Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 May 1.
Published in final edited form as: Appl Psychol Health Well Being. 2023 Dec 10;16(2):723–744. doi: 10.1111/aphw.12513

Meaning in Life Following Service Among Post-9/11 Military Veterans: A Latent Growth Mixture Model Analysis

Shaina A Kumar 1,2, Shelby Borowski 3, Dawne Vogt 1,3
PMCID: PMC11062845  NIHMSID: NIHMS1944938  PMID: 38072466

Abstract

Meaning in life refers to the “sense made of, and significance felt regarding, the nature of one’s being and existence.” Meaningful living promotes well-being, resilience, and personal growth. Yet, much remains unknown about how meaning changes over time and determinants of meaning, particularly during major life transitions. We identified distinct trajectories of meaning using latent growth mixture models and examined prospective predictors of class membership in a military veteran cohort assessed at multiple time points throughout the first three years after leaving service. Three trajectories were identified: consistently high meaning (89.5%; n = 7,025), diminishing meaning (6.1%; n = 479), and strengthening meaning (4.4%; n = 348). Veterans with greater posttraumatic stress symptoms, depression symptoms, and moral injury experienced increased odds of a less adaptive trajectory (i.e., diminishing and/or strengthening versus consistently high meaning), whereas veterans who reported greater psychological resilience, community relationship satisfaction, and intimate relationship satisfaction experienced lower odds of a less adaptive trajectory. Several gender differences were also observed. Results provide insight into veteran subgroups that are more likely to experience lower meaning after leaving military service and thus may benefit from additional support to reduce their risk for poor longer-term health and well-being outcomes.

Keywords: gender, meaning, psychopathology, resilience, social support, well-being

Introduction

Meaning in life refers to the “sense made of, and significance felt regarding, the nature of one’s being and existence” (Steger et al., 2006). The experience of meaning is widely viewed as a critical component of well-being, contributing to the achievement of one’s potential as a human being (Heintzelman & King, 2014; Steger, 2012). Indeed, meaningful living has been directly equated with authentic living, promoting happiness and satisfaction with life, personal strengths, and resilience (Steger et al., 2006; Steger, 2012). Moreover, perceiving that life is worthwhile, having a sense of purpose, and drawing on meaning is contradictory to manifestations of psychological distress and can act as a buffer during times of suffering (Steger, 2012). Because meaningful living encourages greater use of effective coping strategies, experiencing personal growth during times of stress is also possible (Steger, 2012). In this way, it is important to understand how meaning in life might evolve over time and what factors might determine one’s sense of meaning to inform efforts aiming to bolster well-being in populations whose sense of meaning is challenged by major life transitions.

Researchers have theorized that, unless life circumstances change considerably, meaning in life should be stable over time. Empirical research largely supports this assertion. For example, Krause and Hayward (2014) found that one’s sense of meaning is generally consistent across time. Similarly, Steger and Kashdan (2007) found that meaning in life showed a trend toward continuity in everyday life, rather than waxing and waning in response to momentary influences. That said, a considerable amount of unexplained variance remained in their study, and as such they proposed it is still possible that one’s sense of meaning might be sensitive to influential life events. This may be especially true for those who are searching for meaning in their lives and actively making attempts to comprehend their experiences in the face of a significant life transition. In particular, life transitions involving role entry (e.g., getting married, starting college) or exit (e.g., getting divorced, retiring from the workforce) can encourage significant shifts in identity and prompt individuals to reevaluate the meaning they attribute to their lives (Altmaier, 2020). In the current study, we test this possibility in a sample of newly separated military veterans, and we also aim to uncover important predictors of life meaning trajectories.

Keyes (2007) suggests that mental health is not merely the absence of mental illness but also the presence of something positive. Recognizing meaning in life as a key aspect of positive mental health, a great deal of research has been conducted on the many correlates of meaning—both factors that may increase risk for lower meaning in life, otherwise known as “risk factors,” and factors that may foster resilience to lower meaning in life, otherwise known as “protective factors.” In terms of risk factors, research indicates that lower levels of meaning in life are associated with greater negative affect and emotions, including fear, anger, shame, sadness, and hopelessness; moral injury and rumination; posttraumatic stress, depression, anxiety, and general psychological distress; and suicidal ideation and attempts (Braden et al., 2015; Fischer et al., 2020; Fischer et al., 2023a, 2023b, 2023c; Kelley et al., 2021; Steger, 2012). Alternatively, many studies have also linked higher levels of meaning in life to increases in factors that might promote positive adjustment. Some of these factors include positive affect and emotions, including feelings of love, joy, happiness, hope, and optimism; self-esteem, self-acceptance, and self-actualization; ambition and strength; psychological resilience and personal growth; life satisfaction and satisfaction with others; and general well-being (Dezutter et al., 2015; Fischer et al., 2023a; Niu et al., 2016; Steger, 2012). Collectively, these studies provide critical insight into key correlates of meaning in life; however, there is a need to replicate and extend these results in populations whose experiences of meaning are challenged by major life transitions.

One such group is military veterans. The transition from active service to civilian life can raise profound existential questions and prompt reconsideration of one’s life purpose. In fact, Erik Erikson coined the term “identity crisis” following his experience working with veterans, commonly describing periods of instability and insecurity in this group (Orazem et al., 2017). Losing such a significant sense of identity can be challenging, especially when veterans struggle to establish new personal goals and values. This loss of meaning also has substantial implications for their overall well-being and reintegration into society—as noted above, a lack of meaning can be linked to feelings of purposelessness, existential distress, and increased vulnerability to mental health concerns, including suicidal ideation and attempts (Braden et al., 2015; Fischer et al., 2023b, 2023c; Steger, 2012), which veterans are already at increased risk of experiencing given their exposure to potentially stressful and traumatic events during service (Sokol et al., 2021). Indeed, among military veterans, research has shown that lower levels of meaning are inversely related to many negative psychological states, such as posttraumatic stress, depression, guilt, and distress (Park et al., 2023). Notably, this transitional period is marked by many unique challenges to the extent that researchers have coined the term “deadly gap” to describe the transition out of military service to civilian life, where transitioning veterans are at higher risk for suicide compared with both the broader veteran population and the United States public (Sokol et al., 2021). Together, this is all the more reason to explore how meaning might unfold over time and what factors might predict adaptive trajectories of meaning in this group.

Why Meaning in Life?

Although there are a number of positive psychology constructs that would be valuable to examine in the context of major life transitions (e.g., overall psychological well-being), understanding how and when each is most relevant to consider is critical for advancing the science of positive psychology (King & Hicks, 2021). The current study focused on meaning in life as the primary outcome variable for several reasons. First, meaning in life encompasses a broad range of aspects, including coherence, significance, and value in one’s existence (Martela & Steger, 2016). Notably, meaning in life can also include the pursuit of purpose. In this way, it is a more inclusive concept and allows for a more holistic examination of the human existence. Second, meaning in life is commonly viewed as a critical component of eudaimonia, above and beyond other positive psychological constructs (Steger et al., 2013). Researchers argue that individuals are unable to attain optimal functioning in the absence of meaning in life, and thus having a sense of meaning serves as an essential ingredient of flourishing. Third, although meaning in life is relatively stable across time, evidence suggests meaning might evolve as individuals encounter different circumstances and life transitions (Steger & Kashdan, 2007). Studies in this area can therefore shed light on distinct ways that individuals might redefine their sense of meaning in the face of change. In the case of the current study, there is the possibility of expanding our knowledge about for whom (e.g., veterans) and under what circumstances (e.g., transitioning from military to civilian life) meaning might be particularly challenged and, in turn, inform clinical interventions aiming to boost meaning and protect against ill-being.

The Present Study

We aimed to extend recent work and contribute to the growing literature on life meaning in several ways. First, although meaning in life has been extensively studied, empirical work examining changes in meaning are rare. Moreover, to our knowledge, only one other study has modeled trajectories of change in life meaning among military veterans (i.e., Park et al., 2023). Following their investigation, Park et al. (2023) identified three distinct meaning trajectories over time: (1) moderately high and stable, (2) low and increasing, and (3) low and decreasing. Given the dearth of empirical research in this area, we aimed to replicate and extend these findings. Notably, Park et al. (2023) examined trajectories of meaning across 12 months among veterans who separated from service within the last five years, whereas we examined trajectories of meaning up to three years post-separation among veterans who separated from service in the prior 90 days. This is consistent with findings suggesting that veterans may experience even greater changes in well-being in the second and third year following separation from service (Vogt et al., 2022). Through examination of these data, we can gain a better understanding of what life meaning might look like earlier on and over longer periods of time for this group.

