.0% (n; % male) Marital status 357; 23.2% 27; 41.9% 32; 32.7% 33; 47.4% (% unmarried) Race 272; 21.0% 9; 20.3% 15; 25.5% 24; 41.5% (n; % nonCaucasian) Education 260; 32.2% 17; 48.6% 15; 27.6% 12; 20.0% Χ2(3) = 16.63; p (n; % up to high = .001 school diploma) Income 837; 52.9% 36; 58.1% 52; 59.2% 55; 62.8% Χ2(3) = 5.25; p = (n; % less than .16 $60,000) Combat status 1225; 69.3% 36; 55.4% 66; 62.2% 53; 57.9% Χ2(3) = 12.58; p (n; % non-combat) < .01 2 Mental health 302; 15.3% 31; 45.9% 33; 34.7% 58; 74.7% Χ (3) = 241.92; treatment p < .001 (n; % treatment) History of Suicide 54; 3.6% 14; 28.4% 9; 11.3% 17; 25.3% Χ2(3) = 148.98; Attempt p < .001 (n; % attempt) Note: Means and standard deviations are reported for Age. Unweighted n and weighted percentage are reported for each dichotomous variable.

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

31

Table 3. Results of multinomial logistic regression analysis of Wave 1 variables associated with different presentations of suicidal ideation over the 2-year study period Remitted vs. No SI W ald

p

RRR

Sociodemographic Characteristics 1. 7.8 0 Age 3 <.01 3 4. 14. <.00 0 Education 86 1 5 1. 3 Marital Status .83 .36 6

SI Onset vs. No SI

95% CI

W ald

1.011.06

2.8 7

.09

1.998.26

.14

.71

.702.65

.44

.51

p

RRR

95% CI

1. 0 2

1.001.04

.15

.69

.541.53

.14

.71

.712.03

4.01

.04

.652.00

.00

.98

.581.54

1.55

.21

.9 9 1. 5 5

<.00 1

<.00 1

Race

3.6 1

.06

.4 7

.211.02

.21

.65

.9 1 1. 2 0 1. 1 4

Combat Status

.46

.50

.8 0

.421.52

.06

.82

.9 4

Remitted SI vs. Chronic SI

Chronic SI vs. No SI Wal d

p

RRR

95% CI

Wald

p

RR R

95% CI

1. 0 0

.971.02

6.1 7

.01

1.0 4

1.011.07

.381.92

9.4 8

<.0 1

4.7 2

1.7612.68

1.023.93

.85

.36

.68

.301.54

.492.00

2.7 3

.10

.47

.191.15

.783.07

2.4 5

.12

.52

.231.18

3. 7 2

2.864.85

.42

.52

.91

.681.21

1. 6 4

1.322.03

10. 72

.001

.52

.35.77

1.212.04

17. 58

<.00 1

.48

.34.67

.422.67

4.6 9

.03

.32

.11.90

1.31

. 2 5

1.2 4

.861.79

5.53

. 0 2

.56

.34.91

.00

. 9 8

1.0 0

.671.49

.8 6 2. 0 0

Wave 1 Risk Correlates

Psychiatric Distress

80. 49

<.00 1

3. 3 8

2.594.42

9.7 7

<.01

1. 4 8

1.161.89

.41

.8 5

.581.25

20. 47

<.00 1

1. 4 7

1.251.74

.551.02

.23

.63

1. 0 6

.85-1.31

.34-1.81

Physical Health Difficulties

.67

Substance Use History

3.4 7

.06

.7 5

Suicide attempt

5.7 7

.016

.3 4

.14.82

.34

.56

.7 8

.781.47

18. 83

<.00 1

0. 5 7

4.3 7

.04

.7 5

.73

.9 6

9 4 . 9 9 1 9 . 9 2 1 1 . 6 8 . 0 2

.90

1. 5 7 1. 0 6

.37

.8 6

.631.19

.21

.7 8

.541.14

.17

.7 9

.571.10

.001

Wave 1 Protective Correlates Protective Psychosocial

.17

.68

1. 0 7

Social Connectedness

18. 36

<.00 1

.4 4

.30.64

Active Lifestyle

2.1 4

.14

.7 9

.571.08

.12

.45-.74

.57-.98

.75-1.23

. 8 0 1 . 6 1 1 . 9 0

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

32

Note. SI = suicidal ideation. RRR=relative risk ratio. Table 1 describes how each of the independent variables was operationalize

Highlights

    

Nearly 10% of the sample had current SI at Wave 2; 5% had chronic SI. Medical and psychological co-morbidities increase risk of chronic suicidal ideation. Social connectedness was protective for remitted and onset, but not chronic SI. Bolstering protective factors may help mitigate risk for SI onset in this population. For chronic SI, targeting risk factors, physical and mental health, remains important.

GRAPHICAL ABSTRACT

SUICIDAL IDEATION IN U.S. MILITARY VETERANS

33