Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Mar 31.
Published in final edited form as: Psychol Trauma. 2024 Nov 14;17(5):931–940. doi: 10.1037/tra0001810

Military sexual trauma, childhood trauma and combat trauma: Associations with longitudinal posttraumatic growth among U.S. veterans

Liv M Canning a, Jordan P Davis b, John J Prindle a, Carl A Castro a, Eric R Pedersen c, Shaddy K Saba a, Adrian J Bravo d, Reagan E Fitzke e, Alexandra H Mills f, Whitney S Livingston b
PMCID: PMC13035283  NIHMSID: NIHMS2056885  PMID: 39541538

Abstract

Objective:

Military sexual trauma, childhood trauma, and combat trauma are prevalent among U.S. military personnel. Cumulative trauma exposure may hinder posttraumatic growth, a positive psychological change following traumatic events, while social support can facilitate this growth. Understanding the influence of these traumas and social support on longitudinal posttraumatic growth is crucial.

Methods:

We assessed 1,230 veterans at 6, 9, 12, 18, and 24 months post-initial survey. Latent class analysis (LCA) identified trauma experience classes, and latent growth models examined posttraumatic growth trajectories, incorporating social support as a time-varying covariate.

Results:

The LCA revealed four classes: high trauma exposure, moderate childhood trauma – moderate combat trauma, high military sexual trauma – moderate combat trauma, and combat trauma only. Veterans in the combat-only class reported significant posttraumatic growth. The moderate childhood trauma – moderate combat class exhibited consistently low growth. Veterans in the high military sexual trauma – moderate combat class showed slightly higher initial growth, but no significant change over time. The high trauma exposure class experienced a significant decline in growth.

Conclusions:

Social support significantly predicted posttraumatic growth, with varying impacts across trauma classes. Interventions could be vital for survivors of military sexual trauma, childhood trauma, or compounded traumas to enhance posttraumatic growth among military veterans.

Keywords: combat, military, substance use disorder, PTSD, longitudinal

Introduction

Military personnel report high rates of trauma exposure (Beguin-Fernald, 2022), with some studies noting upwards of 65% reporting one or more traumatic events and as many as 39% reporting two or more events (Hourani, 2003). These events can include early childhood trauma (e.g., physical, sexual, and emotional abuse), experiences while deployed (e.g., combat-related trauma), and experiences of military sexual trauma. Prior work reports that veterans who have experienced trauma are at heightened risk of mental health and substance use disorder symptomology (Aronson et al., 2020; Davis et al., 2022). However, literature suggests that traumatic experiences can also catalyze positive change, known as posttraumatic growth. Posttraumatic growth is “positive psychological change experienced as a result of the struggle with highly challenging life circumstances” (Tedeschi & Calhoun, 2004, p. 1). Nevertheless, the degree to which people maintain and report changes in posttraumatic growth and the factors that support these changes remain unclear (Tedeschi et al., 2018).

Posttraumatic growth has been well-documented throughout history (Tedeschi et al., 2018), yet it has only been studied scientifically within the past 15–20 years (Calhoun & Tedeschi, 2006). Typically thought of as positive social, spiritual, or psychological growth after trauma, posttraumatic growth is divided into five categories: personal strength, relating to others, appreciation of life, openness to new possibilities, and spiritual change (Calhoun & Tedeschi, 2006; Mark et al., 2018). Unlike resiliency, which focuses on an ability to adapt or “bounce back” to pre-trauma levels of functioning after traumatic experiences (Smith et al., 2008), posttraumatic growth looks at the ability of an individual to achieve even higher levels of functioning than before the traumatic event. Nevertheless, the mere experience of adversity alone does not always lead to posttraumatic growth – thus, understanding factors associated with this posttraumatic growth is necessary.

In particular, understanding posttraumatic growth over time, regardless of when the trauma occurred, is essential. Firstly, posttraumatic growth is not a static process; it evolves and unfolds over time (Tedeschi et al., 2018). Additionally, longitudinal studies offer insights into the cumulative impact of trauma exposure on posttraumatic growth. Individuals who have experienced multiple traumatic events may exhibit distinct patterns of growth compared to those with single-event traumas (Kira et al., 2013). Examining posttraumatic growth longitudinally allows researchers to assess how the accumulation of trauma over time influences the trajectory and magnitude of growth. In addition, individual factors may impact posttraumatic growth over time. Individuals may experience fluctuations in their perceptions of growth, influenced by various factors such as life events, social support, and coping strategies (Tedeschi et al., 2018).

One area of research that has received considerable attention is the association between social support and posttraumatic growth among veteran and civilian populations (Laffaye et al., 2008). The Functional Descriptive Model posits that social support is a critical facilitator in reconstructing new belief systems, providing comfort and relief during the construction of new, post-trauma schemas (Tedeschi & Calhoun, 1996). In a recent meta-analysis on the association between social support and posttraumatic growth, Ning and colleagues (2022) reported a moderate correlation between social support and posttraumatic growth across 217 studies. Though fewer, some research has explored the associations between social support and posttraumatic growth among veteran samples (Lee et al., 2015; Maguen et al., 2006; Nordstrand et al., 2020; Pietrzak et al., 2010). For example, Nordstrand and colleagues (2020) noted that social support was associated with greater posttraumatic growth following war-related stressors. Similarly, Lee and colleagues (2015) found that social support was crucial for the posttraumatic growth domain of relating to others in a qualitative sample of older veterans. Pietrzak and colleagues (2010) found that being younger and having high perceived support within one’s military unit were positively associated with posttraumatic growth. Unfortunately, these studies utilized cross-sectional approaches that limit our ability to explore how social support might influence changes in posttraumatic growth and, further, if this influence varies by trauma type.