Second, most research to date has conceptualized meaning as a precursor to outcomes of interest (e.g., meaning in life as a predictor of psychosocial adjustment). One theoretical basis for why meaning in life might also be of interest as an outcome is offered by the meaning-making model (Park, 2010). According to this model, stressful experiences and their sequelae can lead to changes in one’s perception of life as meaningful, purposeful, and manageable. In this context, it is possible that associated risk or protective factors may serve as predictors for understanding potential pathways to finding a new sense of meaning (or, alternatively, failing to find meaning) as individuals navigate their experiences of distress and recovery. For that reason, uncovering factors that might predict meaning made following a challenging experience is critical.

That said, little research has explored factors that might contribute to levels of meaning in life over time, much less trajectories of meaning. In the aforementioned study, Park et al. (2023) found that factors related to demographics, deployment experiences, and post-deployment resources predicted belonging to meaning trajectories. We are grateful to Park et al. (2023) for their valuable contributions to this area of study; however, they noted, “We attempted to examine a broad range of determinants of meaning class, but likely omitted important components that were not adequately assessed in our study (e.g., depressive symptoms).” We aimed to expand upon their work by examining an additional range of predictors of meaning that might be malleable to intervention and particularly relevant to veterans, considering both risk and protective factors that have been previously identified as important to veterans’ perceptions of meaning in life (e.g., Fischer et al., 2020; Fischer et al., 2023a; Kelley et al., 2021; Niu et al., 2016), as well as their broader well-being (Vogt et al., 2021). Specifically, we examined three potential risk factors, including posttraumatic stress symptoms, depression symptoms, and experiences of moral injury, as well as three protective factors, including psychological resilience, community relationship satisfaction, and intimate relationship satisfaction.

Finally, to our knowledge, no study has examined gender-stratified models of life meaning trajectories. This is an important consideration, given that researchers have found gender differences between women and men regarding their overall sense of meaning in life (e.g., Dezutter et al., 2015; Park et al., 2023). By examining trajectories of meaning and determinants of meaning in total and gender-stratified samples, we hope to add greater specificity to the meaning literature. These findings might also serve to inform future research directions and clinical intervention efforts for this group.

In sum, the goals of the present study were to (1) identify classes of unique trajectories of change in life meaning among a population-based sample of veterans following separation from military service; (2) determine key predictors of class membership; and (3) assess meaning trajectories and respective predictors for the entire sample and separately for women and men to further advance our knowledge of life meaning. Notably, our sample was comprised of post-9/11 veterans who had recently left military service, a period of time when they must establish a new sense of meaning in life. Results from the current study might especially benefit this group.

Based on theory and prior research, we hypothesized that our sample of veterans would demonstrate at least three trajectories of life meaning over time, such that one class would demonstrate relatively high and stable levels of meaning over time; one class would demonstrate decreasing levels of meaning over time; and one class would demonstrate increasing levels of meaning over time (H1). We also hypothesized that our three risk factors would predict less adaptive trajectory membership (i.e., posttraumatic stress symptoms, depression symptoms, moral injury; H2), whereas our three protective factors would predict more adaptive trajectory membership (i.e., psychological resilience, community relationship satisfaction, intimate relationship satisfaction; H3). No specific hypotheses regarding the gender-stratified models were created prior to conducting the analyses given the lack of research in this area.

Method

Participants and Procedures

All procedures were approved by the local Institutional Review Boards, and all participants provided informed consent. The Veteran Metrics Initiative (TVMI) study (Vogt et al., 2018) is a prospective cohort study that assessed post-9/11 veterans’ well-being throughout the first three years of their transition from military to civilian life. All United States service members who had separated from active duty service or deactivated from the National Guard/Reserves in the prior 90 days and who resided within the continental United States were identified from the Veteran’s Administration/Department of Defense Identity Repository (VADIR) in Fall 2016. Study recruitment used a modified Dillman approach with four invitation/reminder letters (Dillman et al., 2014), inviting all eligible veterans (i.e., the full separation cohort; N = 48,965) to complete an online survey. Potential participants received a $5 pre-incentive with the invitation letter and those who completed the survey received additional incentives for each survey completed and $20 at baseline, with incentives increasing by $5 at each subsequent time point (Coughlin et al., 2011). Data were collected at six time points: baseline (within approximately 3 months of separation; T1) and at approximately 9, 15, 21, 27, and 33-months post-separation (T2–T6). Overall, 9,566 participants (n = 1,743 women, n = 7,823 men) responded at T1. After reducing the denominator by known undeliverable mailings (n = 4,682) and deceased individuals (n = 2), the response rate was 23%, which is consistent with other studies of post-9/11 veterans that typically range from 20% to 30% (Coughlin et al., 2011). Follow-up response rates were 75% at T2 (n = 7,200), 75% at T3 (n = 7,201), 68% at T4 (n = 6,480), 61% at T5 (n = 5,844), and 55% at T6 (n = 5,249). Providing evidence that non-response bias was not a large concern for this study, respondents were similar to the sampling frame on many characteristics (e.g., gender, race and ethnicity, branch of service), although lower-ranking enlisted service members were less likely to participate than officers (see Vogt et al., 2018).

Of veterans in this sample (those who completed at least one life meaning measure at T3–T6; N = 7,852), the majority identified as men (81.8%; n = 6,423) and ranged in age from 18 to 64 years old (M = 34.16, SD = 9.49). Veterans identified as White (78.0%; n = 6,128), Black or African American (12.5%; n = 982), Asian (5.0%; n = 392), Native American or Alaska Native (3.3%; n = 260), Other Pacific Islander (1.5%; n = 121), West Asian, Middle Eastern, or North African (0.4%; n = 34), Native Hawaiian (0.4%; n = 28), and “other” (6.2%; n = 121; more than one race could be selected), with 14.0% (n = 1,102) identifying as being of Hispanic origin. Approximately 12.2% (n = 956) continued to serve in the National Guard/Reserves, while the remainder of veterans reported having separated from the Army (32.9%; n = 2,583), Navy (19.5%; n = 1,525), Air Force (19.0%; n = 1,492), or Marine Corps (16.4%; n = 1,284). They reported a range of paygrades, from E1-E4 junior enlisted (29.0%; n = 2,281), E5-E6 mid-grade enlisted (30.0%; n = 2,359), E7-E9 senior enlisted (17.2%; n = 1,349), W1-W5 warrant officers or O1-O3 junior officers (10.1%; n = 794), to O4-O10 senior officers (13.6%; n = 1,069). The majority of veterans (70.8%; n = 5,552) reported at least one war deployment, and 85.6% (n = 6,718) indicated an honorable separation.

Measures

Meaning in Life

Veterans completed the Purpose in Life Test-Short Form (PIL-SF; Schulenberg et al., 2011) to examine levels of meaning in life at T3–T6. The PIL-SF is a 4-item self-report measure in which each item is anchored to a specific question and rated on a scale from 1 to 7. For example, the second item reads, “My personal existence is,” with response options ranging from 1 (utterly meaningless without purpose) to 7 (very purposeful and meaningful). A sum score was created (possible range: 4 to 28), with higher scores indicating greater meaning in life. Past work has supported the validity of this measure (e.g., Schulenberg et al., 2011).

Prospective Predictors of Meaning in Life

Posttraumatic Stress Disorder Symptoms.

Veterans completed the Primary Care PTSD Screen for DSM-5 (PC-PTSD-5; Prins et al., 2016) to examine levels of posttraumatic stress symptoms at T1. The PC-PTSD-5 is a 5-item self-report measure in which each item is rated on a dichotomous 0 (no) or 1 (yes) scale. Veterans completed this measure if they indicated having experienced a traumatic experience in the past. Veterans who did not endorse a traumatic event received a score of zero on this measure, consistent with scoring guidelines (Prins et al., 2016). A sum score was created (possible range: 0 to 5), with higher scores indicating greater symptoms. Past work has supported the validity of this measure (e.g., Bovin et al., 2021; Prins et al., 2016).

