Military-to-civilian transition strains and risky behavior among post-9/11 veterans

ABSTRACT

Many military veterans face significant challenges in civilian reintegration that can lead to troublesome behavior. Drawing on military transition theory (MTT) and using data from a survey of post-9/11 veterans in two metropolitan areas (n = 783), we investigate previously unexamined relationships between post-discharge strains, resentment, depression, and risky behavior, taking into account a set of control variables, including combat exposure. Results indicated that unmet needs at time of discharge and perceived loss of military identity are associated with increased risky behavior. Much of the effects of unmet discharge needs and loss of military identity are mediated by depression and resentment toward civilians. The results of the study are consistent with insights from MTT, providing evidence of specific ways in which transitions can affect behavioral outcomes. Moreover, the findings highlight the importance of helping veterans meet their post-discharge needs and adapt to changing identity, in order to reduce the risk of emotional and behavioral problems.

What is the public significance of this article?—Our study suggests the need for targeted and coordinated services that not only help veterans meet needs such as housing, employment, and medical care upon discharge, but also help prepare them for the challenges associated with a significant change in identity as part of civilian life.

There are over 4 million post-9/11 veterans in the United States (U.S. Department of Defense, 2018; U.S. Department of Veterans Affairs, 2016). While most successfully reintegrate into civilian life after their service, a substantial portion experience adjustment difficulties, including mental health and substance abuse problems, and involvement with the criminal justice system (Castro & Kintzle, 2018; Hoge et al., 2004; Keeling et al., 2018; McGuire, 2007; Tanielian et al., 2008). About half of recently discharged veterans indicate their adjustment to civilian life was “very” or “somewhat difficult” (Parker et al., 2019). Significant portions of veterans from Operation Iraqi Freedom (OIF) and Operation Enduring Freedom (OEF) meet the diagnostic criteria for post-traumatic stress disorder (PTSD) or depression (Hoge et al., 2004; Ramchand et al., 2015). Mental health problems in particular, put veterans at increased risk for unlawful behavior (Tanielian & Jaycox, 2008). Although not disproportional, about 8% of all persons in jail or prison are veterans (Bronson et al., 2015).

Much of the contemporary study of problematic behavior among military veterans focuses on how trauma exposure leads to mental health problems, such as PTSD, increasing the likelihood of violence toward intimates (e.g., Jones, 2012; Marshall et al., 2005; Sullivan & Elbogen, 2014; Taft et al., 2005). Comparatively, less attention has been paid to a range of other issues that veterans confront after their military service (Castro & Kintzle, 2018). Veterans face unique challenges, such as the ease of transition into civilian roles, especially in terms of finding jobs, housing, and meeting health-care needs. In addition, existing qualitative research suggests the loss of identity cultivated through intensive socialization processes and a highly structured lifestyle can be a source of dislocation and distress for many veterans (Smith & True, 2014; Yanos, 2004). However, there has been limited quantitative research examining how identity loss affects behavioral outcomes. Moreover, public ambivalence about the value of military efforts in the post-9/11 era has led many veterans to believe their sacrifices are underappreciated and not well understood, with adverse consequences for well-being (Markowitz et al., 2020).

At the same time, beyond intimate violence, comparatively less attention has been paid to other forms of harmful behavior among veterans, including nonviolent offenses and risk-taking (Killgore et al., 2008). There are compelling reasons for examining outcomes other than intimate partner violence among military veterans. Evidence suggests there is substantial overlap between criminal violence (e.g., spousal assault), nonviolent criminal behavior (e.g., driving under the influence), and other non-criminal, but potentially harmful behaviors (e.g., gambling, substance abuse; Deane et al., 2005). Moreover, minor antisocial behavior is much more common than serious violence, can damage health, and can be criminogenic – for example, drinking increases the risk of violence (Deane et al., 2005; Felson, 2009). In addition, a growing number of veterans are involved in civil legal proceedings, such as divorce, child-custody disputes, and debt collections (Seamone & Albright, 2015). Given the above considerations, we draw on military transition theory to examine how unmet discharge needs and military-identity loss are related to depression and resentment, which in turn, increase risky behavior among post-9/11 veterans.