Prior work investigating both short and long-term effects of trauma exposure often assess the effect of a single event or experience in isolation (Brooks et al., 2023). Increasingly, research has moved away from a sole focus on differences in outcomes by trauma type and, instead, has begun to identify exposure to multiple types of traumatic experiences (i.e., poly victimization). Poly-victimization theory refers to the effects of experiencing multiple types of trauma resulting in trauma symptoms, behavioral and psychological problems, and other adverse outcomes (Finkelhor et al., 2007; Haahr-Pedersen et al., 2020; Mitchell et al., 2020). Recent literature suggests that while some survivors of multiple traumatic events can experience posttraumatic growth (Brooks et al., 2017, 2019; Jirek & Saunders, 2018; Nuccio & Stripling, 2021), others may become overwhelmed, which can inhibit posttraumatic growth (Butler, 2005; Brooks et al., 2021). For example, in a sample of university students, Jirek and Saunders (2018) found poly-victimization to be positively associated with posttraumatic growth. However, when Brooks and colleagues (2021) qualitatively interviewed participants regarding growth after trauma, they found that several participants reported feeling their previous trauma exposures hampered their ability to grow from subsequent trauma exposures. Understanding the association between poly-victimization and posttraumatic growth is imperative among highly traumatized populations, such as veterans.

To fully understand how posttraumatic growth emerges, it is essential to explore variations in posttraumatic growth for multiple trauma types (Thomas et al., 2021). For example, one study explored the association between posttraumatic growth across multiple trauma types, including survivors of sexual assault, motor vehicle accidents, and those experiencing sudden loss (Shakespeare-Finch & Armstrong, 2010). Those experiencing sudden loss reported higher perceived growth than those in the sexual assault group. Few studies, however, have explored how multiple trauma exposures relate to longitudinal changes in posttraumatic growth. As there is individual variability in the timing and trajectories of posttraumatic growth over time, capturing these changes with a longitudinal design is essential. It may be that specific trauma typologies relate to different trajectories of posttraumatic growth, which may provide needed information on how to support positive growth following traumatic events.

Present Study

Though the number of longitudinal studies of posttraumatic growth is increasing (Maitlis, 2020), research primarily focuses on outcomes after a single or a limited range of adverse events (Brooks et al., 2023). The present study expands this research by exploring the effects of multiple trauma exposures (childhood trauma and military sexual trauma) on changes in posttraumatic growth and how social support may catalyze posttraumatic growth in a veteran population. As such, the present study had three aims. The first aim was to extract heterogeneity in experiences of combat trauma, military sexual trauma and childhood trauma among military veterans using latent class analysis (LCA). The second aim was to understand how changes in posttraumatic growth vary by emergent childhood trauma and military sexual trauma classes. The third aim, in line with the functional descriptive model, was to understand how social support is associated with changes in posttraumatic growth and how these associations may vary by emergent childhood trauma and military sexual trauma classes.

Methods

Procedures

Veterans were recruited via social media advertisements in February 2020 as part of a survey study of veteran attitudes and health behavior. Eligibility criteria for the study were (1) aged 18 to 40 and (2) having separated from the Air Force, Army, Marine Corps, or Navy. Since recruitment was completed solely online, we implemented several verification checks (see Davis et al., 2021) to ensure, to the best of our ability, that participants were not misrepresenting themselves. Checks included noting if participants completed surveys in an impossibly short time, utilizing “insider knowledge” military-related questions that required consistent responses across multiple surveys, and checking to confirm that participants did not complete the survey more than once. The final sample was composed of 1,230 veteran participants at baseline. Participants were sent follow-up surveys via email at 6-months (N = 1,025; 83.3% retention from baseline), 9-months (N = 1,006; 81.8%), 12-months (N = 1,005; 81.7%), 18-months (N = 976; 79.3%), and 24 months (N = 952; 77.4%) after baseline. They received gift cards (ranging from $30 to $50) for completing the baseline and follow-up surveys (see Davis et al., 2021 for more details on participant recruitment and study procedures). The present study uses four waves of data from the 6-month follow-up forward as posttraumatic growth was not assessed at baseline (analytic sample N = 1,025). USC IRB approved all procedures, and all participants provided informed consent.

Participants

Table 1 provides detailed information on participant demographics and the prevalence of trauma typologies. Overall, participants were, on average, 34.5 (SD = 3.67) years old, with 88.7% (n = 1,091) identifying as male. Most of the sample was White (79.3%, n = 975), with 7.3% (n = 90) identifying as Black, 10.9% (n = 134) as Hispanic, and the remainder identified as members of other minority groups (2.5%; n = 31). Veterans served for a mean of 9.94 years (SD = 3.41) in the military. Veterans were out of the military for a mean of 4.66 (SD = 2.95) years when they took the baseline survey in August/September 2020. Most (70%) were within 2–5 years since discharge when they took the baseline survey. At baseline, 292 of the 1025 (28.5%) who completed the survey reported any past 6-month use of mental health or substance use disorder services. In terms of childhood trauma, 13.3% (n = 164) reported sexual abuse, 16.7% (n = 205) reported physical abuse, and 27% (n = 333) reported emotional abuse. Rates of sexual violence or sexual harassment were similar but with harassment more prevalent, ranging from 13% (sexual violence: attempted penetration) to 53.7% (sexual harassment: nonconsensual sharing, showing, or taking pictures). Of the total sample, 94.1% (n = 1,158) reported experiencing combat trauma.

Table 1.

Participant demographics

Variable M(SD) or N(%)
Age 34.5 (3.67)
Sex (male) 1,091 (88.7%)
Race/ethnicity
 White 975 (79.3%)
 Hispanic/Latino/a 134 (10.9%)
 Black 90 (7.3%)
 Asian 13 (1.1%)
 Multiracial/other 18 (1.5%)
Time in the military 9.94 (3.41)
Time out of military 4.66 (2.95)
Combat severity 5.02 (2.35)
Trauma typologies
Combat trauma 5.02 (2.35)
Combat trauma (any) 1158 (94.1%)
Childhood trauma (any) 378 (30.7%)
 Sexual abuse 164 (13.3%)
 Physical abuse 205 (16.7%)
 Emotional abuse 333 (27.1%)
 Witness parent violence 166 (13.5%)
Military sexual trauma (any) 784 (63.7%)
 Nonconsensual penetration 208 (16.9%)
 Nonconsensual attempted penetration 160 (13.0%)
 Nonconsensual touching 241 (19.6%)
 Unwelcomed comments, gestures, or jokes 247 (20.1%)
 Nonconsensual flashing or exposing 198 (16.1%)
 Nonconsensual sharing, showing, or taking
pictures		 660 (53.7%)
Posttraumatic growth 66.5 (19.8)
Social support 3.12 (0.56)
Open in a new tab

Measures

Demographics.