Depression Symptoms.

Veterans completed the Patient Health Questionnaire-2 (PHQ-2; Kroenke et al., 2003) to examine levels of depression symptoms at T1. The PHQ-2 is a 2-item self-report measure in which each item is rated on a scale from 0 (not at all) to 3 (nearly every day). A sum score was created (possible range: 0 to 6), with higher scores indicating greater symptoms. Past work has supported the validity of this measure (e.g., Kroenke et al., 2003).

Moral Injury.

Veterans completed the Moral Injury Events Scale (MIES; Nash et al., 2013) to assess moral injury at T2 (moral injury was assessed at T2 due to space constraints at T1). The MIES is a 9-item self-report measure in which each item is rated on a scale from 1 (strongly disagree) to 6 (strongly agree). A sum score was created (possible range: 9 to 54), with higher scores indicating greater moral injury. Past work has supported the validity of this measure (e.g., Nash et al., 2013; see Szabó et al., 2023 for a review).

Psychological Resilience.

Veterans completed the Brief Resilience Scale (BRS; Smith et al., 2008) to examine levels of psychological resilience at T1. The BRS is a 6-item self-report measure in which each item is rated on a scale from 1 (strongly disagree) to 5 (strongly agree). A sum score was created (possible range: 6 to 30), with higher scores indicating greater psychological resilience. Past work has supported the validity of this measure (e.g., McKay et al., 2021; Smith et al., 2008).

Community and Intimate Relationship Satisfaction.

The Well-Being Inventory (WBI; Vogt et al., 2019) is a self-report measure that assesses well-being across four dimensions: vocation, finances, health, and social relationships. For the purpose of the present study, we examined well-being in the dimension of social relationships, specifically community and intimate relationship satisfaction, at T1. Regarding community relationship satisfaction, veterans rated four items on a scale from 1 (very dissatisfied) to 5 (very satisfied) to determine overall satisfaction within communities and with friends and relatives. Regarding intimate relationship satisfaction, veterans who reported having a significant other rated six items on a scale from 1 (very dissatisfied) to 5 (very satisfied) to determine overall satisfaction with their significant other’s contribution to the relationship. An average score was created for each domain (possible ranges: 1 to 5), with higher scores indicating greater satisfaction. Past work has supported the validity of this measure (e.g., Vogt et al., 2019).

Data Analytic Plan

We first examined descriptives and Pearson’s r correlations among study variables. Next, using Mplus 8 (Muthén & Muthén, 1998-2017), we applied a latent growth modeling approach to identify distinct trajectories of change in life meaning across 15, 21, 27, and 33 months post-separation and predict trajectory assignment. We used maximum likelihood estimation with robust standard errors (MLR) to address potential violations of normality assumptions and maximize data available in the latent growth models. Under MLR, we were able to retain all participants who had at least one data point related to life meaning in the current study (see Curran et al., 2010 for a discussion of maximum likelihood estimation). Finally, to assess for differences in trajectories by gender, we conducted modeling with the total sample and within gender-stratified samples. To note, sample size was determined from the number of veterans who had completed the larger study from which these data were drawn; thus, a priori power analyses could not be conducted. However, recommended data requirements to use growth models include a sample size of at least 100; at least three repeated measures per individual for a sizable portion of cases; and continuous and normally-distributed repeated measures when using maximum likelihood estimation (Curran et al., 2010). In the case of more complex mixture models, researchers recommend a minimum sample size of 500 (Meyer & Morin, 2016). Data used in the current study met these requirements.

Regarding model estimation, we first compared simple growth curve models using chi-square difference tests. For all three samples, a model with intercept (1, 1, 1, 1), linear (0, 1, 2, 3), and non-linear (0, 1, 4, 9) growth parameters generally fit best, and these parameters were included in subsequent models. Second, we regressed a specified number of latent classes onto the growth parameters to establish the optimal number of trajectories. No predictors were included in these models. We examined latent class growth analyses (LCGA; intercept, linear, and non-linear growth parameters fixed) and latent growth mixture models (LGMM; allowing select growth parameters to be freely estimated) with one to five classes. Residual variance in observed life meaning scores was constrained to be equal. Models were evaluated on multiple fit indices: (1) Akaike Information Criterion (AIC), Bayesian Information Criteria (BIC), and Sample Size Adjusted Bayesian Information Criteria (aBIC), with lower values indicating better fit (Jung & Wickrama, 2008); (2) Lo-Mendell-Rubin Adjusted Likelihood Ratio Test (LRT) and Bootstrapped Likelihood Ratio Test (BLRT), with a significant p-value indicating the model with k classes fits the data better than the model with k − 1 classes (Nylund et al., 2007); (3) Entropy and Average Latent Class Posterior Probabilities (APP), with values closer to 1 indicating better accuracy of classification (Lubke & Muthén, 2007; Nagin, 2010); and (4) the proportion of individuals assigned to each class greater than 1% (Hipp & Bauer, 2006). Lastly, we examined plots of average life meaning across time and considered past theory and research to select the optimal number of classes. The best-fitting models were those with the intercept freely estimated and linear and non-linear growth parameters fixed (i.e., LGMM).

Finally, we considered the role of prospective predictors in class membership. In all models, predictors were regressed on the discrete latent class variable using multinomial regression, following the three-step method (RU3STEP) recommended by Asparouhov and Muthén (2014). Results assess predictor associations (odds ratios [ORs]) with most likely class membership. ORs are presented alongside 95% confidence intervals (CIs).

Results

Descriptive and Bivariate Statistics

Descriptive and bivariate statistics among primary study variables are displayed in Table 1. All measures demonstrated adequate reliability. Notably, posttraumatic stress symptoms, depression symptoms, and moral injury were all negatively correlated with life meaning at each time point (rs ranging from −.28 to −.45, ps < .001). In contrast, psychological resilience, community relationship satisfaction, and intimate relationship satisfaction were all positively correlated with life meaning at each time point (rs ranging from .30 to .42, ps < .001). All variables were sufficiently distinct (rs < .70), including life meaning at each time point, and thus there were no concerns about multicollinearity (Tabachnick & Fidell, 1996).

Table 1.

Means, Standard Deviations, Observed Ranges, Reliability Estimates, and Correlations Among Primary Study Variables

1 2 3 4 5 6 7 8 9 10
1. T1 Posttraumatic Stress Symptoms - - - - - - - - - -
2. T1 Depression Symptoms .62** - - - - - - - - -
3. T2 Moral Injury .39** .37** - - - - - - - -
4. T1 Psychological Resilience −.48** −.57** −.36** - - - - - - -
5. T1 Community Relationship Satisfaction −.36** −.47** −.30** .44** - - - - - -
6. T1 Intimate Relationship Satisfaction −.35** −.43** −.25** .37** .49** - - - - -
7. T3 Meaning in Life −.31** −.45** −.31** .41** .42** .36** - - - -
8. T4 Meaning in Life −.31** −.43** −.31** .38** .42** .34** .67** - - -
9. T5 Meaning in Life −.32** −.43** −.29** .39** .41** .33** .63** .69** - -
10. T6 Meaning in Life −.28** −.39** −.28** .34** .38** .30** .60** .64** .68** -
Mean 1.39 1.22 19.09 22.58 3.95 4.04 21.06 21.25 21.22 21.64
SD 1.82 1.69 10.62 4.84 0.88 0.98 4.93 4.84 4.87 4.87
Observed Range 0.00 – 5.00 0.00 – 6.00 9.00 – 54.00 6.00 – 30.00 1.00 – 5.00 1.00 – 5.00 4.00 – 28.00 4.00 – 28.00 4.00 – 28.00 4.00 – 28.00
N 7848 7812 6866 7850 7847 6599 7199 6478 5833 5244
Cronbach’s Alpha (α) .84 .89 .91 .88 .84 .92 .90 .90 .90 .91
McDonald’s Omega (ω) .84 .89 .88 .88 .84 .93 .90 .90 .91 .92
Open in a new tab

Note.

