Abstract
Deployment to a combat zone is a fundamental mission for most military forces, but prior research suggests that there is a complex and nuanced association between deployment and related risk factors for suicide. Deployment and combat experiences vary greatly among military personnel and can affect a variety of protective and risk factors for suicide. This article offers a critical examination of the association among modern U.S. military deployments, suicide attempts, and death while considering the context of a prominent theory of suicide. Although previous work has demonstrated that deployment is not associated with suicide overall in this population, there is growing evidence that risk may be elevated shortly after deployment, and for some subgroups. Specific aspects of combat exposure, including the experience of killing or witnessing death in combat, may be important contributing factors. An analysis of the literature illustrates that deployment-related risk factors for suicide are complex. The limitations of the literature are discussed, and future directions are suggested.
Keywords
As U.S. military suicide rates rose during Operations Enduring Freedom (OEF), Iraqi Freedom (OIF), and New Dawn (OND), 1 many experts worried about the impact of the wars on the well-being of service members. The timing of the increase was certainly persuasive. Since 2001, 63% of active-duty service members, a total of more than 2 million, have experienced a deployment; more than 40% of these service members have been deployed multiple times (Baiocchi, 2013). During the same time period, suicide rates in the military nearly doubled, rising from a base rate of 10.1 per 100,000 persons per year in 2002 to 19.7 in 2009 (Armed Forces Health Surveillance Center, 2012). Significant resources were committed to understanding and preventing military suicide (Blue Ribbon Work Group on Suicide Prevention in the Veteran Population, 2008; Department of Defense Task Force on the Prevention of Suicide by Members of the Armed Forces, 2010; Shulkin, 2016). What evolved from these new research efforts was a complex body of studies suggesting that the relationship between deployment and suicide is not a simple one. Systematic reviews and meta-analyses have been conducted in an attempt to distill common findings, but complex methodological limitations restrict the conclusions that can be drawn (e.g., C. J. Bryan et al., 2015; Ramchand, Rudavsky, Grant, Tanielian, & Jaycox, 2015). Given service members’ elevated risk for suicide and the likelihood that deployments will continue (Besteman & Savell, 2018), an improved understanding of the potential impacts of warfare on service members is imperative. Thus, the current article has three aims: (a) to provide an overview of the extant literature regarding the impact of deployment and combat on rates of death by suicide and suicide attempts, (b) to draw attention to the challenges associated with defining and measuring deployment and combat variables, and (c) to highlight promising directions for future research. 2 To provide a context for this analysis, we start by describing a leading theory of suicide that is helpful in considering the relationship between combat deployment and suicide.
Theoretical Approach to Understanding Suicide
Although there are several theoretical models for suicide (e.g., three-step theory—Klonsky & May, 2015; integrated motivational-volitional model of suicide—O’Connor, 2011), we focus on the interpersonal theory of suicide (ITS; Van Orden et al., 2010) because it is one of the leading empirically supported models in military and veteran populations (Monteith, Menefee, Pettit, Leopoulos, & Vincent, 2013; Selby et al., 2010). This theory posits that the desire for suicide comprises two interpersonal constructs: thwarted belongingness (e.g., feelings of isolation and a sense that one does not belong to a social group) and perceived burdensomeness (e.g., belief that one’s existence places an extreme burden on other people; Van Orden et al., 2010). Individuals arrested for major crimes, for example, may feel very isolated and cut off from their social group. They may also feel that they are an embarrassment to their family and that their death is worth more than their life to loved ones (i.e., they are a burden on others). The theory posits that either of these states alone is not sufficient to create a desire for death. In combination, however, they result in suicidal desire. Studies supporting the ITS have long demonstrated that specific factors associated with thwarted belongingness and perceived burdensomeness, such as social isolation and loneliness, are associated with suicide and suicide attempts (Stravynski & Boyer, 2001; Trout, 1980). In contrast, factors such as marriage and social support are protective, and theorized to decrease these feelings (Kazan, Calear, & Batterham, 2016).