Participants self-reported their biological sex, race/ethnicity, age, and a summed score of combat experiences (Schell & Marshall, 2008).

Childhood trauma.

Five items from the adverse childhood experiences questionnaire (Felitti et al., 2019) were used to assess childhood trauma. Each item asked participants whether a specific traumatic event happened to them (yes/no) before their 18th birthday. The five items are: “Did a parent or other adult often or very often swear at you, insult you, put you down, or humiliate you,” (emotional abuse) “Did a parent or other adult often push, grab, slap, or ever hit you,” (physical abuse) “Did an adult or person at least 5 years older than you fondle you, have you touch their body, or attempt to have intercourse with you?” (sexual abuse) “Were any of your parents or step-parents often or very often pushed, grabbed, slapped, or had something thrown at them, or kicked, bitten, hit with a fist, or hit with something hard or ever repeatedly hit over at least a few minutes or threatened with or hurt by a knife or gun” (witnessing parental violence) “Did you often or very often feel that no one in your family loved you or thought you were important or special, or your family didn’t look out for each other, feel close to each other, or support each other?” (neglect). Three dichotomous items were created to indicate if the participant had ever experienced emotional abuse (created from three items), physical abuse, or sexual abuse.

Military sexual trauma.

In line with the United States Department of Veteran Affairs definition (2021), military sexual trauma was conceptualized as sexual violence and sexual harassment victimization during one’s time in the military. Sexual violence was assessed using five items from the 2014 RAND Military Workplace Study (Morral et al., 2014) to capture nonconsensual sexual violence, including physical assault and attempted physical assault, while the individual was in the military. Two items assessed nonconsensual penetration of the mouth, anus, or vagina; one assessed attempted unsuccessful penetration; and two assessed unwanted sexual touching of one’s private parts without penetration. Sexual harassment during military service was assessed using five items from the Campus Climate Survey Validation Study conducted by the Bureau of Justice Statistics and RTI International (Krebs et al., 2016). One item assessed unwanted sexual advances, jokes, or gestures; one item assessed nonconsensual exposing of private parts; and three items assessed the spreading of sexual rumors or nonconsensual sharing of pictures, videos, or sexual rumors that the individual did not want to see or did not want to be shared. Two dichotomous items were created to indicate if the participant had ever experienced sexual violence or sexual harassment during their time in the military.

Combat trauma.

Severity of combat trauma was assessed at baseline using a measure of 11 items from prior work with veterans (Schell & Marshall, 2008). Example items are: “Having a blow to the head from any accident or injury”, “Engaging in hand-to-hand combat”, and “ever feeling like they were in great danger of being killed”. Participants responded yes/no to each of the 12 items.

Posttraumatic growth.

Twenty-one items from the Posttraumatic Growth Inventory (Tedeschi & Calhoun, 1996) were used to assess posttraumatic growth. Each included statement relates to the five factors of posttraumatic growth. Participants indicated their scores on a six point scale ranging from “0 = I did not experience this as a result of this crisis” to “5 = I experienced this change to a very great degree as a result of my crisis.” Responses were summed to create a composite mean posttraumatic growth score for each participant.

Social support.

Nineteen items from the Social Support Survey Instrument (Sherbourne & Stewart, 1991) were used to assess social support. Participants were prompted to respond on a five-point Likert scale ranging from “1 = none of the time” to “5 = all of the time” to the prompt: “How often is each of the following kinds of support available to you if you need it?” Example items include “Show love and affection,” “Take to Doctor,” “Give you good advice,” “Do something enjoyable with.” Responses were summed to create a composite mean social support score for each participant.

Analytic Plan

We employed a multiple-step model-building process to address our study aims. Supplemental Figure 1 displays a conceptual layout of our final model. To address our first aim, we utilized latent class analysis (LCA) in Mplus version 8.6 (Muthén & Muthén, 1998). LCA is a technique that identifies heterogeneity within a sample and classifies individuals based on the probability of item endorsement. Within our LCA model, we entered childhood trauma items associated with military sexual trauma, and combat trauma. We fit models ranging from one to six classes and examined fit statistics to determine if adding a class improved model fit. We used decreases in Bayesian Information Criteria (BIC) and the sample size adjusted Bayesian Information Criteria (aBIC). to assess model fit. Class selection is based on BIC or aBIC values increasing from the prior model, indicating that a k-1 class better fits the data. Further, we utilized non-significant Vuong-Lo-Mendell-Rubin Likelihood Ratio Test (VLRT), the Lo-Mendell-Rubin adjusted likelihood ratio test (LRT), and the bootstrapped likelihood ratio test (BLRT) to indicate whether a k – 1 class solution is a better fit to the data. Furthermore, following the suggestions on class size (Nylund et al., 2007), solutions with no classes comprising less than 5% of the study sample were prioritized over those with smaller classes to avoid stability issues. Once the best fitting class solution was determined, we followed best practice methods outlined by Nylund Gibson in which classification errors for each individual are computed, and the inverse logits are used as weights rather than using the modal class assignment as an imperfect latent class indicator. The advantage of this method of setting logits for the final class solution is that it is more resistant to class shifts when including the emergent latent classes in further modeling steps (Nylund-Gibson et al., 2019). Once the best-fitting class structure was determined, we estimated a latent growth model for posttraumatic growth for each emergent class. Here, we allowed each class to have its own mean intercept, mean slope, variance, co-variance, and residual co-variances. During the model fitting process, we determined that the best-fitting model was one in which each class had a random intercept and a fixed slope. Once the final model was established, we performed model constraint tests between classes on all intercept and slope estimates. Doing this allows us to determine if classes differ significantly in their starting values (e.g., intercepts) and how posttraumatic growth changes over time (e.g., slopes).