**

p < .001.

Class Selection

Fit indices (e.g., information criteria, LRTs, classification accuracy metrics; see Table 2) and plots of life meaning by assigned class suggested that a three-class solution best described changes in meaning over time for the total and gender-stratified samples. For each sample, the information criteria improved as the number of classes specified increased; APPs favored one- through four-class solutions; and entropy values largely favored three- or four-class solutions. The LRT suggested a three-class solution for the total sample but a one-class solution for women and a two-class solution for men. Given that simulation studies have demonstrated the LRT (original and adjusted versions) is associated with elevated Type I error rates, whereas the BLRT works well consistently, we favored the BLRT findings (Nylund et al., 2007). Considering indicators of fit and parsimony, we debated between the three- and four-class solutions for each sample. We then inspected the plots to adjudicate the results. Notably, the four-class solution contained an additional trajectory that diverged only slightly from trajectories already represented in the three-class solution and information criteria were also minimally different between the two classes. Ultimately, the three-class solution appeared optimal, and we prioritized communicating class structures that were distinguishable and clinically meaningful. The three-class solution also seemed most likely to be replicated in future studies given prior research (Park et al., 2023) and respective sample sizes in each class.

Table 2.

Fit Indices for Unconditional Growth Models

AIC BIC aBIC LRT BLRT Entropy APPs (range)
Total Sample
 1-Class 137511.71 137546.55 137530.66 - - - 1.00
 2-Class 137001.09 137063.81 137035.21 < 0.001 < 0.001 0.70 0.78 to 0.93
 3-Class† 136513.24 136603.83 136562.52 0.005 < 0.001 0.79 0.75 to 0.93
 4-Class 136206.52 136324.98 136270.96 0.135 < 0.001 0.80 0.76 to 0.92
 5-Class 135942.87 136089.21 136022.47 0.157 < 0.001 0.81 0.70 to 0.91
Women
 1-Class 25068.49 25094.82 25078.93 - - - 1.00
 2-Class 24953.23 25000.61 24972.02 0.0753 < 0.001 0.76 0.80 to 0.94
 3-Class† 24862.96 24931.40 24890.11 0.0240 < 0.001 0.77 0.74 to 0.92
 4-Class 24789.47 24878.97 24824.97 0.0292 < 0.001 0.81 0.75 to 0.93
 5-Class 24747.64 24858.20 24791.49 0.3726 < 0.001 0.75 0.70 to 0.89
Men
 1-Class 112440.50 112474.33 112458.45 - - - 1.00
 2-Class 112045.64 112106.55 112077.95 < 0.001 < 0.001 0.69 0.78 to 0.93
 3-Class† 111624.73 111712.71 111671.40 0.1724 < 0.001 0.80 0.76 to 0.94
 4-Class 111349.93 111464.98 111410.96 0.1795 < 0.001 0.80 0.75 to 0.92
 5-Class 111133.16 111275.28 111208.55 0.0122 < 0.001 0.80 0.72 to 0.91
Open in a new tab

Note.

†

Best fitting model.

AIC = Akaike Information Criterion; BIC = Bayesian Information Criterion; aBIC = Sample Size Adjusted Bayesian Information Criterion; LRT = Lo-Mendell-Rubin Adjusted Likelihood Ratio Test; BLRT = Bootstrapped Likelihood Ratio Test; APPs = Average Latent Class Posterior Probabilities. In all models, the intercept is freely estimated and linear and quadratic components are fixed.

Intercept, linear, and quadratic growth parameters for the three-class models are displayed in Table 3. Trajectories of life meaning and respective sample sizes by known class are displayed in Figure 1. Most of the sample (89.5% total, n = 7,025; 87.1% women, 90.4% men) was assigned to a “consistently high meaning” class characterized by elevated levels of meaning over time, which we considered adaptive (the reference class). A “diminishing meaning” class (6.1% total, n = 479; 6.4% women, 5.4% men) demonstrated below average levels of meaning initially that decreased further with time, which we considered less adaptive. A smaller proportion of the sample (4.4% total, n = 348; 6.5% women, 4.2% men) was assigned to a “strengthening meaning” class, reporting low levels of meaning initially that increased over time. Notably, although those in the “strengthening meaning” class experienced an increase in their level of meaning over time, we considered this trajectory less adaptive compared to the “consistently high meaning” class given their initial report of below average life meaning. Collectively, these results align with prior literature and support our first hypothesis (H1).

Table 3.

Growth Parameters by Known Class for Unconditional Growth Models

Classes Total Sample
 Women
 Men
Estimate SE Estimate SE Estimate SE
Consistently High Meaning
 Intercept 22.14*** 0.15 22.53*** 0.26 21.96*** 0.27
 Linear 0.04 0.07 −0.38* 0.17 0.11 0.08
 Quadratic 0.03 0.02 0.18** 0.05 < 0.01 0.03
Diminishing Meaning
 Intercept 16.76*** 0.84 16.61*** 1.17 17.35*** 1.98
 Linear −2.92** 0.94 −4.63* 1.85 −2.93* 1.48
 Quadratic 0.55* 0.23 1.21* 0.58 0.42 0.29
Strengthening Meaning
 Intercept 12.04*** 0.43 13.42*** 0.66 11.68*** 0.51
 Linear 3.72*** 0.91 5.55*** 1.15 3.19** 1.03
 Quadratic −0.36 0.26 −1.12** 0.34 −0.15 0.28
Open in a new tab

Note.

*

p < .05

**

p < .01

***

p < .001.

Figure 1.

Figure 1

Open in a new tab

Life Meaning Trajectories and Sample Sizes by Known Class in Total and Gender-Stratified Samples

Predicting Class Membership in the Total Sample

Table 4 displays results from multinomial logistic regressions predicting assignment to less adaptive meaning trajectories (i.e., diminishing and strengthening meaning) relative to the consistently high meaning trajectory. For the total sample, significant predictors of the less adaptive meaning trajectories included posttraumatic stress symptoms, depression symptoms, and experiences of moral injury. Specifically, the odds of belonging to the diminishing meaning versus consistently high meaning class were higher for veterans who endorsed greater posttraumatic stress symptoms (OR = 1.19 [95% CI = 1.11, 1.27]), depression symptoms (OR = 1.20 [95% CI = 1.09, 1.33]), and moral injury (OR = 1.04 [95% CI = 1.03, 1.05]). Similarly, the odds of belonging to the strengthening meaning versus consistently high meaning class were also higher for veterans who endorsed greater posttraumatic stress symptoms (OR = 1.26 [95% CI = 1.16, 1.37]), depression symptoms (OR = 1.30 [95% CI = 1.20, 1.41]), and moral injury (OR = 1.05 [95% CI = 1.03, 1.06]).

Table 4.

Multinomial Logistic Regressions Predicting Class Membership

Total Sample
Predictors Diminishing Meaning
 Strengthening Meaning
OR 95% CI OR 95% CI
T1 Posttraumatic Stress Symptoms 1.19 [1.11, 1.27] 1.26 [1.16, 1.37]
T1 Depression Symptoms 1.20 [1.09, 1.33] 1.30 [1.20, 1.41]
T2 Moral Injury 1.04 [1.03, 1.05] 1.05 [1.03, 1.06]
T1 Psychological Resilience 0.92 [0.89, 0.95] 0.89 [0.86, 0.93]
T1 Community Relationship Satisfaction 0.67 [0.56, 0.79] 0.65 [0.55, 0.77]
T1 Intimate Relationship Satisfaction 0.80 [0.68, 0.93] 0.66 [0.56, 0.77]
Women
Predictors Diminishing Meaning
 Strengthening Meaning
OR 95% CI OR 95% CI
T1 Posttraumatic Stress Symptoms 1.31 [1.14, 1.50] 1.23 [1.06, 1.43]
T1 Depression Symptoms 1.25 [1.09, 1.45] 1.31 [1.14, 1.52]
T2 Moral Injury 1.06 [1.03, 1.08] 1.04 [1.01, 1.08]
T1 Psychological Resilience 0.89 [0.85, 0.94] 0.92 [0.87, 0.98]
T1 Community Relationship Satisfaction 0.50 [0.35, 0.70] 0.58 [0.42, 0.80]
T1 Intimate Relationship Satisfaction 0.76 [0.57, 1.01] 0.87 [0.65, 1.17]
Men
Predictors Diminishing Meaning
 Strengthening Meaning
OR 95% CI OR 95% CI
T1 Posttraumatic Stress Symptoms 1.08 [0.99, 1.19] 1.23 [1.11, 1.35]
T1 Depression Symptoms 1.15 [1.03, 1.29] 1.34 [1.22, 1.47]
T2 Moral Injury 1.02 [0.997, 1.03] 1.03 [1.01, 1.05]
T1 Psychological Resilience 0.95 [0.91, 0.99] 0.90 [0.87, 0.94]
T1 Community Relationship Satisfaction 0.79 [0.65, 0.96] 0.68 [0.56, 0.82]
T1 Intimate Relationship Satisfaction 0.92 [0.76, 1.11] 0.64 [0.54, 0.77]
Open in a new tab

Note. Odds ratios represent the change in odds of assignment to a comparison class relative to the reference class (i.e., consistently high meaning) per unit increase in each predictor. Bolded odds ratios indicate significance (i.e., the 95% CI does not contain 1.00).