The ITS also proposes that even if suicidal desire is present, an individual is unlikely to make a lethal or near-lethal suicide attempt unless the individual also possesses the capability for suicide (Van Orden et al., 2010), a concept common across recent suicide theories (e.g., Klonsky & May, 2015). In the ITS, the capability for suicide is hypothesized to originate from elevated pain tolerance and fearlessness regarding death. The theory states that this capability for suicide is acquired. The human instinct to live is viewed as a powerful force that is decreased over time by repeated exposure to injuries, painful experiences, or provocative events that expose (and therefore habituate) an individual to the pain and fear associated with enacting lethal behavior. Through the lens of the ITS, repeated experience with deliberate self-harm, for example, creates repeated exposures to pain that decrease the fear of injuries, pain, and death (Hamza, Stewart, & Willoughby, 2012; Van Orden et al., 2010).
A significant body of literature supports the ITS. A recent meta-analysis of 122 studies found that thwarted belongingness and perceived burdensomeness were significantly associated with suicidal ideation, and that thwarted belongingness, perceived burdensomeness, and capability for suicide had a significant interaction effect on the number of prior suicide attempts (Chu et al., 2017).
Research has also begun to explore how the ITS may specifically relate to military suicide. The possible effects of deployment and combat on the development of the capability for suicide are intuitive; the ITS predicts that exposure to violent combat experiences will increase this acquired capability. Indeed, research has found elevated levels of capability for suicide in multiple samples of service members (Brenner et al., 2008; C. J. Bryan, Morrow, Anestis, & Joiner, 2010; Monteith et al., 2013). A possible explanation is that military training may increase practical capability for suicide through increased familiarity with and availability of firearms, which have been linked to increased rates of death by suicide (Brent & Bridge, 2003; Kellermann et al., 1992); however, this possibility has received only limited research attention (e.g., Ursano et al., 2016). Additionally, the experience of combat exposure that included injury, death, and violence has been linked to acquired capability for suicide (C. J. Bryan & Cukrowicz, 2011). However, a separate study found that this capability did not increase over the course of deployment, as one would expect (C. J. Bryan, Sinclair, & Heron, 2016). This finding, coupled with research with adolescent male twins indicating a strong genetic contribution to the development of the capability for suicide, has led some scholars to question whether this capability is acquired or is somehow innate (Smith et al., 2012). As it stands, the complex, often contradictory, empirical findings raise important questions about what aspects of the military experience may elevate suicide risk. In this article, we offer a critical examination of the empirical literature to better characterize the multifaceted associations between deployment and suicide, and we use the central tenets of the ITS as a reference for potentially explaining variations in reported relationships.
Deployment
Although the results are mixed, there is recent consensus that an association between deployment and suicide does not provide a simple explanation for the large rise in U.S. military suicide rates. This is surprising given that within the ITS framework, exposure to the experiences of deployment (e.g., threat of harm, weapons training) is anticipated to increase acquired capability for suicide and, thus, suicide risk (e.g., Selby et al., 2010). However, the lack of a relationship between deployment status and suicide has been seen in service members from multiple wars, including the Vietnam and Persian Gulf wars (Kang & Bullman, 2009; Michalek, Ketchum, & Akhtar, 1998). Most research among OEF and OIF personnel has also failed to show an association between suicide and deployment (both when deployment is defined as any deployment and when it is treated as the number of deployments). For example, a large prospective longitudinal study of service members (more than 700,000 person years) showed no association between suicide and the number of deployments or the total number of days of deployment (LeardMan et al., 2013). A retrospective population-based cohort study that included all suicides from the start of OEF through 2009 found no relationship between suicide and any deployment to OEF or OIF and no relationship between suicide and multiple deployments, even after including individuals who separated from the military (e.g., because of postdeployment mental-health problems; Reger et al., 2015). Similar results were obtained in other studies with OEF and OIF service members (Griffith & Bryan, 2017; Kang et al., 2015; Phillips, LeardMann, Vyas, Crum-Cianflone, & White, 2017).
Studies have also examined whether the association between OEF or OIF deployment and suicide may change over time. Although their study also found no overall association between deployment and suicide, Shen, Cunha, and Williams (2016) found that OEF and OIF service members had elevated suicide rates for the first 4 years after their deployment compared with service members who had not been deployed. Further, results differed across the military services. Suicide rates following deployment were not elevated in the Marine Corps and were elevated for only 9 months (from Year 3 to Year 3.75) in the Navy. The study also found a decreased risk of suicide during deployment. Similarly, Griffith and Bryan (2017) found similar suicide distributions of suicides over time among Army National Guard members who had and had not been deployed. For service members who were deployed, most suicides occurred 1 year or more following a deployment. However, a study with contrasting findings found a higher suicide rate in service members who were currently deployed (20.0 per 100,000) or previously deployed (20.5 per 100,000) compared with those who had never been deployed (15.6 per 100,000; Schoenbaum et al., 2014).