Reintegrative strains and risky behavior

Military transition theory (MTT) is a recently developed framework for understanding how transition may lead to either positive or adverse outcomes by focusing on the quality and success of the military-to-civilian transition (Castro & Kintzle, 2018). According to the theory, transition processes involve changes in relationships, social networks, resources, and identity that can entail both short- and long-term challenges (e.g., having access to healthcare and benefits, changes in sense of self in civilian context; Castro & Dursun, 2019). Successful transition to civilian life is thus a function of several factors, including having a job, access to health care, stable housing, and a sense of belonging. The well-being of service members is conceptualized as a process of social reintegration from a highly structured lifestyle to one where the means of meeting one’s goals must be established or reestablished. While MTT has been used to understand how disconnection among military veterans places them at risk for suicide (Castro & Kintzle, 2014), research has not yet taken advantage of this approach to examine how transition processes are related to other outcomes, such as risky behavior.

Military transition theory (MTT) asserts that transitions are inherently stressful for a variety of reasons. The unique mission of the military requires provision of full employment, housing, medical care, and other needs. Returning service personnel thus face challenges in meeting these needs, while grappling with a shift in sense of “belonging” and purpose. The theory predicts that when veteran reintegration is particularly difficult, these challenges can be sources of strain that have potential adverse consequences for well-being (Castro & Kintzle, 2018; Schonfeld et al., 2015). MTT is consistent with insights from general strain theory, which asserts that strain can result from not only blocked goals (e.g., financial strain from not having a job), but also the loss of positively valued stimuli (e.g., relationship breakup). These sources of strain lead to negative emotions, such as anger, depression, and hostility, which are then expressed in risky and unlawful behaviors (Agnew, 1992; DeCoster & Heimer, 2001; Link et al., 2015).

We focus on two sources of strain associated with civilian reintegration that may lead to mental health and behavioral problems among veterans – unmet discharge needs and perceived loss of military identity. As military transition theory highlights, during military service, needs, such as housing, food, medical care, and income are met. Thus, upon discharge, veterans experience the loss of those things that, for some, can be difficult goals to regain. Consequently, a cumulative number of unmet financial, housing, healthcare, and other needs following discharge has the potential to increase negative emotions and sentiments that lead to troublesome behavior.

Military service also entails development of an important and valued identity (Atuel & Castro, 2019; Kleykamp et al., 2021; Lancaster et al., 2018; Smith & True, 2014; Stets & Burke, 2014). Military identity is a highly salient role identity that develops in military culture – a sense of self that is reinforced and supported by others, in terms of values and behavioral expectations of military personnel, including strong in-group identification (Atuel & Castro, 2018; Castro & Dursun, 2019; Grimell, 2015). Despite the hardships – especially deployment to combat – most military service personnel derive a great deal of pride and satisfaction from their service and endorse it (Igielnik, 2019). Therefore, when service ends, it can lead to an overwhelming loss of sense of self, as the role of “soldier” is a significant part of one’s identity (Smith & True, 2014). Reconciling the shift in identity from “soldier” to “civilian” can lead to feelings of dislocation and depression (Mitchell et al., 2020; Orazem et al., 2017; Smith & True, 2014; Tarbet et al., 2020).

Qualitative research indicates that, while military transitions affect role identity, they may or may not lead to a “crisis.” Rumann and Hamrick (2010) find that, for many veterans, returning to work or college, for example, presents challenges in forming new friendships to replace those formed over time in the military. Qualitative interviews reveal that new friends often do not share an “understanding” of military experience, particularly for those who were deployed to combat. Thus, the loss of existing networks of those with shared understanding can be disorienting. Moreover, transition to civilian life involves adjustment to the loss of status and certain privileges accorded to those in uniform (e.g., discounts and expressions of gratitude). For some, the loss of these things can create resentment toward civilians. The consequences of this variation – subjective in nature – may include feelings of depression and resentment at having been “promised” something (Grimell, 2017a, 2017b). As Atuel and Castro (2019) point out, many of these processes have not been examined in a systematic way. Building on this research, we examine whether the perceived loss of military identity after discharge is associated with depression and resentment toward civilians, which may be expressed in risky behavior.