Finally, we introduced social support as a time-varying co-variate in our model to address our final aim. We regressed our social support time-varying co-variate onto the contemporaneously observed posttraumatic growth scores for each emergent class at each time point. Doing so allowed us to determine the effect of social support on posttraumatic growth over time and how this effect varied by emergent childhood trauma and military sexual trauma classes. We first introduced our social support time-varying co-variate as a freely estimated predictor within each class (e.g., the time-varying co-variate was allowed to be freely estimated over time across all emergent classes). We then tested this model against one where our social support time-varying co-variate was systematically constrained to be equal over time within classes. For example, constraining effects of social support to be equal within Class 1 (compared to the model in which it was freely estimated across all classes), then constraining effects of social support to be equal in Class 2 and comparing to the previous model (constrained social support in Class 1), constraining effects of social support to be equal within Class 3 and comparing to the previous model (constrained social support in Class 2) so on and so forth. This process was repeated for all emergent classes until all constraints were tested. We used the difference in negative two log-likelihood ratio tests to determine if the constrained versus freely estimated model fit the data best. This process allowed us to test if each of our time-varying co variates had consistent, stronger, or weaker influences over time within each class. A final model was estimated employing all decisions on constraints for each class. All models following Aim 1 (LCA) included our control variables (i.e., biological sex, race/ethnicity, and age). Missing data ranged from 0%−19% across time. To address missing data, we utilized maximum likelihood estimation techniques in Mplus.

Results

Emergent childhood trauma and military sexual trauma classes

Fit indices for the childhood trauma and military sexual trauma LCA can be found in Supplemental Table 1. The non-significant VLRT, LRT, and BLRT values for the seven-class solution indicated that a six-class solution might be the best-fitting model. However, upon reviewing the plotted classes, the six-class solution had two classes with nearly identical patterns of item endorsements and two classes with less than 5% of the sample, indicating this solution is likely unstable (Nagin, 2005). Given this, we plotted the five-class solution to see if we retained the same amount of information regarding class sizes and structure. Once again, the five-class solution had two classes with very small sample sizes (< 5%) and two nearly identical classes. Thus, we plotted the four-class solution to see if we could retain the same amount of information with a more stable class solution. The four-class solution retained the same class information (i.e., similar class sizes and structure). It also eliminated the repeat class that emerged in the five and six-class solutions and provided classes with larger sample sizes. Thus, given the similar model fit criteria from the five and six-class solutions, and the class information resulting from the removal of the repeated class, we chose the four-class solution.

Figure 1 presents item probability plots for each of the four classes. The high trauma exposure class (solid line) represented 9.2% (n = 113) of participants. Within this class, participants had a high probability of endorsing all trauma typologies. The moderate childhood trauma moderate combat class (dash-dot line) represented 27% (n = 333) of the sample. Veterans in this class endorsed only moderate levels of childhood trauma items (range 0.42 – 0.86) and combat trauma with a very low endorsement of military sexual trauma (harassment and violence) (range 0.02 – 0.09). The high military sexual trauma – moderate combat class (dotted line) represented 11.1% (n = 136) of the sample. Veterans in this class reported relatively high levels of military sexual trauma (range 0.47– 0.71) with moderate endorsement of most combat items and nearly no experiences of childhood trauma (range 0.04 – 0.23). Our final class, labeled combat only (dashed line), represented 52.7% (n = 648) of the sample. In this class, veterans have a relatively low endorsement of childhood and military sexual trauma (range 0.004 – 0.31) but high probability of endorsing most combat items.

Figure 1.

Figure 1.

Open in a new tab

Emergent latent classes for childhood trauma, combat trauma, and military sexual trauma.

Changes in posttraumatic growth across emergent trauma classes

Results can be found in Figure 2. Regarding changes in posttraumatic growth, our results indicated that veterans in the moderate childhood trauma moderate combat class had the lowest starting value for posttraumatic growth among all classes (intercept = 53.4, SE = 1.17, p < 0.01) with the slope of posttraumatic growth showing a steady decrease (slope = −1.44, SE = 0.30, p = <0.001). Veterans in the high trauma exposure class had a starting value of 58.7 (SE = 1.53, p < 0.01) for posttraumatic growth and a steady, significant decrease over time (slope = −2.81, SE = 1.13, p = 0.01). Veterans in the high military sexual trauma moderate combat class reported slightly higher starting values in posttraumatic growth (intercept = 59.9, SE = 2.18, p < 0.01) but a non-significant decrease throughout the study (slope = −0.23, SE = 0.88, p =0.80). Finally, veterans in the combat only class had the highest starting value for posttraumatic growth (intercept = 68.0, SE = 0.32, p < 0.01) and a rapid increase in posttraumatic growth throughout the study (slope = 1.81, SE = 0.11, p < 0.01). We also conducted a Wald test of parameter constraints between each intercept and slope value for each class (see Supplemental Table 1). Doing this allows us to determine if the values for intercepts and slopes significantly differ between classes. Results indicated that all intercept values between classes were significantly different, except for one paring: between veterans in the high trauma exposure class and the high military sexual trauma moderate combat class. Regarding slopes, the only differences to emerge were between veterans in the low trauma exposure class and all other classes. These results can be found in Supplemental Table 2.

Figure 2.

Figure 2.

