Article Text

Download PDFPDF

Post-traumatic growth in the military: a systematic review
  1. Katharine M Mark1,
  2. Sharon A M Stevelink1,
  3. Jeesoo Choi2,
  4. Nicola T Fear1,3
  1. 1 King’s Centre for Military Health Research, King’s College London, London, UK
  2. 2 School of Medicine, King’s College London, London, UK
  3. 3 Academic Department of Military Mental Health, King’s College London, London, UK
  1. Correspondence to Dr Katharine M Mark, King’s Centre for Military Health Research, King’s College London, London, SE5 9RJ, UK; katharine.mark{at}kcl.ac.uk

Abstract

Background Post-traumatic growth is defined as positive psychological, social or spiritual growth after a trauma.

Objectives This systematic review aimed to identify studies that quantitatively measured post-traumatic growth among (ex-) military personnel, to determine whether there is evidence of growth in this context and whether such growth is associated with any sociodemographic, military, trauma or mental health factors.

Data sources The electronic databases PsycInfo, OVIDmedline and Embase were searched for studies published between 2001 and 2017.

Study eligibility criteria and participants Papers were retained if they involved military or ex-military personnel, where some had been deployed to Iraq or Afghanistan.

Study appraisal Quality assessment was conducted on all studies.

Results 21 studies were retained. The Post-Traumatic Growth Inventory was employed by 14 studies: means ranged from 32.60 (standard deviation = 14.88) to 59.07 (23.48). The Post-Traumatic Growth Inventory Short Form was used by five studies: means ranged from 17.11 (14.88) to 20.40 (11.88). These values suggest moderate growth. Higher levels of social support, spirituality and rumination and minority ethnicity were most frequently associated with more post-traumatic growth.

Limitations The involved studies may lack generalisability and methodological quality.

Conclusions Overall, this paper confirms that negative reactions to trauma, particularly post-traumatic stress disorder, are not the only possible outcomes for service personnel, as moderate post-traumatic growth can also be observed.

Implications of key findings Interventions aimed at helping current and former armed forces personnel to identify and promote post-traumatic growth post-conflict may be beneficial for their well-being.

  • military personel
  • post-traumatic growth
  • veterans
  • systematic review

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Key messages

What is already known about this subject?

  • Post-traumatic growth has been shown to occur following a range of traumatic experiences. It is related to an individual’s affect and values prior to the distressing event, as well as to external details, such as social, financial, mental health and demographic factors. For example, growth has been linked with higher levels of social support and lower levels of depression. In terms of military-related post-traumatic growth specifically, existing studies have focused on various past conflicts and service subgroups. However, there has not been a systematic review conducted into the presence of post-traumatic growth, or the factors associated with the phenomenon, in military and ex-military personnel.

What are the new findings?

  • For the first time, we investigated post-traumatic growth in the military using a systematic review design. Moderate levels of growth were found across the 21 included studies, indicating positive change in military and former military personnel following trauma. There was a fairly large range in post-traumatic growth scores across the papers. In terms of factors associated with post-traumatic growth, we found the most frequently reported association was between minority group ethnicity and higher levels of growth and the strongest reported association was between time since the traumatic event and higher levels of growth. Post-traumatic growth was also associated with social support and rumination across a number of studies.

Key messages

How might this impact on policy or clinical practice in the foreseeable future?

  • As well as focusing on the negative consequences of trauma in a military context, this systematic review suggests that treatment provisions for armed forces members, post-deployment, can usefully encourage more positive outcomes. Our findings indicate that interventions aimed at helping current and former military personnel to identify and promote post-traumatic growth may be beneficial for their psychological well-being. In line with the associated factors identified here, clinicians should be advised to provide and encourage additional social support for those returning from conflict zones who identify as Caucasian and who report low levels of spirituality and rumination.