Unmet needs and loss of military identity may not only lead to depression, but these strains likely entail resentment toward civilians, negative sentiments that may find their expression in risky and unlawful behavior (Bernard, 1990; Umberson et al., 2002). Evidence indicates substantial portions of veterans harbor resentment toward civilians (Igielnik, 2019), yet how post-discharge strains affect resentment and its behavioral consequences have not been examined. Many veterans feel their struggles are misunderstood and their sacrifices are underappreciated (Iraq and Afghanistan Veterans of America, 2019). Veterans constitute a smaller portion of the population compared to earlier eras (U.S. Department of Veterans Affairs, 2016). While the public holds the military as an institution in high regard, public support for engagements in Iraq and Afghanistan was limited, due to the extended length and low efficacy of these conflicts (Oliphant, 2018). Moreover, because of inaccurate perceptions of dangerousness and (often well-intentioned) media focus on veterans’ problems, there is a degree of stigma associated with veteran status (“brave but broken”), resulting in social rejection that can limit prospects for employment and the extent of social network ties (Hipes et al., 2015; MacLean & Kleykamp, 2014).

Recent research also shows that perceived negative public attitudes have adverse consequences for veterans’ mental and social well-being (Markowitz et al., 2020). Likewise, veterans experiencing adjustment difficulties, including the inability to meet certain needs and losing a valued part of their identity, may not only feel more depressed, but this may enhance their feelings of resentment toward civilians (Orazem et al., 2017; Tarbet et al., 2020). Feelings of resentment may be expressed in maladaptive behavior (Agnew, 1992; Link et al., 2015).

In sum, following insights from MTT, military transition-specific strains, including unmet discharge needs and identity loss, can be predicted to increase negative emotions and sentiments, increasing the likelihood of troublesome behavior. Specifically, we hypothesize that veterans who (1) report a greater number of unmet needs at time of discharge and (2) perceive a greater level of military identity loss will report (a) a greater level of depressive symptoms and (b) increased resentment toward civilians, and (c) engaging in more risky behavior. We further predict that the effects of unmet needs and identity loss on risky behavior will be at least partly mediated by depression and resentment. Figure 1 illustrates the relationships that we examine in a sample of post-9/11 veterans. Post-9/11 veterans face unique challenges compared to service members from other eras, including serving for longer time periods, multiple deployments, and many having returned during a post 2008-recession economy (Faberman & Foster, 2013).

Figure 1.

Post-discharge strains, negative sentiments, and risky behavior.

Note: Effects of control variables not shown.

Methods

Sample

The data used come from a cross-sectional survey of veterans in two major metropolitan areas, Chicago and San Francisco. Sampling at both sites used a targeted strategy to recruit from the veteran population in the Chicago metropolitan (“Chicagoland”) and San Francisco Bay areas. Several sampling strategies were used. The first involved agencies that serve veterans, as well as college veteran groups. Respondents were recruited online and in-person. For the online sample, agencies sent out an invitation and survey link to veterans in their databases. For the in-person sample, agencies worked with researchers to organize data collection at group meetings. Respondents were also sampled through a national veteran’s organization e-mail list. The final sampling strategy used print advertisements and social media (e.g., Facebook, Twitter, and LinkedIn) within Chicagoland and the San Francisco Bay area to promote the survey to potential participants. The surveys were administered both online and in person, on paper, and respondents received a $15 gift card. All data collection procedures were IRB approved. Respondents in the study had to have served in the military, but were no longer serving. The median time since discharge was 6 years. After listwise deletion of missing data, a total of 783 post-9/11 veteran cases are used in the analysis.¹

Measures

Military transition theory and focus group assessments identify a set of key needs at the time of discharge that shape the degree of successful transition to civilian life (Castro & Kintzle, 2018). Unmet discharge needs were measured by adding the number of transition needs that were unmet “when I left the military,” and included: not having a job, a place to live, access to health care, access to educational benefits, medical and service records, physical and mental health needs met, legal issues resolved, and financial problems. The number of unmet needs was added to form an index with a range from 0 to 9 (alpha = .82).²