Overall, the findings indicate that the effects of deployment on suicide risk may differ among the service branches, their reserve and National Guard components, and over time. Some of the findings are counterintuitive. For example, service members in the Army and Marine Corps experienced more ground combat than did most of those in the Navy and Air Force. Therefore, the finding that the Marine Corps did not have elevated rates of suicide following deployment but the Navy did is challenging to understand; it raises the important question of whether deployment and combat exposure are the same thing. We explore this question in detail later in this article. It is possible, however, that some of the differences between the military services and components relate to the postdeployment environment. For example, Reservists and National Guard personnel rapidly demobilize and reintegrate into civilian life after a deployment. In contrast, Active personnel remain in an active-duty status with their unit. Therefore, their social support and postdeployment adjustment likely differ significantly. Research is needed to examine how the postdeployment environment may explain some of the differences in suicide risk reported.
Research with OEF and OIF service members has also elucidated how the relationship between deployment and suicide attempts may shift over time. A study of enlisted U.S. Army personnel who served from 2004 to 2009 found that the majority of suicide attempts were made by those who had never been deployed (61.1%); those who had been previously deployed made 29.2% of attempts, and those who were currently deployed had the lowest percentage (9.7%; Ursano et al., 2016). A study of Canadian military forces obtained similar findings (Sareen et al., 2017). Timing of first suicide attempt also appears to vary by deployment status. The study of enlisted U.S. Army personnel found that those who had never been deployed had the highest rates of attempted suicide in their second month of service, those who had previously been deployed had their highest rates in the 5th month following deployment, and those who were currently deployed had their highest risk in their 6th month of deployment (Ursano et al., 2016). Therefore, the risk-assessment process may benefit when the deployment context and timing, in addition to traditional mental-health factors, stressors, and history of suicide behaviors, are taken into consideration.
Overall, these studies detail a complex relationship between deployment and suicide behaviors. Most research studies to date have not found an overall association between deployment and death by suicide. A recent meta-analysis also found no significant relationship (C. J. Bryan et al., 2015). However, as we described earlier, the association between suicide and deployment may vary across subgroups and may fluctuate over time since deployment. Research is also starting to examine factors that may influence the association between deployment and suicide. For example, in one study, attempted suicide by firearm, a suicide method estimated to be lethal 82.5% to 92% of the time (Card, 1974; Spicer & Miller, 2000), was higher in currently deployed and previously deployed service members than in those who had never been deployed (Ursano et al., 2016). Additionally, associations between suicide and deployment differ by Army occupation; higher suicide rates were found in never deployed and previously deployed infantry and combat engineers compared with currently deployed service members (Kessler et al., 2015). Given that infantry and combat engineers are more likely to gain training and experience with weaponry, regardless of whether they have been deployed, it may be that training and familiarity with lethal weapons heightens acquired capability for suicide or the likelihood of utilizing firearms in a suicide attempt. The differences in suicide rates between service branches and components (e.g., Griffith & Bryan, 2017; Shen et al., 2016) may reflect important differences in deployment or training experiences that are worthy of exploration.
One likely contributor to the varied findings regarding the association between deployment and suicide is the difficulty defining and measuring deployment. Laypeople, leadership, and even some researchers believe that research on military “deployment” is a direct examination of the effects of combat. However, that is not an accurate assumption. A deployment means that a service member has moved to support a military mission in an operational area. The precise way this definition is operationalized in research studies is important. Consider the following questions: How long must one be deployed for the assignment to count as a deployment? How much time must there be between two movements for them to count as two deployments? What geographic locations count in identifying deployments for a given war? For Navy deployments, what water locations count (or do any)? These questions, and others, demonstrate the wide variability that is possible in definitions of “deployment” and resultant research findings. Therefore, research results based on any one definition may have fairly limited generalizability. To further illustrate this point, we present two case examples.