In contrast, significant predictors of the consistently high meaning trajectory included psychological resilience, community relationship satisfaction, and intimate relationship satisfaction for the total sample. Specifically, the odds of belonging to the diminishing meaning versus consistently high meaning class were lower for veterans who reported greater psychological resilience (OR = 0.92 [95% CI = 0.89, 0.95]), community relationship satisfaction (OR = 0.67 [95% CI = 0.56, 0.79]), and intimate relationship satisfaction (OR = 0.80 [95% CI = 0.68, 0.93]). Similarly, the odds of belonging to the strengthening meaning versus consistently high meaning class were also lower for veterans who reported greater psychological resilience (OR = 0.89 [95% CI = 0.86, 0.93]), community relationship satisfaction (OR = 0.65 [95% CI = 0.55, 0.77]), and intimate relationship satisfaction (OR = 0.66 [95% CI = 0.56, 0.77]). These results align with prior literature and support our second (H2) and third (H3) hypotheses.

Gender Differences in Predicting Class Membership

Gender-stratified multinomial logistic regression results are displayed in Table 4. Like the full sample, depression symptoms, psychological resilience, and community relationship satisfaction had similar impacts on life meaning trajectories among both women and men veterans. Higher levels of depression symptoms increased odds of belonging to the diminishing or strengthening classes compared to the consistently high meaning class, and psychological resilience and community relationship satisfaction decreased odds of belonging to the diminishing or strengthening classes compared to the consistently high meaning class.

Notably, several gender differences were also observed. Unique findings appeared when examining posttraumatic stress symptoms, moral injury, and intimate relationship satisfaction as predictors of life meaning trajectories among women and men. Regarding posttraumatic stress symptoms and moral injury, these factors did not predict membership in the diminishing meaning class compared to the consistently high meaning class for men, whereas these factors were significant predictors of membership in the diminishing meaning class for women. Alternatively, intimate relationship satisfaction did not predict membership in the strengthening meaning class compared to the consistently high meaning class for women, whereas this factor was a significant predictor of membership in the strengthening meaning class for men. Finally, neither women nor men’s intimate relationship satisfaction was associated with odds of belonging to the diminishing meaning class compared to the consistently high meaning class.

Discussion

The current study provides a number of important insights regarding how meaning changes over time and what factors predict changes in meaning during significant life transitions, in this case, during military veterans’ transition from service to civilian life. Relatively little research has been conducted on meaning in life among veterans to date, and studies examining changes in meaning in life over time are rarer still. A key contribution of this work is demonstrating that the majority of veterans (89.5%) experience high levels of meaning despite going through the major life transition of leaving military service and needing to rebuild their identities within the civilian world. This finding is consistent with research in the civilian population, which has shown that even when faced with major life stressors most individuals experience high levels of meaning in life (Heintzelman & King, 2014). It is encouraging that most veterans endorsed having clear goals and aims, progression toward achieving life goals, and a purposeful and meaningful existence across the study period. Because meaning in life is associated with a multitude of positive outcomes, such as higher quality of life and greater use of adaptive coping strategies (Heintzelman & King, 2014; Steger, 2012), drawing on one’s sense of meaning could serve as a robust protective factor for this population.

Despite this positive finding, a substantial minority of veterans in the sample demonstrated less adaptive meaning trajectories following military service (10.5%). Veterans in both the diminishing and strengthening meaning groups initially reported below average levels of life meaning compared to those in the consistently high meaning group, indicating they were less likely to endorse having goals or aims in life, working toward life goals, and having purpose and meaning in their lives during the early transition period. That said, those in the strengthening meaning group appeared to experience some level of adaptive meaning-making and growth over the course of the study, such that their sense of meaning increased as time passed (although these increases were still not at the same level of those in the consistently high meaning group). Veterans of most concern are those who were classified into the diminishing meaning group (6.1%), who demonstrated below average levels of meaning initially that decreased further with time. This is troubling in part because low levels of meaning are a known risk factor for the development or worsening of a range of mental health concerns (Braden et al., 2015; Steger, 2012).

Findings also provided critical insight regarding predictors of change in meaning over time. The finding that veterans who endorsed greater posttraumatic stress symptoms, depression symptoms, and moral injury were at elevated odds of belonging to the diminished meaning and strengthening meaning classes (as compared to the consistently high meaning class) is consistent with the meaning-making model (Park, 2010). Essentially, these three risk factors may serve as additional disruptions to core beliefs that enable people to establish meaning in life in the context of a major life transition. Specifically, it is possible that veterans with greater levels of posttraumatic stress, depression, and moral injury were less successful in establishing their sense of the world as meaningful after leaving service because they did not have sufficient resources to navigate the added challenges introduced by a stressful life transition. This possibility is supported by prior research, which indicates that greater levels of these three factors are inversely associated with meaning (e.g., Fischer et al., 2020; Fischer et al., 2023a; Kelley et al., 2021). It is also noteworthy that these factors predicted both a diminishing and strengthening sense of meaning, which could be driven by veterans’ initially below average levels of meaning in both groups. Indeed, heightened symptoms in these areas may erode one’s sense of meaning by halting progress toward life goals and discouraging engagement in meaningful activities, influencing the ability to develop a sense of purpose during a major transition.

In contrast, the finding that veterans who reported greater psychological resilience, community relationship satisfaction, and intimate relationship satisfaction were at decreased odds of experiencing less adaptive trajectories aligns with prior work indicating that these factors can promote a significant sense of meaning (e.g., Fischer et al., 2023a; Niu et al., 2016). In keeping with the meaning-making model, the process of successful meaning making might be further facilitated by one’s access to intrapersonal and interpersonal resources such as those studied here. These protective factors might encourage more adaptive cognitive-emotional processing, allowing the stress associated with challenging life events to be transformed into positive emotions and meaning made (Park, 2010). Indeed, viewing these antecedents of meaning in life through a meaning-making lens, these findings enrich our understanding of protective factors that might counter stressful life experiences to positively influence how purposeful and meaningful individuals view their lives. This information can also serve to inform clinical intervention efforts focused on meaning-making, an important contribution given research showing that veterans who experience a loss of meaning are more likely to seek health care services from Veterans Affairs (Fontana & Rosenheck, 2005).

Lastly, while gender differences in identified classes and predictors of class membership were minimal overall (i.e., same class structure and similar predictors of meaning trajectories), some nuanced differences were observed. In particular, it appeared that there was some variability in the strength and curve of meaning trajectories among women and men. Notably, linear slope estimates were relatively stronger for women compared to men, and quadratic slope estimates were only significant for women. Relatedly, it appeared that women who belonged to a less adaptive meaning trajectory experienced some level of stabilization in meaning toward the end of the study period, whereas men continued to experience changes in meaning as time elapsed. Given the novelty of these findings, any detailed explanations of these differences are speculative, although one possibility is that women’s greater family and caregiving responsibilities (on average) challenge their ability to dedicate the necessary time and effort to developing a stable sense of meaning in the context of a major life transition. Alternatively, men veterans may experience unique challenges associated with greater exposure to combat and other unexpected traumatic events. In this way, the transition from military to civilian life could bring to the surface concerns that invariably impact their sense of meaning if left unaddressed. Future research is needed to further explore and confirm the cause of these observed gender differences.