Introduction

Background

Historically, research has focused on post-traumatic stress disorder (PTSD) to understand the negative implications of trauma on behaviour, cognition and emotions.1 However, evidence of positive reactions to distressing events has been observed, termed post-traumatic growth (PTG).2 The phrase is defined as positive psychological, social or spiritual growth after a traumatic incident. Its individual elements are broadly classified as: personal improvement, altered priorities, improved relationships and finding meaning in life.3 PTG has been reported following cancer,4 natural disasters,5 abuse6 and military deployment.7

The degree of positive change experienced is known to be linked to both internal and external factors. The former is an individual’s ‘personal system’, which refers to one’s affect and values prior to the traumatic event. The latter may be the network of support available, certain social, financial and demographic backgrounds and factors related to the event itself.8 For example, the internal factors of searching for answers,9 personality traits10 and depression11 and the external factors of age12 and social support13 have all been linked to PTG. Focusing on two of the most strongly endorsed of these specific factors, the literature documents a negative association between PTG and depression14— with higher levels of PTG occurring for individuals with lower levels of depression — and a positive association between PTG and social support15— with higher levels of PTG occurring for individuals with higher levels of social support.

When considering military-related PTG specifically, the literature spans from past battles, such as the World Wars16 and the Vietnam War,17 to the most recent conflicts in Iraq and Afghanistan.18 Studies have sometimes limited their focus to a subsample of the military — for example, infantry,19 chaplains20 or medical personnel21 — or to a specific type of service-related trauma, for example — amputation22 or brain injury.23 Although psychological difficulties are present for a number of returning service personnel,24 there is an increased interest in PTG as a positive consequence of deployment. Investigating such positive outcomes in a military context is important considering the substantial risk of trauma exposure and potential for mental health problemswithin this population.24 A more thorough understanding of PTG in military personnel may also have implications for clinical practice, by confirming whether PTG should be incorporated into psychological treatments for service members and veterans.25 Indeed, programmes and training, such as ‘Comprehensive Soldier Fitness’,26 ‘Higher Ground’27 and ‘Battlemind’,28 which help facilitate well-being, resilience and decompression in post-deployment military personnel, are starting to acknowledge PTG.

Objectives

While there have been systematic reviews of PTSD among military and ex-military personnel,29 there have, to the best of our knowledge, been no comparable reviews focusing solely on PTG in these populations. To address this gap in the literature, the current paper systematically reviewed studies, published between 2001 and 2017, that quantitatively measure PTG in previously deployed (ex-) military personnel. It aimed to identify whether PTG was present, as well as the factors associated with the phenomenon, within this specific group.

Method

This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) checklist (see online supplementary file 1).

Supplemental material

Search strategy

The literature search was carried out in December 2017. The electronic databases PsycInfo, OVIDmedline and Embase identified studies published between January 2001 and December 2017. Search terms used were: ‘Post-traumatic growth’; ‘PTG’; ‘Trauma’; ‘Growth’; ‘Stress related growth’; ‘Perceived benefit’; ‘Benefit finding’; ‘Military’; ‘Veteran’; ‘Deployment’; ‘Combat’; ‘War’; ‘Army’; and ‘Armed forces’ (see online supplementary file 2). The reference lists of included studies were checked for further relevant papers. Authors were contacted to obtain additional information when needed.

Selection strategy

A total of 449 articles were retrieved from the above bibliographic searches. Two hundred and eighty-five papers were removed as duplicates. A further 122 were rejected after reviewing paper titles and abstracts. The final 42 papers were read in full, of which 21 were deemed to be relevant to the search criteria and appropriate for assessing our research objective (see online supplementary file 3).

Inclusion criteria

Inclusion criteria for the systematic review were: (1) studies measuring PTG using any quantitative tool; (2) studies focusing on military samples where at least some included personnel had been deployed to Iraq and/or Afghanistan; (3) studies published between 2001 and 2017, to cover the entirety of the conflicts in Iraq and Afghanistan and any papers published subsequently; and (4) studies published in English.

Exclusion criteria

Exclusion criteria were: (1) reviews, PhD dissertations, conference proceedings, abstracts, unpublished studies and books; (2) randomised controlled trials and pilot, case and intervention study designs; and (3) studies of the families of service personnel.