Case 1—John
I was deployed to Iraq from 2003 to 2004. Information about how long I’d be there was always changing. During the combat operation, I was often stationed far from any military installations with only a handful of other soldiers. I usually slept in the sand and got dozens of bug bites each night. I experienced frequent sandstorms with 3 days of sustained winds at 40 to 60 miles an hour with nothing but a poncho. During the combat portion, we drank scalding hot water from black rubber bags that were dropped from the air to us. Communication home was never better than severely limited, and mail was always the best source of information. I won’t describe the combat I saw, but you can imagine. Soldiers around me were dying. The best phrase I can think of to describe the whole thing is “mind-numbing,” but that is inadequate.
Case 2—Robert
I was deployed to Iraq in 2005 for a year. I was assigned to a large air base with thousands of other soldiers. The area faced occasional mortar attacks, but these rarely hit anything. I did not worry too much about combat unless I was on the road, which was rare. I slept in something we called a hooch, where each room had a portable air conditioner installed. The food was better than what I ate at home. Fridays were usually steak nights, and every dinner had three choices of Baskin & Robbins ice cream. My unit worked hard when on a mission, but on most evenings, I had time to head to the gym to work out. When I wasn’t working out, I went to the Morale, Welfare and Recreation tent and played video games, or talked to my family. I would have volunteered to deploy again if it weren’t for the separation from my family.
These two examples have been edited and include some experiences common among service members to ensure anonymity, but they accurately document the facts. The comparison between the two deployments is striking; many factors can play a significant role in how a deployment is experienced. The variability evident in these examples calls into question how well deployment can be expected to serve as a risk factor for suicide among people who served during OEF, OIF, or OND.
From a theoretical perspective, it is clear that a deployment can be associated with diverse and opposing influences on thwarted belongingness, perceived burdensomeness, and acquired capability for suicide. Lifelong friendships are often developed with unit members on a deployment (Hinojosa & Hinojosa, 2011), and these likely improve a sense of belonging. However, service members are also separated from friends, family, and other psychosocial support; marriages sometimes suffer as couples struggle both during deployment and after reunion (Meadows, Tanielian, & Karney, 2016). Service members earn more pay on deployment to support their families, which can decrease their sense of burdensomeness, but they may also feel “useless” when it comes to home-front problems (e.g., children’s problems at school) and regret imposing problems and new roles on loved ones. As the case examples illustrate, some service members experience oppressive environmental conditions on deployment (which might be expected to increase acquired capability for suicide), whereas others do not. From a theoretical perspective, the simple number or length of deployments is unable to act as a proxy for the range of deployment experiences that may relate to important theoretical constructs for suicide. Therefore, additional research is needed to examine specific deployment experiences and their association with important theoretical constructs.
In addition, the use of a single, unified deployment construct may have methodological limitations related to factors outside deployment itself. For example, it is possible that deployed service members are healthier than the nondeployed (e.g., they do not have physical or mental conditions that make them unfit to deploy). It may be possible to reduce the impact of these biases by carefully defining eligibility for deployment and removing people who could not have been deployed from analysis altogether (Kessler et al., 2015). However, it is important that deployment not be considered a proxy for combat exposure, because combat exposure varies widely across deployed service members. Combat infantry units reported high levels of combat after returning from Iraq and Afghanistan in 2003, early in the wars (Hoge et al., 2004), whereas many U.S. service members experienced minimal or no combat exposure in those countries, especially during certain phases of the wars. Given the important distinction between deployment and combat exposure, we next consider the association between combat exposure and suicide behaviors.
Combat Exposure
One way in which suicide risk may be conferred is through combat exposure. At the most basic level, combat exposure implies experiencing or witnessing engagement with a military combatant, or exposure to associated experiences (e.g., an improvised explosive device, medical treatment of the wounded). To date, only a few studies have examined whether combat exposure is associated with suicide. Farberow, Kang, and Bullman (1990) found that Vietnam veterans who died by suicide were not more likely to have had combat experience or to have served in a combat-related position in comparison with Vietnam veterans who died by car accident. LeardMann et al. (2013) obtained similar findings in a prospective longitudinal study of OEF and OIF personnel; broadly defined, experience of combat was not related to eventual death by suicide. The recent meta-analysis by C. J. Bryan and colleagues (2015) was unable to shed light on this finding as the researchers were able to identify only one study that investigated the relationship between combat experience and suicide, and found no studies that examined the association between specific combat experiences and suicide.