Identity loss is measured using the subscale of the Warrior Identity Scale (Lancaster & Hart, 2015). The four items are as follows: “By leaving the military I lost a family,” “I miss my military friends,” “The most important things that have happened in my life involve the military,” and “I wish I could go back to the military.” Each item is coded from 1 (“strongly disagree”) to 4 (“strongly agree”). In two studies with different samples (Lancaster & Hart, 2015; Lancaster, Castro & Kintzle, 2018), the items showed discriminant validity with other aspects of military identity (e.g., public regard and private regard). Items were summed and divided by the number of items, so that higher values indicate greater identity loss (alpha = .77).

Given our predictions regarding the consequences of post-discharge strains, we developed a measure of resentment toward civilians, based on focus group data where themes regarding veterans’ public perceptions emerged. The measure includes level of agreement with seven items: “Civilians don’t work as hard as veterans” “Civilians don’t appreciate the sacrifices that veterans made for them,” “Civilians are not team players,” “Civilians don’t understand the problems faced by veterans,” “Over the past few years, veterans have gotten less than they deserve,” “People think I’m ‘screwed up’ because of my military service,” and “My military skills and experience are often dismissed.” Each item is coded from 1 (“strongly disagree”) to 5 (“strongly agree”). Items were summed and divided by the number of items, so that higher values indicate greater resentment (alpha = .81).³

Depression is measured using the PHQ-9 (Patient Health Questionnaire), a nine-item scale that measures symptoms in the last two weeks (Kroenke et al., 2001). The items ask respondents to indicate the frequency of symptoms, such as “little interest or pleasure in doing things,” “feeling down, depressed, or hopeless,” “trouble falling or staying asleep, or sleeping too much,” “poor appetite or overeating,” “trouble concentrating,” and “thoughts that you would be better off dead, or of hurting yourself.” Each item was coded 0 = “not at all” 1 = “several days” 2 = “more than half the days” 3 = “nearly every day” and then summed across the items (alpha = .94).

Risky behavior is measured using the scale developed by Adler et al. (2011). The scale is a summed total of whether the respondent engaged in any of the following in the last 12 months (each coded 0 for “no” and 1 for “yes”): “excessive speeding,” “involved in a car accident,” “driven recklessly,” “looked to start a fight,” “carried a weapon outside of work duties,” “took unnecessary risks to life,” “took unnecessary health risks,” “risked getting a sexually transmitted disease,” “drove after several drinks,” “rode with a drunk driver,” “gambled with money you could not afford to lose,” “smoked tobacco,” and “used smokeless tobacco.” The scale can range from 0 to 13 (alpha = .82).

We controlled for a set of demographic, service-related, and other variables that may be associated with degree of successful post-military transition, stress exposure, depressive symptoms, and involvement in risky and illegal behavior (Cesur et al., 2016; Kennedy, 2020; MacLean & Elder, 2007; Mirowsky & Ross, 2003; White et al., 2011). Demographic variables include age (years), race (1 = Black, 0 = non-Black), sex (1 = male, 0 = female), education (1 = “some high school” to 8 = “doctorate”), and whether the respondent was divorced or separated (1 = yes). Service-related variables included number of deployments greater than 30 days, combat exposure (1 = yes, 0 = no), rank, based on the 21-level military pay-grade scale at time of discharge, years of service, and years since discharge. Controlling for combat exposure and number of deployments allows us to isolate the effects of post-discharge strains on outcomes that are also likely to be affected by adverse service-related experiences. Veterans with higher rank advancement may benefit from longer-term security and stability upon discharge.