With respect to predictors, posttraumatic stress symptoms, moral injury, and intimate relationship satisfaction differentially impacted meaning trajectories among women and men veterans. Specifically, posttraumatic stress symptoms and moral injury did not predict membership in the diminishing meaning class compared with the consistently high meaning class for men, whereas they did predict membership in this class for women. These findings point to the possibility that women might have been more impacted by the negative aftereffects of posttraumatic stress and moral injury compared to men in our sample. One reason for this finding might be that some women are adversely affected by societal perceptions that they are not “real veterans” or that they are not exposed to “real danger” compared to men veterans (Street et al., 2009). In this way, they may feel unsupported, invalidated, or unappreciated in the face of their symptoms and experiences, which may impact their self-identity and sense of meaning. In contrast, intimate relationship satisfaction did not predict membership in the strengthening meaning class compared to the consistently high meaning class for women, whereas it did predict membership in this class for men. This pattern of findings is consistent with research suggesting that men might derive more benefit than women from being in a supportive intimate relationship (Monin & Clark, 2011), which in turn might have implications for relationship satisfaction during a transitional period. In this way, it is possible that intimate relationship satisfaction might provide additional benefits for men veterans, such that they are able to grow their sense of meaning from initially lower levels post-separation. Finally, neither women nor men’s intimate relationship satisfaction was associated with odds of belonging to the diminishing meaning class compared to the consistently high meaning class—perhaps intimate relationship satisfaction was not influential enough to predict diminished meaning over time compared to other variables studied here, such as depression symptoms. Veterans may use other protective coping strategies (e.g., engagement in values-based action and mindfulness) or possess strong levels of psychological resilience and community relationship satisfaction (as seen here) that guard against diminished meaning, above and beyond satisfaction with their intimate relationship. Collectively, these data suggest there is value in deconstructing determinants of meaning by gender among veterans; however, these results should be interpreted with caution in light of their novelty. Replication of these findings is critical, and more research is needed to understand why these predictors serve unique functions for different people.

Limitations and Future Directions

Although the current study contributes to the study of life meaning, there are several limitations worth noting. To begin, the current study used a 4-item measure to explore veterans’ sense of meaning across time. Notably, this measure was utilized because of its strong psychometric properties, particularly its reliability and correlation with other commonly administered measures of meaning, in addition to its psychometric contribution above and beyond these common measures (Schulenberg et al., 2011). The shortened version of the measure also allowed for decreased participant burden over time, likely contributing to retention of veterans across the study period. However, some researchers have conceptualized meaning as a multidimensional construct reflecting multiple elements, such as coherence, purpose, and significance (Martela & Steger, 2016), which might not be fully captured by the current 4-item measure. Further, this measure did not link specific life domains to participants’ report of meaning (e.g., family, friends, vocation), which might have provided greater insight into individual differences in meaning over time. Taken together, future work in this area should aim to include a more comprehensive measure of meaning in life to expand upon our findings.

Relatedly, all measures in the current study relied on self-report, which might be influenced by one’s capacity to introspect and report objectively on emotional experiences, and we did not administer clinician-rated measures to supplement participant responses. Further, close-ended response options on each measure did not allow participants to describe specific behaviors or the context of their experiences. That said, veterans are in the best position to report on their functioning across multiple life domains, especially their own level of meaning in life (Heintzelman & King, 2014), and behaviorally-specific prompts in our measures of some variables address these limitations in part. Future researchers might consider strengthening their work in this area by complementing close-ended responses with open-ended prompts that allow participants to describe their experiences in detail, adding greater nuance to the literature.

Notably, our sample also lacked certain elements of diversity that might impact the generalizability of results, with the majority of participants identifying as White men. Although examination of this sample is important through its contribution to our knowledge about unique ways in which life meaning might change over time among veterans, future researchers should sample more racial, ethnic, and gender minorities, as well as other subsets of veterans, in order to assess whether findings in the current study are consistent among those with intersecting minoritized identities. Recruiting more diverse samples would also allow for additional nuanced analyses examining the evolving nature of life meaning across different groups of individuals.

In addition, it should be noted that we limited our analyses to time points that would allow us to examine prospective predictors of meaning in life (i.e., predictors at T1–T2, meaning at T3–T6). We view this as a strength, as many prior studies of positive psychology constructs have assessed predictors and outcomes concurrently, precluding conclusions about temporality. That said, balancing time constraints and favoring retention at each point in the larger study, we did not administer the meaning in life measure at T1 or T2, and therefore could not conduct additional analyses using these time points. It is possible that there is more variability in meaning to be captured closer to separation, and thus future researchers should consider examining meaning and related constructs in closer proximity to this major life transition. Gathering data on meaning in life from the onset of separation would also allow for more robust research questions related to change. For example, doing so would allow for an examination of how changes in each predictor correspond to changes in meaning from the time of separation.

Finally, we encourage researchers to examine other factors that might impact trajectories of meaning in life following this major transition. Building from the work conducted by Park et al. (2023), demographic variables such as age, income, separation status, and relationship status are worth continued examination. Relatedly, we only included veterans who indicated currently being in an intimate relationship in the analyses examining relationship satisfaction as a predictor of meaning trajectories. That said, of the 7,852 veterans in the sample, the majority (84%; n = 6,600) reported being in a relationship and thus were retained. Although this means some veterans were not included, we still found this factor important to examine given prior research supporting the positive link between intimate relationship satisfaction and well-being among veterans (e.g., Vogt et al., 2019). Taken together, it is important to continue investigating how both partnered and non-partnered veterans might achieve a sense of meaning following military separation given our findings. Future studies might also consider exploring factors that promote adaptive life meaning trajectories at both the intrapersonal and interpersonal level. Based on recent research (e.g., Fischer et al., 2023a), candidates might include gratitude, optimism, and openness to experience. Uncovering personal strengths that individuals can draw on to promote optimal functioning during stressful life transitions may serve to inform the development and refinement of intervention programs for those struggling to find meaning in their lives.

Clinical Implications and Conclusion

To our knowledge, this is the first study to examine veterans’ mental health and satisfaction with interpersonal relationships as prospective predictors of life meaning through the transition and reintegration period. These findings have important implications for clinical intervention. The finding that a substantial minority of veterans in the current sample demonstrated less adaptive meaning trajectories following military service highlights the need for targeted approaches among those who are struggling to find meaning. For example, at the time of separation, veterans might be encouraged to identify their personal strengths and then use those strengths to serve a larger purpose (e.g., mentoring other veterans through the transition experience). Veterans might also find involvement in organizations that foster connection and serve to better the well-being of others beneficial, such as Habitat for Humanity and Team Red, White & Blue, as examples. For those who would benefit from structured treatment, clinical interventions such as goal-focused positive psychotherapy (Conoley & Scheel, 2018) use a strengths-based framework to assist individuals in building psychological resiliency and connectedness with others, which were both found to be important predictors of a greater sense of meaning in this study. Alternatively, veterans at risk for low meaning due to posttraumatic stress symptoms, depression symptoms, or experiences of moral injury might benefit from a referral to empirically-supported treatments such as cognitive processing therapy (Resick et al., 2017). Through active engagement with these types of services, veterans may gain more mental bandwidth to cultivate and enhance their sense of meaning.

More generally, findings also highlight the need to attend to positive indicators of mental health as well as indicators of ill-being to foster a more complete model of mental health among veterans and other populations. For example, efforts to nurture the personal strengths and interpersonal relationships of veterans may not only decrease their suffering but also increase their well-being. This balanced perspective is important to consider given prior research suggesting that individuals who do not experience mental illness but have a low sense of meaning are at risk for long-term impairment and disability (see Keyes, 2007 for a review).

Victor Frankl (1946) wrote, “There is nothing in the world, I venture to say, that would so effectively help one to survive even the worst conditions as the knowledge that there is meaning in one’s life.” Indeed, enhancing one’s meaning has the potential to prevent poor longer-term well-being outcomes that sometimes follow from stressful life experiences, including major life transitions. For this reason, greater attention to factors that improve veterans’ post-military adjustment is warranted, in addition to research examining the efficacy of intervention efforts aiming to bolster life meaning in this group. In this way, we can gain a better understanding of how to strengthen veterans’ resiliency across multiple stages of their lives.