Data from multiple war eras are available for investigating the relationship between the experience of any combat and suicide attempts; individual studies have obtained very different results. Studies investigating this relationship in Vietnam veterans have been mixed; one study found a small positive relationship (Fontana, Rosenheck, & Brett, 1992), whereas another study found no relationship (Maguen et al., 2012). Research on this association in OEF and OIF service members is also mixed. C. J. Bryan, McNaughton-Cassill, and Osman (2013) found a small, positive relationship between the intensity of combat and history of suicide attempts in active-duty Air Force Security Forces personnel. Similarly, a study of Canadian military forces found a small, positive association between the number of deployment-related traumatic events (e.g., combat, witnessing death) and suicide attempts, although this association was null after accounting for previous trauma and co-occurring mental-health disorders (Sareen et al., 2017). On the other hand, C. J. Bryan, Ray-Sannerud, Morrow, and Etienne (2013) found a small negative relationship in a sample of U.S. Air Force personnel seeking outpatient mental-health care. In short, research has offered mixed results regarding the association between combat and suicide attempts.
The mixed results regarding combat experience may be due in part to the diversity of combat experiences, a possibility that points to the need to better assess specific combat experiences that may influence suicidality, rather than to assess combat exposure as a unified construct. As is true for deployment, exposure to combat is not a uniform experience. For instance, in the case examples, John reported the experience of soldiers dying around him as “mind-numbing,” whereas Robert indicated that combat was “rare” and something he did not worry about. Accurately defining combat exposure is a challenge given that exposure can vary in several distinct ways. First, exposure can differ according to how actively involved an individual is in combat; at one extreme are experiences of actively killing someone, and at the other extreme are experiences of being under threat of attack but having no direct enemy contact and exposure to the aftermath of battle (e.g., human remains; Van Winkle & Safer, 2011). Second, combat exposure can vary according to distance from the combat event. For instance, individuals who pilot Predator or Reaper remotely piloted aircrafts, commonly referred to as drones, report higher emotional distress than noncombatant service members, but less than those deployed directly to combat zones (Chappelle et al., 2014; Chappelle, McDonald, Thompson, & Swearengen, 2012). Third, the target of combat can have important ramifications for service members. In combat, they are sometimes forced to engage with civilians (e.g., women, children), an experience that has been related to later guilt and suicide attempts (Hendin & Haas, 1991). Finally, the length of exposure can vary substantially, ranging from a brief one-time exposure to prolonged exposure to a set of experiences (LeardMann et al., 2013). It is important to consider the diversity of combat experiences in detailing the relationship between combat exposure and suicide. In the rest of this section, we review research in which the diversity of combat experiences has been taken into account.
Experience of killing or of death in combat
There is some support for the idea that suicide behaviors are associated with killing and exposure to death and injury in combat. A meta-analysis in which these types of experiences were aggregated found a positive effect size (r = .11) indicating that a history of attempted suicide had a small relationship with killing during combat or seeing people who were wounded or died in combat (C. J. Bryan et al., 2015).
To date, no studies have assessed the independent impact of killing in combat on death by suicide, but studies have begun to explore associations between killing in combat and suicide attempts. A study on members of the Army National Guard found a small association between killing or witnessing killing and suicide attempts (Griffith & Vaitkus, 2012). One study of Vietnam veterans also showed a small relationship between killing during combat and history of suicide attempts (Fontana et al., 1992), whereas another showed that killing during combat had a positive association with suicide ideation, but not attempts (Maguen et al., 2012). Notably, the nonsignificant findings of Maguen and colleagues may have been the result of limited power given their relatively small sample size (260 veterans, 12 of whom had attempted suicide) compared with the studies that supported an association, with samples of more than 5,000 veterans. An additional study found no correlation between killing-related combat experiences and history of suicide attempts in previously deployed, treatment-seeking, OEF and OIF active-duty service members who were diagnosed with posttraumatic stress disorder (PTSD; McLean et al., 2017). Given the restricted sample, however, this finding may not generalize to the broader population of OEF and OIF service members. Overall, it appears that there is a positive association between killing and suicide attempts, but it is notable that the effect sizes are small, an indication of the resilience of service members even after experiencing intense combat.