Analysis plan

All analyses were conducted using Stata 16.1 (StataCorp, 2019). First, we present descriptive statistics for all study variables, noting how our sample compares to known characteristics of post-9/11 veterans. Next, we estimate regression equations to examine relationships between transition strains and depression and resentment. We then estimate the equations for risk-taking with and without the potentially mediating effects of depression and resentment. We then use the “medsem” procedure to estimate the indirect effects of the strain variables on risky behavior to examine the extent to which these effects are statistically mediated by depression and resentment (Zhao et al., 2010). All equations control for the full set of demographic and service-related variables. Given the continuous measures of our dependent variables, we use OLS regression to estimate the equations.⁴

Because unmet needs are referenced to time of separation, whereas identity loss, depression, and risky behavior are measured as more recent, we are estimating how unmet needs at time of discharge are associated with subsequent emotional and behavioral outcomes. MTT holds that transition experiences can affect well-being following discharge and throughout later life. To consider the possibility that there may be stronger (i.e., more “acute”) effects for those discharged more recently, compared to those longer ago, we examine the robustness of our findings in two ways. In supplementary analyses, we examine the sensitivity of our findings to length of time since discharge by examining whether the effect of unmet transition needs depends on time since discharge and whether its effect is similar in a sub-sample of more recently discharged veterans.

Results

Descriptive statistics

Descriptive statistics for all study variables are shown in Table 1. When compared to the national demographic profile of post-9/11 veterans as indicated by U.S. Department of Defense (2018) estimates, our sample is comparable, with some slight differences. Our sample includes a slightly lower proportion of males (77% vs. 82%) and Black veterans (13% vs. 15%). Some of these differences may be a result of our sampling methods that include veteran groups, as research finds women and whites are generally more likely to join voluntary organizations (Thoits & Hewitt, 2001). Our sample had very similar proportions of veterans with some college or higher, compared to national estimates (47% vs. 46%). In terms of age, on average, our somewhat was younger than the known characteristics of post-9/11 veterans (national estimates indicate that 75% are under age 45, in our sample about 86% were under age 45). This may be due to the inclusion of college veteran groups and online recruiting in our sample.

Table 1.

Descriptive statistics (N = 783).

	Mean	Std. Deviation
Strain Variables
Unmet discharge needs (0–9)	3.76	2.23
Identity loss (1–4)	2.91	.56
Outcome Variables
Resentment (1–5)	3.50	.78
Depression (0–27)	11.00	7.00
Risky behavior (0–13)	3.84	3.12
Control Variables
Age (years)	35.79	7.92
Sex (male)	.77	-
Race (Black)	.13	-
Education (1–8)	5.11	1.47
Divorced/separated	.12	-
Combat	.58	-
Deployments	3.21	1.50
Years since discharge	6.92	3.86
Rank (1–21)	6.86	4.90
Years served	9.03	6.73

In terms of the strain measures, the average number of unmet discharge needs is near the lower end of the possible range. The modal number of unmet needs was 2, with only about 9% indicating that all of their needs were met at time of discharge. However, civilian resentment is somewhat closer to the upper range of its scale. In terms of the outcome measures, on average, level of depression is near the lower end of the possible-scale range. The average frequency of risky behavior is about 4 of the 15 possible behaviors reported in the last 12 months, although the mode is “0” (20%). Fifty-eight percent of the respondents served in combat settings, similar to profiles found in other post-9/11 veteran samples (Igielnik, 2019). Our sample had a slightly higher average number of deployments, however (3 vs. 2).

Strain and outcome equations

The results from the series of equations showing the relationships between adjustment strains and each of the outcomes are presented in Table 2. Equation 1 shows the effects of unmet discharge needs and identity loss on depression, net of the control variables. As predicted, both have statistically significant positive effects. We also find that veterans with combat experience and those with a greater number of deployments report increased levels of depression. Consistent with general population studies, veterans with higher levels of education report lower levels of depression. Veterans with higher rank at time of discharge and longer service also report lower levels of depression. About 25% of the variation in current depression symptoms is attributable to the strain and control variables examined.

Table 2.

Post-discharge strains and risky behavior equations.