Funding Statement:

This research drew from data collected as part of a previously funded project managed by the Henry M. Jackson Foundation for the Advancement of Military Medicine, Inc. (HJF), and collaboratively sponsored by the Bob Woodruff Foundation, Health Net Federal Services, The Heinz Endowments, HJF, Lockheed Martin Corporation, May and Stanley Smith Charitable Trust, National Endowment for the Humanities, Northrop Grumman, Philip and Marge Odeen, Prudential, Robert R. McCormick Foundation, Rumsfeld Foundation, Schultz Family Foundation, Walmart Foundation, Wounded Warrior Project, Inc., and the Veterans Health Administration Health Services Research and Development Service (Co-PI: Vogt). Shaina A. Kumar was supported by a T32 award from the National Institute of Mental Health (T32MH019836). The views expressed are those of the authors and do not necessarily represent the views or policy of the Department of Veterans Affairs or National Institute of Mental Health.

Footnotes

Conflict of Interest Disclosure: The authors have no conflicts of interest to disclose.

Ethics Approval Statement: This study was approved by the local Institutional Review Boards, and all participants provided informed consent.

Data Availability Statement:

De-identified data specific to the current paper are available from the last author upon reasonable request, pending approval from the Department of Veterans Affairs.

References

  1. Altmaier EM (2020). Meaning in life amidst life transitions. In Altmaier EM (Ed.), Navigating life transitions for meaning (pp. 3–12). Elsevier. [Google Scholar]
  2. Asparouhov T, & Muthén B (2014). Auxiliary variables in mixture modeling: Three-step approaches using Mplus. Structural Equation Modeling, 21(3), 329–341. 10.1080/10705511.2014.915181 [DOI] [Google Scholar]
  3. Bovin MJ, Kimerling R, Weathers FW, Prins A, Marx BP, Post EP, & Schnurr PP (2021). Diagnostic accuracy and acceptability of the Primary Care Posttraumatic Stress Disorder Screen for the Diagnostic and Statistical Manual of Mental Disorders (Fifth Edition) among US veterans. JAMA Network Open, 4(2), 1–11. 10.1001/jamanetworkopen.2020.36733 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Braden A, Overholser J, Fisher L, & Ridley J (2015). Life meaning is associated with suicidal ideation among depressed veterans. Death Studies, 39(1), 24–29. 10.1080/07481187.2013.871604 [DOI] [PubMed] [Google Scholar]
  5. Conoley CW, & Scheel MJ (2018). Goal focused positive psychotherapy. Oxford. [DOI] [PubMed] [Google Scholar]
  6. Coughlin SS, Aliaga P, Barth S, Eber S, Maillard J, Mahan C, Kang HK, Schneiderman A, DeBakey S, Vanderwolf P, & Williams M (2011). The effectiveness of a monetary incentive on response rates in a survey of recent U.S. veterans. Survey Practice, 4(1), 1–8. 10.29115/SP-2011-0004 [DOI] [Google Scholar]
  7. Curran PJ, Obeidat K, & Losardo D (2010). Twelve frequently asked questions about growth curve modeling. Journal of Cognition and Development, 11(2), 121–136. 10.1080/15248371003699969 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Dezutter J, Luyckx K, & Wachholtz A (2015). Meaning in life in chronic pain patients over time: Associations with pain experience and psychological well-being. Journal of Behavioral Medicine, 38(2), 384–396. 10.1007/s10865-014-9614-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Dillman DA, Smyth JD, & Christian LM (2014). Internet, phone, mail, and mixed-mode surveys: The tailored design method (4th ed.). Wiley & Sons. [Google Scholar]
  10. Fischer IC, Feldman DB, Tsai J, Harpaz-Rotem I, Lucas KA, Schulenberg SE, & Pietrzak RH (2023a). Identifying significant correlates of purpose in life in older US military veterans: Results from the National Health and Resilience in Veterans Study. International Psychogeriatrics, 35(10), 560–565. 10.1017/S1041610222001223 [DOI] [PubMed] [Google Scholar]
  11. Fischer IC, Nichter B, Feldman DB, Na PJ, Tsai J, Harpaz-Rotem I, Schulenberg SE, & Pietrzak RH (2023b). Purpose in life protects against the development of suicidal thoughts and behaviors in US veterans without a history of suicidality: A 10-year, nationally representative, longitudinal study. Journal of Affective Disorders, 340, 551–554. 10.1016/j.jad.2023.08.040 [DOI] [PubMed] [Google Scholar]
  12. Fischer IC, Shanahan ML, Hirsh AT, Stewart JC, & Rand KL (2020). The relationship between meaning in life and post-traumatic stress symptoms in US military personnel: A meta-analysis. Journal of Affective Disorders, 277, 658–670. 10.1016/j.jad.2020.08.063 [DOI] [PubMed] [Google Scholar]
  13. Fischer IC, Tsai J, Harpaz-Rotem I, McCutcheon VE, Schulenberg SE, & Pietrzak RH (2023c). Perceived purpose in life, mental health, and suicidality in older US military veterans: Results from the National Health and Resilience in Veterans Study. The American Journal of Geriatric Psychiatry, 31(2), 87–93. 10.1016/j.jagp.2022.09.010 [DOI] [PubMed] [Google Scholar]
  14. Fontana A, & Rosenheck R (2005). The role of loss of meaning in the pursuit of treatment for posttraumatic stress disorder. Journal of Traumatic Stress, 18(2), 133–136. 10.1002/jts.20014 [DOI] [PubMed] [Google Scholar]
  15. Frankl VE (1946). Man’s search for meaning. Beacon Press. [Google Scholar]
  16. Heintzelman SJ, & King LA (2014). Life is pretty meaningful. American Psychologist, 69(6), 561–574. 10.1037/a0035049 [DOI] [PubMed] [Google Scholar]
  17. Hipp JR, & Bauer DJ (2006). Local solutions in the estimation of growth mixture models. Psychological Methods, 11(1), 36–53. 10.1037/1082-989X.11.1.36 [DOI] [PubMed] [Google Scholar]
  18. Jung T, & Wickrama AS (2008). An introduction to latent class growth analysis and growth mixture modeling. Social and Personality Psychology Compass, 2(1), 302–317. 10.1111/j.1751-9004.2007.00054.x [DOI] [Google Scholar]
  19. Kelley ML, Chae JW, Bravo AJ, Milam AL, Agha E, Gaylord SA, Vinci C, & Currier JM (2021). Own soul’s warning: Moral injury, suicidal ideation, and meaning in life. Psychological Trauma: Theory, Research, Practice, and Policy, 13(7), 740–748. 10.1037/tra0001047 [DOI] [PubMed] [Google Scholar]
  20. Keyes CLM (2007). Promoting and protecting mental health as flourishing: A complementary strategy for improving national mental health. American Psychologist, 62(2), 95–108. 10.1037/0003-066X.62.2.95 [DOI] [PubMed] [Google Scholar]
  21. King LA, & Hicks JA (2021). The science of meaning in life. Annual Review of Psychology, 72, 561–584. 10.1146/annurev-psych-072420-122921 [DOI] [PubMed] [Google Scholar]
  22. Krause N, & Hayward RD (2014). Assessing stability and change in a second-order confirmatory factor model of meaning in life. Journal of Happiness Studies, 15(2), 237–253. 10.1007/s10902-013-9418-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kroenke K, Spitzer RL, & Williams JB (2003). The Patient Health Questionnaire-2: Validity of a two-item depression screener. Medical Care, 41(11), 1284–1292. 10.1097/01.MLR.0000093487.78664.3C [DOI] [PubMed] [Google Scholar]
  24. Lubke G, & Muthén BO (2007). Performance of factor mixture models as a function of model size, covariate effects, and class-specific parameters. Structural Equation Modeling, 14(1), 26–47. 10.1080/10705510709336735 [DOI] [Google Scholar]