Although exposure to death in combat has been understudied as a predictor of suicide, research has supported a modest relationship between such experiences and suicide risk factors (C. J. Bryan & Cukrowicz, 2011; Fontana & Rosenheck, 1999; Thoresen & Mehlum, 2008), suggesting that killing or witnessing death in combat may be a more robust predictor of suicide compared with exposure to combat or deployment more generally. Overall, such findings are consistent with the ITS framework, which conceptualizes exposure to painful events as contributing to acquired capability for suicide. For instance, John, in Case 1, reported multiple exposures to dying soldiers and defined the experience as “mind-numbing.” Repetitively witnessing killing or death may habituate individuals to the pain and fear associated with enacting lethal behavior, thus increasing their capability for suicide (e.g., Klonsky & May, 2015; Van Orden et al., 2010).
However, firm conclusions regarding the role of killing or witnessing death cannot be reached without research on more specific features of such experiences. Although measures such as the Combat Exposure Scale (Keane et al., 1989) contain items assessing passive observation of violence (“seeing dead bodies or dead body parts”) as opposed to more active participation in violence (“shooting or directing fire at the enemy”), research has generally not examined how the association between suicide risk and killing or witnessing death in combat might vary with an individual’s specific role in killing or who the specific decedents were. It is reasonable to hypothesize that a service member who killed another friendly service member or child may be more susceptible to guilt, shame, and suicidal ideation compared with an individual who killed a clear enemy combatant. Additionally, service members who have been exposed to dead bodies of innocent victims (e.g., women and children) may be more susceptible to mental-health problems than individuals who have not. These aspects of combat experience will be important to incorporate in future research efforts.
Combat injury
One additional way to extend the understanding of the relationship between combat exposure and suicide is by utilizing other related objective data. For example, combat injuries are documented evidence of combat exposure, and, given the health-care and personal-adjustment challenges they can face, wounded service members constitute a population of interest. Several studies suggest that injury may have an important association with suicide. In one study, OEF and OIF service members who were injured during combat had higher crude suicide rates than those who had been deployed but not injured, although after adjustment for demographics, the findings were no longer statistically significant (Reger et al., 2017). Notably, many of the injuries included in this study were presumed to be minor. A study of Vietnam veterans indicated that the extent of injuries may play a role (Bullman & Kang, 1996). Vietnam veterans who were injured more than once and hospitalized at least one time for a combat injury had an increased risk for suicide compared with those who were wounded but not hospitalized. However, veterans who were wounded just once during combat and hospitalized for their wound were not at increased risk of suicide compared with those who were wounded but not hospitalized. To the best of our knowledge, no empirical study has determined whether injury sustained during combat may confer risk for suicide attempts. However, there is some evidence of a relationship between history of traumatic brain injury and suicide attempts in veterans (e.g., Brenner, Ignacio, & Blow, 2011; C. J. Bryan & Clemans, 2013), although few of these studies specified whether the injury resulted from combat (Adams, Corrigan, & Larson, 2012), and the association may have been due in large part to co-occurring psychiatric conditions, such as PTSD (Fonda et al., 2017).
Because this body of research is relatively immature, theoretical evidence can be helpful in developing hypotheses (e.g., C. J. Bryan, Morrow, Anestis, & Joiner, 2010). For instance, an injury (and subsequent medical treatments) may increase feelings of being a burden on loved ones or other military personnel, and consequently lead to increased perceived burdensomeness as conceptualized by the ITS. Indeed, civilians with physical disabilities report elevated levels of perceived burdensomeness (Khazem, Jahn, Cukrowicz, & Anestis, 2015). Disabilities can also affect social support and isolation, thereby increasing a sense of thwarted belongingness. Thus, we expect future research to find that combat injuries are associated with suicide behaviors.