Open in a new tab

Changes in posttraumatic growth by emergent childhood trauma, combat trauma, and military sexual trauma classes

Social support as a time-varying co-variate

Finally, we sought to understand how social support was associated with changes in posttraumatic growth and how these associations may vary by emergent childhood trauma and military sexual trauma classes. To do this, we entered social support as a time-varying co-variate within each class. We estimated a series of models that employed equality constraints on the effect of social support within each class. Our model-building process results can be found in Supplemental Table 3. In our final model (see Table 2), social support was allowed to be freely estimated as a time-varying co-variate for all classes. The effect of social support within the high trauma exposure class was relatively large and positive at each time point (B range = 14.1 – 20.5), indicating that individuals reporting greater social support at a specific time point had a “boost” in posttraumatic growth at that same time point. Interestingly, the effect of social support increased over time, and at the 24-month follow-up, the effect was the largest. Using a Wald-test of parameter constraints that compared values for the effect of social support at the first time point (value of 14.1) with the last time point (value of 20.5), we found no differences, indicating that, while the effects do increase over time, the increase is not significantly different (Wald Test = 1.77, p = 0.19). For those in the moderate childhood trauma and moderate combat class, the effects of social support started relatively large and had a significant, positive effect on posttraumatic growth through the first two time points (16.7, 13.9). However, the effect of social support faded over time and by the 24-month follow-up was no longer significant. Wald test indicates a significant difference between the first and last time points (Wald test = 15.5; p <0.001), indicating that the effect of social support fades significantly over time. For veterans in the high military sexual trauma moderate combat class, social support had positive effect on posttraumatic growth at each time point (B range = 6.5 – 10.6) and increased over time. Wald test indicates significant difference between the first and last time points, indicating the effect of social support is significantly greater over time (Wald test = 3.8; p = 0.04). Finally, for veterans in the combat only class, social support had a minimal (B range = −1.65 – 1.24) effect, which was not significant after the 12-month follow-up.

Table 2.

Effect of social support as a time-varying co-variate on posttraumatic growth by emergent trauma classes

High trauma exposure B SE p
Social support – 6mth 14.1 2.3 <0.01
Social support – 9mth 17.3 1.4 <0.01
Social support – 12mth 18.8 0.9 <0.01
Social support – 18mth 19.7 1.6 <0.01
Social support – 24mth 20.5 2.4 <0.01
Moderate Childhood trauma and Moderate combat B SE p
Social support – 6mth 16.7 3.2 <0.01
Social support – 9mth 13.9 2.8 <0.01
Social support – 12mth 7.6 3.4 0.03
Social support – 18mth 1.49 4.2 0.72
Social support – 24mth 0.04 4.9 0.99
High military sexual trauma moderate combat B SE p
Social support – 6mth 6.50 2.5 <0.01
Social support – 9mth 7.17 2.2 <0.01
Social support – 12mth 9.19 2.2 <0.01
Social support – 18mth 10.0 2.3 <0.01
Social support – 24mth 10.6 2.7 <0.01
Combat only B SE p
Social support – 6mth −1.65 0.80 0.04
Social support – 9mth 1.27 0.58 0.03
Social support – 12mth 1.16 0.62 0.07
Social support – 18mth 1.10 0.80 0.17
Social support – 24mth 1.24 1.00 0.22
Open in a new tab

Note: Time-varying co-variates were tested for equality constraints across each class. Each class was initially allowed to have freely estimated time-varying co-variates. We then systematically constrained the effects of social support to be equal over time by class, using reductions in negative two log-likelihood as a test of model fit. When the constrained model did not result in a significantly worse model fit, the effect of social support was constrained to be equal over time (childhood trauma only, military sexual trauma only).

Discussion

The present study aimed to understand the longitudinal effects of childhood trauma, military sexual trauma, combat trauma, and social support on posttraumatic growth among US veterans. We used LCA to reveal a four-class solution: high trauma exposure, moderate childhood trauma – moderate combat, high military sexual trauma – moderate combat, and combat only. One of the main aims of the present study was to extract heterogeneity in experiences of military sexual trauma, combat trauma, and childhood trauma among military veterans. More recent conceptualizations of traumatology suggest that exposure to multiple types of trauma may place individuals at heightened risk of adverse outcomes (Finkelhor et al., 2007). Consistent with this idea, our results suggest additional support for high-risk individuals who have experienced multiple types of trauma. In particular, the high trauma exposure class of veterans reported the highest prevalence of early childhood, combat, and military sexual trauma. This finding aligns with prior work that has also revealed similarly high-risk classes of veterans. For example, in a recent study, a class of veterans that was represented by high rates of combat trauma, childhood adversity, and sexual trauma reported the greatest levels of hazardous alcohol and cannabis use as well as the highest number of symptoms of depression and PTSD (Davis et al., 2022). Thus, given this and numerous studies noting that exposure to multiple trauma typologies places individuals at greater risk of adverse outcomes (Croft et al., 2018; Copeland et al., 2018), one might expect veterans in this higher-risk class to display the lowest levels of posttraumatic growth. However, our model also extracted classes that represented combat trauma in isolation and combination with other traumatic experiences, allowing us to explore differential effects on changes in posttraumatic growth.

Studies that examine specific types of trauma help answer complex questions about how individuals adjust after these experiences, and although some research suggests that the type of event experienced is associated with variation in certain aspects of posttraumatic growth (Karanci et al., 2012; Shakespeare-Finch & Armstrong, 2010), these studies are limited in their ability to explore changes in posttraumatic growth by trauma type. Thus, we estimated intercept and slope values of overall posttraumatic growth over two years to understand how changes in posttraumatic growth vary by emergent trauma classifications. Those with high trauma exposure started with a moderately high posttraumatic growth but also experienced a significant decline. The decline in posttraumatic growth could indicate the cumulative negative impact of multiple trauma exposures over time (Finkelhor et al., 2007). Veterans in the moderate childhood trauma – moderate combat class exhibited the lowest initial posttraumatic growth and a steady decline, suggesting that early trauma may have a lasting detrimental impact on growth and recovery (Anda et al., 2006). In contrast, veterans in the combat only class had the highest initial posttraumatic growth and a notable increase over time, potentially indicating a resilience or adaptive process specific to combat-related experiences (Tedeschi & Calhoun, 2004). Veterans with high military sexual trauma and moderate combat trauma showed slightly higher initial posttraumatic growth but no significant change over the study period. This stability might imply that while these veterans do not deteriorate, they do not experience continued growth. It may be that experiencing severe forms of trauma, such as sexual assault, have lasting and mitigating effects on positive change following experiences of traumatic events. Prior work among survivors of adult sexual assault has shown that trauma-related characteristics such as perception of life threats, negative reactions to disclosure, and characterological self-blame are all associated with lower levels of posttraumatic growth (Ullman, 2014). It may be that individuals who experience military sexual trauma are also experiencing higher levels of these essential trauma-related characteristics, thus mitigating positive posttraumatic growth.