	(1)	(2)	(3)	(4)
	Depression	Resentment	Risky Behavior
Strain variables
Unmet discharge needs	.993*** (.104)	.039** (.012)	.286*** (.046)	.107* (.044)
Identity loss	1.915***	.407***	.576**	.064
	(.387)	(.044)	(.172)	(.164)
Negative emotions/sentiments
Depression				.162***
				(.015)
Resentment				.521***
				(.135)
Control variables
Age	.160	.002	−.067**	−.093***
	(.104)	(.006)	(.023)	(.021)
Sex (male)	−.222	.089	1.110***	1.105***
	(.557)	(.063)	(.246)	(.222)
Race (black)	−.340	−.116	−.382	−.245
	(.687)	(.078)	(.304)	(.274)
Education	−.771***	−.056**	−.181*	−.033
	(.171)	(.019)	(.075)	(.068)
Divorced/separated	1.486*	.058	1.305***	1.075***
	(.702)	(.079)	(.039)	(.280)
Combat	1.070*	.188**	.510*	.213
	(.477)	(.054)	(.210)	(.191)
Deployments	1.055***	−.014	.273***	.108
	(.163)	(.018)	(.072)	(.067)
Rank	−.122*	−.011	−.052*	−.023
	(.053)	(.006)	(.023)	(.021)
Years since discharge	−.115	.002	.078*	.090**
	(.070)	(.008)	(.031)	(.028)
Years served	−.226	.009	.032	.061*
	(.065)**	(.007)	(.028)	(.026)
R2	.247	.175	.201	.354

*p < .05 **p < .01 ***p < .001 (two-tailed tests).

Unstandardized OLS regression coefficients with standard errors in parentheses.

Equation 2 shows the relationship between adjustment strains and resentment toward civilians, net of the control variables. Here, we also find significant positive effects of unmet discharge needs and identity loss. Combat experience is also associated with increased resentment, while veterans that are more educated express less resentment. About 18% of the variation in resentment is attributable to the two strain variables and the control variables considered.

Equation 3 shows the effects of post-discharge strains on the number of risky behaviors in the last year. As predicted, unmet discharge needs and identity loss are significantly associated with greater risky behavior, controlling for the other variables. Not surprisingly, male veterans are more likely to engage in risky behavior, compared to their female counterparts, while older, more educated, and higher-rank veterans engage in less risky behavior. It is noteworthy that veterans with combat exposure and those with more deployments also report engaging in more risky behavior. Together, about 20% of the variation in risky behavior is explained by the sources of strain and the control variables.

Equation 4 estimates the effects of post-discharge adjustment strains on risky behavior, controlling for depression and resentment. In line with our expectations, we find both depression and resentment are significantly associated with increased risky behavior. When depression and resentment are added to the equation, the effect of unmet discharge needs is reduced by 63%, yet remains statistically significant. The effect of identity loss is reduced by 89%, and is no longer significant.

Table 3 shows the indirect effects of unmet needs and identity loss on risky behavior through depression and resentment. Using the “medsem” procedure, both Sobel (1987) and bootstrapped standard error-based tests (Zhao et al., 2010) for the significance of indirect effects showed that all indirect effect were statistically significant, indicating that depression and resentment mediate much of the effects of the strain and identity loss variables on risky behavior (Mehmetoglu, 2018).

Table 3.

Indirect effects of transition strains on risky behavior.

	Indirect effect	95% confidence interval
Unmet needs
via depression	.160***	(.116, .204)
	(.023)
via resentment	.020*	(.004, .036)
	(.008)
Identity loss
via depression	.292***	(.158, .426)
	(.068)
via resentment	.220***	(.101, .340)
	(.061)

*p < .05 ***p < .001

Standard errors in parentheses.

Supplementary analyses

Our measure of unmet needs is referenced to time of separation, whereas identity loss, depression, and risky behavior are measured more recently. As MTT specifies, conditions at time of discharge may have both short-term and long-term outcome trajectories. However, these effects may be stronger or more “acute” for those discharged more recently, compared to those longer ago. Although we controlled for the main effects of age and time since discharge, we examined the robustness of our findings in two ways. First, we examined the interaction between unmet needs and years since discharge (and an interaction between identity loss and years since discharge). We found neither effect was significant, indicating the effects of unmet needs and resentment on the outcomes are not conditional upon length since discharge in our sample. Second, we limited our analysis to a sub-sample of those veterans (32%, n = 283) who were discharged no more than 4 years ago. Our findings are substantively very similar, consistent with the findings from the interaction terms mentioned above. Together, this suggests that post-discharge strains may exert both short- and long-term cumulative effects.