Conclusions and Future Directions
OEF, OIF, and OND sparked a new generation of research examining the relationships between war and suicide. Our Case Examples set the stage for a discussion about suicide, deployment, and combat, highlighting the diversity of service members’ experiences during deployments. Although limited, the literature to date has provided valuable information about suicide’s relationship with deployment and combat. Considering the research as a whole can offer important insights into this area of national concern. First, the increase in deployments since 2001 does not provide a simple explanation for the rise in military suicide rates. Even when relationships between deployment or combat and suicide have been reported, the effect sizes observed have generally been modest (C. J. Bryan et al., 2015), and suicide remains an extraordinarily rare event statistically. It can be a disservice to military members and veterans to assume that anyone who has been deployed must be at high risk for suicide. The stereotype that postdeployed military members are at high risk for suicide may actually increase their risk because such stereotypes can affect employment opportunities and new social relationships. Further, sensational and often misleading media coverage of military suicide (e.g., Price, 2015) may directly increase suicide rates through contagion effects (Stack, 2003). It may also increase the stigmatization of suicidal individuals, which in turn may increase their risk of suicide (Niederkrotenthaler, Reidenberg, Till, & Gould, 2014; Oexle et al., 2017).
Second, the mixed findings regarding suicide’s connection with deployment and combat experiences highlight the variability in what may precipitate suicide. The research we have summarized examined risk among large cohorts in the hope of identifying common indications of suicide risk that can be used to improve prevention for large groups. However, clinical risk assessments must continue to consider the impact of deployment and other stressors on a case-by-case basis. There are well-known examples in which combat and deployment appear to have played important roles in the pathway that led ultimately to suicide (e.g., Somers, 2013). Thus, although deployment may not be a primary predictor of death by suicide or of suicide attempts in general, treatment for postdeployment adjustment problems and combat-related PTSD remains an important consideration.
Overall, the literature indicates that combat exposure is a better predictor of suicide behaviors than is deployment more generally. However, a focus on combat exposure has shortfalls similar to those of a focus on deployment, including definitional difficulties and minimal empirical research. The studies we have reviewed do reveal two main points: First, the experience of killing or witnessing death in combat may have stronger support as a risk factor for attempted suicide compared with other aspects of deployment (general combat exposure or history of deployment), and second, more research is needed, especially studies that consider predictors of suicide mortality. Studies to date offer a foundational understanding of how deployment factors may be associated with suicide, enabling more targeted research in the future. Next, we outline several potentially fruitful foci of future research.
Improving measurement
The large, high-quality studies that have been conducted to date have increased our appreciation for the limitations in the field, including the challenge of defining variables, methodological problems, and the need to expand the range of constructs examined. It will be necessary for future research to move beyond the current approach to data collection, which largely relies on the collection of self-report data at the time the data are needed. Not only are such self-reports limited for a variety of methodological reasons (e.g., recall problems, mental-health disorders, subjectivity of the scale; Sternke, 2011), but self-reported suicide-related behaviors and thoughts are believed to be underreported (Anestis, Mohn, Dorminey, & Green, 2017). Additionally, researchers are often unable to validate these data, so the data may not be reliable (e.g., Garvey Wilson, Hoge, McGurk, Thomas, & Castro, 2010; Polusny et al., 2011; Wessely et al., 2003). The retrospective nature of these data likely contributes to the mixed findings in the literature given that suicide risk and its predictors likely fluctuate over time (Griffith & Bryan, 2017; Shen et al., 2016; Walsh, Ribeiro, & Franklin, 2017). Potentially critical information regarding predictors and periods of high risk may be obscured when data collection is several months or years removed from deployment or a suicide attempt. Prospective research on suicide behaviors, in which combat-exposure data are collected shortly after combat, would be challenging and require a large sample. However, the current priority afforded suicide prevention on a national level may make such methodological improvements feasible. Because suicide risk can oscillate over time, longitudinal studies could track suicide behaviors over the months and years following combat.
Investigating individual responses to combat
In future research, it will be important to consider that an individual’s personal response to combat exposure may be key to suicide risk. Research indicates that feelings of shame and guilt are important predictors of suicidal ideation in military personnel, particularly those who have experienced combat (C. J. Bryan, Ray-Sannerud, et al., 2013). Additionally, moral injury or moral guilt (i.e., acting against one’s deeply held moral beliefs, such as harming a child in an act of war) has been shown to be a significant predictor of PTSD, suicidal ideation, and suicidal behavior in military personnel (A. O. Bryan, Bryan, Morrow, Etienne, & Ray-Sannerud, 2014). It is reasonable to hypothesize that service members who witness or are the agents of violence and killing and who respond to these acts with internalized feelings of guilt, shame, and moral injury are at higher risk for emotional distress and suicidality compared with those who do not have internalized reactions.