Data extraction

The following data were extracted and then checked and verified by the research team: title; author(s); publication year; study location; study design; sample type, defined here as either: (1) representative — a general military sample from the specific population being studied; (2) medical — individuals enrolled with a medical centre, such as the United States of America (USA) Department of Veterans Affairs (VA), for physical, and not psychological/emotional, care; or (3) clinical — individuals referred to a behavioural health clinic for psychological/emotional care or those diagnosed with PTSD; number of participants; gender distribution; service status, defined here as either: (1) active duty; (2) national guard/reservist; or (3) veteran; deployment location; response rate; PTG measure (and the traumatic event referred to in its wording); and means and standard deviations (SD) of PTG scores (the core summary measures). Data relating to factors associated with PTG were also extracted.

Quality assessment

The Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies was adapted30 to assess the quality of, and risk of bias within, each study. This appraisal tool is based on quality assessment methods, concepts and scales developed by various stakeholders in the field.30 It has been used to assess study quality across multiple systematic reviews.31 32 Two raters (KMM and SAMS) separately graded each study according to 13 criteria (no = 0, yes = 1). There was a maximum quality score of 13 and this score was used to create a quality rating of ‘poor’, ‘fair’ or ‘good’. In order to make the rating system as simple as possible, it was decided among the authors that four criteria would be considered key when scoring cohort/longitudinal articles and three criteria would be considered key when scoring cross-sectional articles. A study that met one or none of these items received a quality rating of ‘poor’; a study that met two items received a quality rating of ‘fair’; and a study that met three (or three or four for longitudinal studies) items received a quality rating of ‘good’ (see online supplementary file 4).

Results

Study demographics

Table 1 shows demographic information for each of the 21 included studies. All but one33 of the studies were conducted in the USA. Thirteen18 19 21 22 34–42 out of the 21 studies were cross-sectional in nature and 1818–20 22 33–36 39–48 had mixed gender samples. The average number of respondents was 1143 (and the range was 5622 to 530243). Eleven18–20 22 34 36–38 42–44 of the studies recruited military personnel who served in Iraq and/or Afghanistan and 1422 33 34 36 38–42 45–49 included veterans (defined here, in line with the USA definition, as individuals who had previously been deployed in combat, who were not currently deployed but who were still employed by the services or as individuals who had been deployed in the military and had subsequently left). As well as including personnel who had been deployed during their time in the military, three35 45 47 studies included personnel who had not been deployed, with the prevalence of this non-deployed group ranging from 8% of the sample35 to 66% of the sample.47 Finally, 1319 21 34 35 37 39–42 44 47–49 studies recruited what we term representative military samples, that is, a general military sample from the specific population being studied.

Table 1

Demographic information and quality rating for each included study

Quality assessment

Five19 33 38 39 41 of the 21 included studies received a rating of ‘good’; 1218 34–37 40 42 43 45 47–49 received a rating of ‘fair’; and 4 21 22 44 46 received a rating of ‘poor’(see table 1 and online supplementary file 5).

Across the included studies, all 21 fulfilled criteria number one of the quality assessment measure — was the research question clearly stated? Numbers five — was a sample size justification, power description or variance and effect estimate provided?; six — were the independent variables measured prior to the outcome being measured?; and 10 — were the independent variables assessed more than once over time? — of the quality assessment measure were least often endorsed. Three20 36 38 out of a possible 21 studies, three31 46 47 out of a possible 21 studies and two31 46 out of a possible 20 studies (this criterion was not applicable for one study) fulfilled these criteria, respectively.

Across the four studies that received a quality rating of ‘poor’,21 22 44 46 all four fulfilled critera numbers one; two — was the study population clearly defined?; eight — for independent variables that can vary in amount or level, did the study examine different levels as related to the outcome?; and nine — were the independent measures clearly defined, valid, reliable and implemented consistently? — of the quality assessment measure. Numbers six; seven — was the timeframe sufficient to see an association between independent variable(s) and outcome?; and 10 of the quality assessment measure were least often endorsed, with none of the four studies fulfilling these criteria.