Discussion

Consistent with insights from MTT, we found that unmet needs at time of discharge and loss of military identity were associated with increased symptoms of depression and resentment toward civilians among post-9/11 veterans. In addition, unmet discharge needs and loss of military identity are associated with increased risky behavior. Much of the effects of unmet needs and identity loss on risky behavior were mediated by increased depression and resentment. Together, these findings offer support for applying MTT to understand some of the processes that lead to risky and criminogenic behavior among military veterans.

Where prior studies focus on sources of strain, such as negative life events, or school difficulties (e.g., Agnew & White, 1992; Broidy, 2001; Paternoster & Mazerolle, 1994; Peck, 2013; Simons et al., 2003), we examined sources of strain identified by MTT as particularly relevant to veterans. Our findings regarding resentment suggest that, in addition to adverse effects of negative public perceptions held by veterans, unmet post-discharge needs may also lead to some degree of emotional upset and maladaptive behavior.

Our results also show that the loss of valued military identity may have adverse consequences manifested in psychological distress and troublesome behavior. When a salient component of self-concept for many, built up through intensive socialization processes and years of service, is lost, some veterans may feel adrift, which can lead them to act in potentially harmful ways. Our findings complement qualitative research on the challenges associated with identity transformation, providing insight into specific relationships between identity loss and outcomes. To the extent that what is “lost” in the military is eventually offset by other sources of valued identity based in new roles (e.g., education, work, family, and community), adverse consequences of identity-related transition strains may be mitigated (Grimell, 2017a, 2017b; McCormick et al., 2019; Rumann & Hamrick, 2010; Yanos, 2004).

Although prior research on strain and delinquency emphasizes anger as the key emotion that results from strain, our results show that post-discharge strains can impact depression, which in turn, may affect risky behavior among veterans. It is important to recognize that anger, depression, anxiety, and unhappiness are well-correlated aspects of distress that can result from situations perceived as unfair and frustrating (Aseltine et al., 2000; Mirowsky & Ross, 2003). Focusing on sources of transition strain helps build on clinically oriented research on the relationships between trauma, mental disorder, and family violence, and extends socio-criminogenic approaches to post-discharge risky and unlawful behavior among military veterans.

Because of the mental health consequences of adjustment strains, some veterans can be at increased risk for conflict with family members and others, which may result in aggressive behavior (Silver, 2002). Veterans under stress experiencing emotional upset, may also turn to maladaptive patterns of substance use and a number of other risky behaviors, such as drunk driving, gambling, and unsafe sexual behaviors that involve short-term gains or pleasure, yet entail potential long-term harmful consequences (Bucciol & Zarri, 2015; Semple et al., 2010).

While it is tempting to think of veterans as “acting out” in response to transition strains, further study is needed to understand to what extent others (family members, in particular) may be frustrated with veterans’ failure to “get going” upon return, leading to disputes and physical altercations (Solomon et al., 2005). Because of the interactional nature of violence, it is likely that transition-related stress leads persons to act in ways that result in their being the target of aggression. For example, persons who are under strain or depressed may act in ways that violate norms of interaction (e.g., unpleasant and uncooperative), leading to verbal and physical aggression by others, which then leads to retaliatory violence (Felson, 1992; Silver, 2002).

It is also important to note that there is likely a subset of persons with more impulsive personalities who have difficulty in adjustment prior to, while serving, and after discharge from the military (Sampson & Laub, 1990; Snowden et al., 2017). Indeed, the majority of military service personnel and veterans are law-abiding. Unfortunately, illegal behavior and untreated mental health problems among some veterans contributes to the stigmatization of others (Hipes et al., 2015). Our study suggests the need to examine processes that are more elaborate, whereby transition strains adversely affect mental health outcomes, in turn, increasing the possibility of risky and criminal behavior among veterans.