Exploring differences among service branches
Available studies suggest that results differ by branch and component (active duty vs. National Guard or Reserve) of the military service (Griffith & Bryan, 2017; Shen et al., 2016). However, there is currently little explanation for this variation, in part because branches and components differ in a range of ways, from core values to postdeployment reintegration support. It is difficult to accurately identity what factors may be driving the differences between branches and components. For instance, we are not aware of research that describes how combat experiences differ by branch or by active-duty status. However, broad military statistics do support the idea that members of the Army and Marines are more likely to be deployed, wounded, and killed than are members of other branches (DeBruyne, 2017), which indicates it is likely that these two branches have a higher frequency or intensity of combat. Similarly, National Guard and Reserve forces made up one third of deployed service members in OEF, but constituted only 14% of those wounded or killed (DeBruyne, 2017). Thus, it appears that the Active component may experience a higher frequency (or intensity) of combat than the Guard or Reserves.
Alternatively, it is possible that differences in suicide rates between services and components are due to differences in demographics, social support, deployment length, or training. In order to better understand suicide risk in military populations, it is important to examine the unique experiences of service members across branches and components. Additionally, it should be noted that the majority of the studies we have reviewed focused on the U.S. military. Future research would benefit from assessing the generalizability of important findings to military personnel from other countries.
Investigating the role of social support
As highlighted in the case studies, service members experience unique constellations of interpersonal stressors while deployed. John’s deployment included infrequent communication with family and limited opportunities to develop relationships with other unit members outside of their missions. Robert’s communication with his family was much more frequent and reliable, and he had recreation programs that may have helped him develop supportive relationships with his unit. The ITS provides a lens through which such experiences can be viewed. For example, the simple separation from loved ones during deployment may play a role in the development of thwarted belongingness and perceived burdensomeness (and thus the desire for suicide). Maintaining a meaningful relationship during deployment is a challenge (Carter & Renshaw, 2016), and interpersonal distress can, in turn, exacerbate suicide risk (Kazan et al., 2016). On the other hand, social support and interpersonal connection during deployment may serve as protective factors; in theory, they could decrease feelings of thwarted belongingness. In line with this reasoning, studies have shown that postdeployment mental health tends to be better among service members who communicate with a romantic partner or have strong unit cohesion during a deployment compared with those who do not experience such forms of social support (Carter et al., 2011; Mitchell, Gallaway, Millikan, & Bell, 2012). These forms of connection, like the experience of combat, likely vary greatly from deployment to deployment. The relations between levels and types of social support during deployment and subsequent suicide risk are worthy of future exploration.
Recommendations
As research on the association between military deployment and suicide is sparse, and findings are mixed, recommendations are difficult at this stage. However, policymakers and military leaders can be instrumental in ensuring that data are available for future research needs. Military data-collection efforts, and initiatives for screening service members can be planned to ensure that a rich source of deployment and combat data is available in the future. It is not clear that suicide-prevention initiatives should focus on deployed service members more than others (at least in the United States), but there are theoretical reasons to think that some aspects of combat exposure may prove important in the future, and results of early studies lean in that direction. Finally, it should be emphasized that research demonstrates that suicide is preventable (Mann et al., 2005). The nation must do more to improve prevention efforts, especially for service members and veterans (Reger, Pruitt, & Smolenski, 2018). Clinicians and researchers have the opportunity and responsibility to contribute to prevention efforts for this high-need, high-risk population.
Footnotes
Acknowledgements
The contents of this article do not represent the views of the U.S. Department of Veterans Affairs or the U.S. Government. Portions of this article were presented at the NATO Military Suicide Prevention Symposium, Riga, Latvia, April 3, 2017.
Action Editor
Brad J. Bushman served as action editor and June Gruber served as interim editor-in-chief for this article.
Declaration of Conflicting Interests
The author(s) declared that there were no conflicts of interest with respect to the authorship or the publication of this article.