It should be noted that these trauma-related characteristics have previously been shown to map onto differential trauma experiences (e.g., reporting experiences of sexual abuse is associated with high rates of each of the aforementioned trauma-related characteristics) and predict long-term substance use and psychopathology (Davis et al., 2019; Davis et al., 2022). It is also critical to consider the timing of trauma experiences. Tedeschi et al. (2018) posit that posttraumatic growth inherently influences schemas about oneself and the role of trauma in one’s life. Thus, the developmental timing of a traumatic event might determine the influence of posttraumatic growth trajectories, such that if a trauma occurs at an earlier period as childhood trauma does, the event may shape developing schemas instead of changing an existing schema that allows for posttraumatic growth (Tedeschi et al., 2018). The age at which trauma occurs may help explain why veterans who experienced childhood trauma reported the lowest rates of posttraumatic growth that significantly decreased over time, as evidenced by the high all and moderate childhood trauma – moderate combat classes. However, there is most likely a wide variability in the timing of the last trauma exposure among the sample, impeding the ability to draw significant conclusions related to timing.

The final aim of our study was to understand how social support influences changes in posttraumatic growth and how these effects may vary by emergent trauma classes. Social support emerged as a significant predictor of posttraumatic growth, with its impact varying across classes. For the high trauma exposure class, social support consistently enhanced posttraumatic growth, highlighting its critical role in mitigating the adverse effects of multiple trauma types (Cohen & Wills, 1985). The moderate childhood trauma – moderate combat class benefited from social support initially. Still, this effect diminished over time, suggesting that ongoing support might not be necessary to sustain posttraumatic growth in this group. In contrast, the combat only class showed minimal response to social support after 12 months, implying that other factors may drive posttraumatic growth in veterans with primarily combat-related trauma. In the high military sexual trauma – moderate combat trauma, the impact of social support increased over time, suggesting that not only is social support important throughout time for survivors of MST and combat trauma, but its importance grows in magnitude over time. These findings underscore the need for tailored interventions that address the unique needs of each trauma class.

Limitations and Conclusion

Although our study is novel in its methods and examination of compounded trauma’s effects on posttraumatic growth, it includes limitations. Our study design is strictly observational, hindering causal inference. In addition, measures are self-reported, which may be subject to bias. Furthermore, as traumatic events such as childhood, combat, and military sexual trauma can be confounded with other variables, it can be challenging to make generalizable conclusions. For example, Helgeson et al. (2006) propose that an event’s perceived severity and stressfulness are more likely to be associated with posttraumatic growth than the event’s objective characteristics. In addition, a key limitation of our study is the wide variability in the time since participants’ last trauma exposure, which complicates the interpretation of changes in posttraumatic growth. The variability in the time elapsed since participants’ last trauma event means that individuals may be at different stages of their recovery and growth processes. This temporal variation can affect the degree and trajectory of posttraumatic growth, making it challenging to draw significant conclusions about the rate of change in posttraumatic growth in this observational sample. Finally, our study has demographic limitations as our sample was predominantly male, limited in racial diversity, and non-clinical. These limitations are essential to highlight, as female veterans are more likely to experience military sexual violence (Castro et al., 2015), and minority veterans are more likely to experience a traumatic event than their white counterparts (Aponte et al., 2017). In addition, as most veterans (70%) in our community sample were within 4–7 years post-separation at the time of the last (Time 6) survey, only a fifth sought any care for any MH or SUD reason in the past 6 months. Thus, our findings may not be generalizable to clinical populations such as veterans involved in clinical Veteran Affairs settings.

Despite these limitations, our study ultimately advances our understanding of the intersections between posttraumatic growth, childhood trauma, combat trauma, and military sexual violence. Our results indicate that military veterans who have experienced childhood trauma may be at an increased risk of experiencing low levels of posttraumatic growth and may be less susceptible to the positive impacts of social support. Conversely, veterans with primarily combat-related trauma exhibited higher initial levels of posttraumatic growth and continued growth over time, highlighting a resilience or adaptive process specific to combat-related experiences. Future studies should examine the underlying mechanisms that may drive the effects of childhood trauma, military sexual violence, combat trauma, and social support. Interventions for veterans who have experienced childhood trauma, military sexual violence or compounded trauma, should increase access to formal and informal social support. Strategies to increase awareness of formal support services offered by the VA among veterans and healthcare providers remain essential. Additionally, efforts to strengthen informal support networks, such as peer support groups, community-based organizations, and initiatives that foster social connections and camaraderie, can complement formal care and provide survivors of trauma with additional sources of support and understanding. By acknowledging the importance of informal social support in promoting posttraumatic growth among trauma survivors and integrating strategies to enhance both formal and informal support networks, interventions can better address the diverse needs of veterans affected by cumulative trauma.

Supplementary Material

supplemental material

Clinical impact statement:

Our results highlight the vital role of social support in aiding posttraumatic growth among veterans, especially those dealing with compounded traumas like military sexual trauma, combat trauma, and childhood trauma. Findings suggests that personalized social support interventions could be crucial for their recovery. By understanding the varying impacts of trauma and support, healthcare providers can better tailor their approaches to meet the specific needs of veterans, promoting better mental health outcomes.

Acknowledgments:

This research was funded by grant R01AA026575 from the National Institute on Alcohol Abuse and Alcoholism (NIAAA), supplement R01AA026575–02S1, and a Keck School of Medicine COVID-19 Research Funding Grant awarded to Eric R. Pedersen.

Footnotes

Conflict of interest: none declared.