  25. Martela F, & Steger MF (2016). The three meanings of meaning in life: Distinguishing coherence, purpose, and significance. The Journal of Positive Psychology, 11(5), 531–545. 10.1080/17439760.2015.1137623 [DOI] [Google Scholar]
  26. McKay S, Skues JL, & Williams BJ (2021). Does the Brief Resilience Scale actually measure resilience and succumbing? Comparing artefactual and substantive models. Advances in Mental Health, 19(2), 192–201. 10.1080/18387357.2019.1688667 [DOI] [Google Scholar]
  27. Meyer JP, & Morin AJS (2016). A person-centered approach to commitment research: Theory, research, and methodology. Journal of Organizational Behavior, 37(4), 584–612. 10.1002/job.2085 [DOI] [Google Scholar]
  28. Monin JK, & Clark MS (2011). Why do men benefit more from marriage than do women? Thinking more broadly about interpersonal processes that occur within and outside of marriage. Sex Roles, 65, 320–326. 10.1007/s11199-011-0008-3 [DOI] [Google Scholar]
  29. Muthén LK & Muthén BO (1998-2017). Mplus user’s guide (8th ed.). Muthén & Muthén. [Google Scholar]
  30. Nagin DS (2010). Group-based trajectory modeling: An overview. In: Piquero AR & Weisburd D (Eds.), Handbook of quantitative criminology (pp. 53–67). Springer. [Google Scholar]
  31. Nash WP, Marino Carper TL, Mills MA, Au T, Goldsmith A, & Litz BT (2013). Psychometric evaluation of the Moral Injury Events Scale. Military Medicine, 178(6), 646–652. 10.7205/MILMED-D-13-00017 [DOI] [PubMed] [Google Scholar]
  32. Niu CC, Huang HM, Hung YY, & Lee HL (2016). A study of interpersonal intimacy and meaning of life among elderly institutionalized veterans. Journal of Nursing Research, 24(4), 311–320. 10.1097/JNR.0000000000000130 [DOI] [PubMed] [Google Scholar]
  33. Nylund KL, Asparouhov T, & Muthén BO (2007). Deciding on the number of classes in latent class analysis and growth mixture modeling: A Monte Carlo simulation study. Structural Equation Modeling, 14(4), 535–569. 10.1080/10705510701575396 [DOI] [Google Scholar]
  34. Orazem RJ, Frazier PA, Schnurr PP, Oleson HE, Carlson KF, Litz BT, & Sayer NA (2017). Identity adjustment among Afghanistan and Iraq war veterans with reintegration difficulty. Psychological Trauma: Theory, Research, Practice, and Policy, 9(1), 4–11. 10.1037/tra0000225 [DOI] [PubMed] [Google Scholar]
  35. Park CL (2010). Making sense of the meaning literature: An integrative review of meaning making and its effects on adjustment to stressful life events. Psychological Bulletin, 136(2), 257–301. 10.1037/a0018301 [DOI] [PubMed] [Google Scholar]
  36. Park CL, Sacco SJ, Kraus SW, Mazure CM, & Hoff RA (2023). Demographic, deployment and post-deployment experiences predict trajectories of meaning in life in OEF/OIF/OND veterans. The Journal of Positive Psychology, 18(1), 167–177. 10.1080/17439760.2021.1991452 [DOI] [Google Scholar]
  37. Prins A, Bovin MJ, Smolenski DJ, Marx BP, Kimerling R, Jenkins-Guarnieri MA, Kaloupek DG, Schnurr PP, Kaiser AP, Leyva YE, & Tiet QQ (2016). The Primary Care PTSD Screen for DSM-5 (PC-PTSD-5): Development and evaluation within a veteran primary care sample. Journal of General Internal Medicine, 31(10), 1206–1211. 10.1007/s11606-016-3703-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Resick PA, Monson CM, & Chard KM (2017). Cognitive processing therapy for PTSD: A comprehensive manual. Guilford. [Google Scholar]
  39. Szabó D, Békés V, Lévay EE, Salgó E, & Unoka ZS (2023). Moral injury in psychiatric patients with personality and other clinical disorders: Development, psychometric properties, and validity of the Moral Injury Events Scale-Civilian Version. European Journal of Psychotraumatology, 14(2), 1–13. 10.1080/20008066.2023.2247227 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Schulenberg SE, Schnetzer LW, & Buchanan EM (2011). The Purpose in Life Test-Short Form: Development and psychometric support. Journal of Happiness Studies, 12(5), 861–876. 10.1007/s10902-010-9231-9 [DOI] [Google Scholar]
  41. Street AE, Vogt D, & Dutra L (2009). A new generation of women veterans: Stressors faced by women deployed to Iraq and Afghanistan. Clinical Psychology Review, 29(8), 685–694. 10.1016/j.cpr.2009.08.007 [DOI] [PubMed] [Google Scholar]
  42. Smith BW, Dalen J, Wiggins K, Tooley E, Christopher P, & Bernard J (2008). The Brief Resilience Scale: Assessing the ability to bounce back. International Journal of Behavioral Medicine, 15(3), 194–200. 10.1080/10705500802222972 [DOI] [PubMed] [Google Scholar]
  43. Sokol Y, Gromatsky M, Edwards ER, Greene AL, Geraci JC, Harris RE, & Goodman M (2021). The deadly gap: Understanding suicide among veterans transitioning out of the military. Psychiatry Research, 300, 1–8. 10.1016/j.psychres.2021.113875 [DOI] [PubMed] [Google Scholar]
  44. Steger MF (2012). Experiencing meaning in life: Optimal functioning at the nexus of well-being, psychopathology, and spirituality. In Wong PTP (Ed.), The human quest for meaning: Theories, research, and applications (pp. 165–184). Routledge. [Google Scholar]
  45. Steger MF, Frazier P, Oishi S, & Kaler M (2006). The meaning in life questionnaire: Assessing the presence of and search for meaning in life. Journal of Counseling Psychology, 53(1), 80–93. 10.1037/0022-0167.53.1.80 [DOI] [Google Scholar]
  46. Steger MF, & Kashdan TB (2007). Stability and specificity of meaning in life and life satisfaction over one year. Journal of Happiness Studies, 8(2), 161–179. 10.1007/s10902-006-9011-8 [DOI] [Google Scholar]
  47. Steger MF, Shin JY, Shim Y, & Fitch-Martin A (2013). Is meaning in life a flagship indicator of well-being? In Waterman AS (Ed.), The best within us: Positive psychology perspectives on eudaimonia (pp. 159–182). American Psychological Association. 10.1037/14092-009 [DOI] [Google Scholar]
  48. Tabachnick BG, & Fidell LS (1996). Using multivariate statistics (3rd ed.). HarperCollins. [Google Scholar]
  49. Vogt D, Borowski SC, Godier-McBard LR, Fossey MJ, Copeland LA, Perkins DF, & Finley EP (2022). Changes in the health and broader well-being of U.S. veterans in the first three years after leaving military service: Overall trends and group differences. Social Science & Medicine, 294, 1–10. 10.1016/j.socscimed.2022.114702 [DOI] [PubMed] [Google Scholar]
  50. Vogt D, King MW, Borowski S, Finley EP, Perkins DF, & Copeland LA (2021). Identifying factors that contribute to military veterans’ post-military well-being. Applied Psychology: Health and Well-Being, 13(2), 341–356. 10.1111/aphw.12252 [DOI] [PubMed] [Google Scholar]
  51. Vogt D, Perkins DF, Copeland LA, Finley EP, Jamieson CS, Booth B, Lederer S, & Gilman CL (2018). The Veterans Metrics Initiative study of US veterans’ experiences during their transition from military service. British Medical Journal Open, 8(6), e020734. 10.1136/bmjopen-2017-020734 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. Vogt D, Taverna EC, Nillni YI, Booth B, Perkins DF, Copeland LA, Finley EP, Tyrell FA, & Gilman CL (2019). Development and validation of a tool to assess military veterans’ status, functioning, and satisfaction with key aspects of their lives. Applied Psychology: Health and Well-Being, 11(2), 328–349. 10.1111/aphw.12161 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

De-identified data specific to the current paper are available from the last author upon reasonable request, pending approval from the Department of Veterans Affairs.

RESOURCES