Measures

Table 2 shows information on the outcome of PTG for each included study and these details are also shown in a Forest plot in figure 1. Fourteen  18 19 21 22 33 36–38 40–43 45 46 studies used the Post-Traumatic Growth Inventory (PTGI2; see online supplementary file 6). This questionnaire consists of 21 items and yields a score ranging from 0 to 105. A higher score indicates greater PTG. Factorial stability,50 internal consistency,51 test–retest reliability2 and convergent and discriminant validity52 are high for both the total scale and the five individual subscales of the PTGI. Confirmatory factor analysis has further validated use of the PTGI with a recently deployed population.18

Table 2

PTG information for each included study

Figure 1

A Forest plot graph to show mean PTG scores and associated 95% CIs for each included study. Errors bars show CIs: lower cap means difference between mean value and lower CI; upper cap means difference between upper CI and mean value. All studies are shown. Square data points mean studies using the full PTGI (with a maximum score of 105); triangle data points mean studies using the PTGI-SF (with a maximum score of 50); circle data points mean studies using neither the PTGI nor the PTGI-SF. The study by Tsai and Pietrzak49 does not have CIs shown here. These could not be calculated, because the study was lacking an SD value. PTG, post-traumatic growth; CI, confidence interval; PTGI, Post-Traumatic Growth Inventory; PTGI-SF, Post-Traumatic Growth Inventory Short Form. 

The mean score found across the 14 studies employing the PTGI was 45.48 (SD = 23.25). There was a fairly large range in PTG scores across these papers, with a 26.47-point difference between the lowest PTGI value (mean  = 32.60; SD = 14.88)33 and the highest PTGI value (mean 59.07; SD = 23.48).22 All but three19 33 46 of these papers reported Cronbach’s alphas for their samples on the PTGI, in order to assess the internal reliability of the scale. All values were excellent, with ranges between 0.9042 and 0.96.21 37 None of the studies employing the PTGI reported on the validity or reliability of the scale within their sample.

Six studies used a shortened version of the PTGI. Five of these39 44 47–49 used a 10-item version — the PTGI Short Form (PTGI-SF).50 This mirrors the strong psychometric qualities of the PTGI. For example, confirmatory factor analysis on the items of the PTGI-SF replicates the five-factor structure supported by the PTGI46. The short form also reproduces relationships between PTG and variables of interest among various trauma-afflicted samples49 and produces a total scale internal consistency coefficient of 0.89.49 Excellent reliability, factor structure and concurrent validity for the PTGI-SF have also been shown in a previously deployed military sample.44

Of the five studies that employed the PTGI-SF, one49 did not report an SD value for the mean score on this measure. The five studies that reported a PTGI-SF score had a mean of 18.50 (out of 50; SD = 11.51 for the four studies that reported this statistic). All five of these papers reported Cronbach’s alphas for their samples on the PTGI-SF, in order to assess the internal reliability of the scale. All values were excellent, with ranges between 0.9044 and 0.95.39 47–49

One34 study used a shorter six-item version of the PTGI, specifically designed by the authors. While the researchers reported a high internal consistency score, it is worth nothing that this new scale has yet to be used by others. The study reported a mean PTG score of 17.10 (out of 30; SD = 0.75) and the Cronbach’s alpha was 0.86.

Finally, one35 study used a self-designed and unvalidated four-item study questionnaire for measuring PTG and the benefits of deployment and found a mean PTG score of 3.10 (out of 4; SD = 0.92). Questions put forward two statements about pride: (1) ‘A feeling of pride for having served our country’ and (2) ‘A sense of accomplishment for a job well done’; as well as two about money: (1) ‘I earned more while deployed’ and (2) ‘I had more health care/retirement benefits’.