References

  1. Anda RF, Felitti VJ, Bremner JD, Walker JD, Whitfield CH, Perry BD, Dube SR & Giles WH (2006). The enduring effects of abuse and related adverse experiences in childhood: A convergence of evidence from neurobiology and epidemiology. European archives of psychiatry and clinical neuroscience, 256, 174–186. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Anda RF, Butchart A, Felitti VJ, & Brown DW (2010). Building a framework for global surveillance of the public health implications of adverse childhood experiences. American journal of preventive medicine, 39(1), 93–98. [DOI] [PubMed] [Google Scholar]
  3. Aponte M, Garin T, Glasgow D, Lee T, Newsome E III, Thomas E, Ward B (2017). Minority Veteran Report: Military Service History and VA Benefit Utilization Statistics. Department of Veteran Affairs. https://www.va.gov/vetdata/docs/SpecialReports/Minority_Veterans_Report.pdf [Google Scholar]
  4. Aronson KR, Perkins DF, Morgan NR, Bleser JA, Vogt D, Copeland LA, Finley EP, Gilman CL (2020). Use of health services among post-9/11 veterans with mental health conditions within 90 days of separation from the military. Psychiatric services, 71(7), 670–677. [DOI] [PubMed] [Google Scholar]
  5. Beguin-Fernald CA (2022). Cultivating Posttraumatic Growth and Community Reintegration: A Veteran-To-Farmer Program Evaluation (Doctoral dissertation, The University of Arizona). [Google Scholar]
  6. Brooks M, Graham‐ Kevan N, Lowe M, & Robinson S (2017). Rumination, event centrality, and perceived control as predictors of post‐ traumatic growth and distress: The Cognitive Growth and Stress model. British Journal of Clinical Psychology, 56(3), 286– 302. [DOI] [PubMed] [Google Scholar]
  7. Brooks M, Graham-Kevan N, Robinson S, & Lowe M (2019). Trauma characteristics and posttraumatic growth: The mediating role of avoidance coping, intrusive thoughts, and social support. Psychological Trauma: Theory, Research, Practice, and Policy, 11(2), 232–238. [DOI] [PubMed] [Google Scholar]
  8. Brooks M, Graham-Kevan N, Robinson S, & Lowe M (2021). “I get knocked down, but I get up again”–A qualitative exploration of posttraumatic growth after multiple traumas. Traumatology, 27(3), 274–284. [Google Scholar]
  9. Brooks M, Taylor E, & Hamby S (2023). Polyvictimization, polystrengths, and their contribution to subjective well-being and posttraumatic growth. Psychological trauma: theory, research, practice, and policy. Advance online publication. [DOI] [PubMed] [Google Scholar]
  10. Butler LD, Blasey CM, Garlan RW, McCaslin SE, Azarow J, Chen XH, Desjardins JC, DiMiceli S, Seagraves DA, Hastings TA and Kraemer HC& Spiegel D. (2005). Posttraumatic growth following the terrorist attacks of September 11, 2001: Cognitive, coping, and trauma symptom predictors in an internet convenience sample. Traumatology, 11(4), 247–267. [Google Scholar]
  11. Calhoun LG, & Tedeschi RG (2006). The foundations of posttraumatic growth: An expanded framework. In Calhoun LG & Tedeschi RG (Eds.), Handbook of posttraumatic growth: Research and practice (pp. 3 – 23). [Google Scholar]
  12. Calhoun LG, & Tedeschi RG (2014). The foundations of posttraumatic growth: An expanded framework. In Handbook of posttraumatic growth (pp. 3–23). Routledge. [Google Scholar]
  13. Castro CA, Kintzle S, Schuyler AC, Lucas CL, & Warner CH (2015). Sexual assault in the military. Current psychiatry reports, 17, 1–13. [DOI] [PubMed] [Google Scholar]
  14. Copeland WE, Shanahan L, Hinesley J, Chan RF, Aberg KA, Fairbank JA, ... & Costello EJ. (2018). Association of childhood trauma exposure with adult psychiatric disorders and functional outcomes. JAMA network open, 1(7), e184493-e184493. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Croft J, Heron J, Teufel C, Cannon M, Wolke D, Thompson A, ... & Zammit S (2019). Association of trauma type, age of exposure, and frequency in childhood and adolescence with psychotic experiences in early adulthood. JAMA psychiatry, 76(1), 79–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Davis JP, Dworkin ER, Helton J, Prindle J, Patel S, Dumas TM, & Miller S (2019). Extending poly-victimization theory: Differential effects of adolescents’ experiences of victimization on substance use disorder diagnoses upon treatment entry. Child Abuse & Neglect, 89, 165–177. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Davis JP, Prindle J, Castro CC, Saba S, Fitzke RE, Pedersen ER (2021). Changes in alcohol use during the COVID-19 pandemic among American veterans. Addictive Behaviors 122, 107052. 10.1016/j.addbeh.2021.107052. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Davis JP, Lee DS, Saba S, Fitzke RE, Ring C, Castro CC, & Pedersen ER (2022). Applying polyvictimization theory to veterans: Associations with substance use and mental health. Psychology of addictive behaviors, 36(2), 144. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Felitti VJ, Anda RF, Nordenberg D, Williamson DF, Spitz AM, Edwards V, Koss MP, & Marks JS (2019). Reprint of: relationship of childhood abuse and household dysfunction to many of the leading causes of death in adults: the adverse childhood experiences (ACE) study. American journal of preventive medicine, 56(6), 774–786. [DOI] [PubMed] [Google Scholar]
  20. Finkelhor D, Ormrod RK, & Turner HA (2007). Polyvictimization and trauma in a national longitudinal cohort. Development and psychopathology, 19(1), 149–166. [DOI] [PubMed] [Google Scholar]
  21. Haahr-Pedersen I, Ershadi AE, Hyland P, Hansen M, Perera C, Sheaf G, Bramsen RH, & Vallières F (2020). Polyvictimization and psychopathology among children and adolescents: A systematic review of studies using the Juvenile Victimization Questionnaire. Child Abuse & Neglect, 107, Article 104589. [DOI] [PubMed] [Google Scholar]
  22. Helgeson VS, Reynolds KA, & Tomich PL (2006). A meta-analytic review of benefit finding and growth. Journal of consulting and clinical psychology, 74(5), 797–817. [DOI] [PubMed] [Google Scholar]