Limitations

While we are able to provide evidence of associations between certain post-discharge strains and subsequent outcomes, because the data is cross-sectional, it is difficult to establish the timing of effects without more precise sequencing of measures. Our analysis showed how unmet needs at time of discharge are associated with subsequent emotional and behavioral outcomes. Although our findings were robust when we limited the sample to a sub-set of more recently discharged veterans, the ability to sort out more “acute” effects from longer-terms ones is limited. There are also possible reciprocal effects between strain and outcome variables. For example, behavioral problems often create complications that can worsen symptoms of depression (Canada & Peters, 2017; Karp, 2001). Some veterans, who engage in risky and unlawful behavior, may feel as though they have failed to live up to standards emphasized in military culture, such as honor and integrity, resulting in depression and loss of self-esteem. Ideally, longitudinal data is needed that permit closer examination of the timing of effects, stability, and change in outcomes over time, and reciprocal effects between outcomes.

While unmet transition needs and identity loss are central aspects of MTT, arguably, other needs, such as family and community acceptance/integration also warrant examination. Further research is also needed to examine whether the effects of military transition strains are mitigated by “coping resources” such as self-efficacy and (instrumental and emotional) social support (Mirowsky & Ross, 2003; Pearlin et al., 1981).

Although our mostly urban sample does not vary markedly from known demographic characteristics of post-9/11 veterans, it does not include veterans from some areas (e.g., southern and rural), where there may be differences in economic opportunities, community attitudes toward veterans, and availability of services. However, given that veterans can be a very difficult to access population, as with other studies on special groups (e.g., persons with serious mental illness, ex-offenders), there is some tradeoff between representativeness and the opportunity to include unique measures not available in larger administrative data, in order to advance our understanding of certain social psychological processes (Thornberry & Krohn, 2000).

Implications for services

Our study helps confirm that, as veterans transition to civilian life, access to, and engagement in quality support services, including mental health treatment, job training, housing, and educational support are critical to help manage potential strains (Castro & Kintzle, 2018; Elliott et al., 2011). However, the cultural environment of the military emphasizes masculinity, toughness, and showing courage in the face of adversity (Castro, 2006). While valued in active-duty military contexts, these same attitudes may prevent veterans from seeking help to deal with mental health problems and other struggles, that when not addressed, can lead to risky and illegal behavior (Ben-Zeev et al., 2012). Our findings further suggest that strategies for effectively managing resentment toward civilians and coping with civilian misunderstanding of reintegration challenges may be a concern for many veterans. In programs to support veterans, attention should also be paid to acknowledging potential strain of identity change and finding meaning in new roles.

Oftentimes, services for military populations are disconnected, leaving transitioning service members to navigate resources from the government, state, and local communities. The transition from Department of Defense (DoD) to Veteran Affairs (VA) or state and local care can be difficult as it is the responsibility of the veteran to manage. Providing these suggested services to transitioning veterans is the responsibility of the DoD, VA, state and local government, and local service providers. Results of the study highlight the need for coordinated care to address transition needs not only at the time of discharge, but that transition-specific services (e.g., counseling) should ideally be available on a sustained basis. These services can be a mixture of professional and peer-led, and participation should be incentivized, especially given reluctance of many veterans to seek help.

Conclusion

In sum, consistent with insights from MTT, our findings indicate that previously unexamined veteran reintegration strains – unmet discharge needs and identity loss – are associated with depression and resentment, after taking into account a number of potentially confounding variables. Our study highlights the need to examine complex causal processes associated with risky, criminogenic behavior after military service. Ideally, focus is needed on the longer-term trajectories and relationships between a range of pre- and post-service adjustment factors, and well-being outcomes. Practically, our findings suggest that targeted and coordinated services that help ease the potentially challenging transition from military to civilian are vital for the well-being of veterans, their families, and for community safety.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Requests for the data that support the findings of this study may be made to author, SK. The data are not publicly available due to their containing information that could compromise the privacy of research participants.