  23. Hourani LL, Yuan H, & Bray RM (2003). Psychosocial and health correlates of types of traumatic event exposures among US military personnel. Military Medicine, 168(9), 736– 743. [PubMed] [Google Scholar]
  24. Jirek SL, & Saunders DG (2018). Cumulative adversity as a correlate of posttraumatic growth: The effects of multiple traumas, discrimination, and sexual harassment. Journal of Aggression, Maltreatment & Trauma, 27(6), 612–630. [Google Scholar]
  25. Karanci AN, Işıklı S, Aker AT, Gül Eİ, Erkan BB, Özkol H, & Güzel HY (2012). Personality, posttraumatic stress and trauma type: Factors contributing to posttraumatic growth and its domains in a Turkish community sample. European journal of psychotraumatology, 3(1), Article 17303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Kira IA, Aboumediene S, Ashby JS, Odenat L, Mohanesh J, & Alamia H (2013). The dynamics of posttraumatic growth across different trauma types in a Palestinian sample. Journal of Loss and Trauma, 18(2), 120–139. [Google Scholar]
  27. Laffaye C, Cavella S, Drescher K, & Rosen C (2008). Relationships among PTSD symptoms, social support, and support source in veterans with chronic PTSD. Journal of Traumatic Stress: Official Publication of The International Society for Traumatic Stress Studies, 21(4), 394–401. [DOI] [PubMed] [Google Scholar]
  28. Lee H, Mason D, Holden BE, Adams P, Guardiola L, & Buetikofer E (2015). Social Support and posttraumatic growth (PTG) among OEF-OIF and American Korean War Veterans: A mixed research study. International Journal of Humanities and Social Science, 5(8), 154–165. [Google Scholar]
  29. Maitlis S (2020). Posttraumatic growth at work. Annual Review of Organizational Psychology and Organizational Behavior, 7, 395–419. [Google Scholar]
  30. Maguen S, Vogt DS, King LA, King DW, & Litz BT (2006). Posttraumatic growth among Gulf War I veterans: The predictive role of deployment-related experiences and background characteristics. Journal of Loss and Trauma, 11(5), 373–388. [Google Scholar]
  31. Mark KM, Stevelink SA, Choi J, & Fear NT (2018). Posttraumatic growth in the military: a systematic review. Occupational and environmental medicine, 75(12), 904– 915. [DOI] [PubMed] [Google Scholar]
  32. Mitchell KJ, Moschella EA, Hamby S, & Banyard V (2020). Developmental stage of onset, poly-victimization, and persistence of childhood victimization: Impact on adult well-being in a rural community–based study. Child maltreatment, 25(1), 20–31. [DOI] [PubMed] [Google Scholar]
  33. Nagin DS (2005). Group-based modeling of development. Cambridge, MA: Harvard University Press. [Google Scholar]
  34. Ning J, Tang X, Shi H, Yao D, Zhao Z, & Li J (2022). Social support and posttraumatic growth: A meta-analysis. Journal of Affective Disorders. 320, 117–132. [DOI] [PubMed] [Google Scholar]
  35. Nordstrand AE, Bøe HJ, Holen A, Reichelt JG, Gjerstad CL, & Hjemdal O (2020). Social support and disclosure of war-zone experiences after deployment to Afghanistan— Implications for posttraumatic deprecation or growth. Traumatology, 26(4), 351–360. [Google Scholar]
  36. Nuccio AG, & Stripling AM (2021). Resilience and posttraumatic growth following late life polyvictimization: A scoping review. Aggression and Violent Behavior, 57, 101481. [Google Scholar]
  37. 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: A multidisciplinary Journal, 14(4), 535–569. [Google Scholar]
  38. Nylund-Gibson K, Grimm RP, & Masyn KE (2019). Prediction from latent classes: A demonstration of different approaches to include distal outcomes in mixture models. Structural equation modeling: A multidisciplinary Journal, 26(6), 967–985. [Google Scholar]
  39. Pietrzak RH, Goldstein MB, Malley JC, Rivers AJ, Johnson DC, Morgan CA III., & Southwick SM. (2010). Posttraumatic growth in veterans of operations enduring freedom and Iraqi freedom. Journal of affective disorders, 126(1–2), 230–235. [DOI] [PubMed] [Google Scholar]
  40. Schell TL, & Marshall GN (2008). Survey of individuals previously deployed for OEF/OIF. In Tanielian T, & Jaycox LH (Eds.). Invisible wounds of war: Psychological and cognitive Injuries, their consequences, and services to assist recovery Santa Monica, CA: RAND MG-720. [Google Scholar]
  41. Shakespeare-Finch J, & Armstrong D (2010). Trauma type and posttrauma outcomes: Differences between survivors of motor vehicle accidents, sexual assault, and bereavement. Journal of Loss and Trauma, 15(2), 69–82. [Google Scholar]
  42. Sherbourne CD, & Stewart AL (1991). The MOS social support survey. Social science & medicine, 32(6), 705–714. [DOI] [PubMed] [Google Scholar]
  43. 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, 194–200. [DOI] [PubMed] [Google Scholar]
  44. Tedeschi RG, & Calhoun LG (1996). The Posttraumatic Growth Inventory: Measuring the positive legacy of trauma. Journal of traumatic stress, 9, 455–471. [DOI] [PubMed] [Google Scholar]
  45. Tedeschi RG, & Calhoun LG (2004). Posttraumatic growth: Conceptual foundations and empirical evidence. Psychological Inquiry, 15, 1–18. [Google Scholar]
  46. Tedeschi RG, Shakespeare-Finch J, Taku K, & Calhoun LG (2018). Posttraumatic growth: Theory, research, and applications. Routledge. [Google Scholar]
  47. Thomas EA, Owens GP, & Keller EM (2021). Relationships among non‐ interpersonal and interpersonal trauma types, posttraumatic stress, and posttraumatic growth. Journal of Clinical Psychology, 77(11), 2592–2608. [DOI] [PubMed] [Google Scholar]
  48. Ullman SE (2014). Correlates of posttraumatic growth in adult sexual assault victims. Traumatology, 20(3), 219–224. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. United States Department of Veteran Affairs (2021) Military Sexual Trauma [Fact sheet]. https://www.mentalhealth.va.gov/docs/mst_general_factsheet.pdf

Associated Data

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

Supplementary Materials

supplemental material
NIHMS2056885-supplement-supplemental_material.docx (117KB, docx)

RESOURCES