Of the 14 studies that used the PTGI, six18 19 37 38 42 43 asked subjects to focus on their deployment/combat experience. The mean PTG score of these 14 studies was 45.81 (SD = 22.66). Of the five studies that employed the PTGI-SF, four39 47–49 asked subjects to focus on the worst traumatic event suffered in their lifetime. The mean PTG score of these four studies was 18.03 (SD = 11.38 for the three studies that reported this statistic). See online supplementary file 7 and online supplementary file 8 for graphs showing mean levels of PTG on the PTGI and PTGI-SF, respectively, as a function of their wording.

Out of the 14 studies that used the PTGI, six19 21 37 40–42 recruited representative samples. These six studies had a mean PTG score of 42.79 (SD = 23.67). When medical samples were recruited, the mean PTG score was 51.34 (SD = 24.14). When clinical samples were recruited, the mean PTG score was 45.19 (SD = 22.20).

Associated factors

As shown in table 3, the most commonly reported, in six19 37 39 42 44 47 (out of a possible seven) studies, statistically significant association was between ethnicity and PTG, with less PTG occurring for Caucasians than for ethnic minority groups. Three (out of a possible three) studies each also reported that higher levels of social support22 36 49 and rumination22 40 41 were associated with higher levels of growth (see online supplementary file 9).

Table 3

Factors associated with PTG for each included study, along with their direction of association

Other factors significantly associated with growth were: perceived threat (positive relationship found in two38 44 out of a possible two studies); challenges to core beliefs (positive relationship found in two39 41 out of a possible two studies); military rank (negative relationship found in two19 37 out of a possible two studies); and suicidal ideation (negative relationship found in two19 43 out of a possible two studies).

One paper40 (out of a possible three) showed a positive relationship between PTG and time since the traumatic event. However, this association was the strongest statistically significant association of all those tested, with the largest effect size, an F-test statistic of 49.60 (see online supplementary file 9).

Two factors showed mixed associations with PTG: combat exposure and PTSD. Five (out of a possible 10) studies reported a significant association between combat exposure and PTG, with higher levels of combat exposure being associated with higher levels of growth in four19 37 42 43 studies and a curvilinear relationship being reported in one21 study. However, the other five34 36 38 39 44 studies found a non-significant association. Of the 14 studies that investigated the association between PTSD and PTG, five19 33 36 38 45 reported a non-significant relationship. However, six34 38–41 49 studies reported a positive association, with higher levels of PTSD being associated with higher levels of growth and three42 43 48 reported a negative association, with higher levels of PTSD being associated with lower levels of growth.

Discussion

Key findings

This systematic review found moderate levels of PTG across the 21 included studies, indicative of positive change in military (or former military) personnel. There was a fairly large range in PTG scores across the included papers. For the 14 studies that employed the PTGI, there was a 26.47-point difference between the lowest value33 and the highest value.22 Perhaps this difference is attributable to the mental health statuses of the veterans in these two studies. Specifically, the participants from the former study33 had been diagnosed with, and were being treated for, service-related PTSD symptoms at a military charity within the United Kingdom (UK), while the latter group were a non-clinical and randomly picked USA veteran sample.22

A range of factors were shown to be associated with PTG across the 21 studies. The most frequently reported association was between minority group ethnicity and higher levels of PTG and the strongest reported association between time since the traumatic event and PTG. PTG was also associated with social support and rumination across a number of studies.

Comparisons with previous research

PTG scores, measured using the PTGI, of populations involved in historical conflicts have reported growth values close to the range of those in this review (32.60 (SD = 14.88) to 59.07 (SD = 23.48)). For example, World War II veterans reported an average PTG score of 63.27 (SD = 20.69)16 and Yugoslavia war veterans reported an average PTG score of 35.82 (SD = 18.09).11 When considering traumatic experiences within the civilian population, higher growth scores than those reported in military-based studies have been documented. For example, bereaved individuals have reported an average PTG score of 79.72 (SD = 19.50)51 and cancer survivors have reported an average PTG score of 73.00 (SD = 21.00).52

Perhaps there are distinctive factors at play for military and non-military groups. For example, searching for answers,9 personality traits10 and age12 have all been associated with PTG. Studies of armed forces members tell us that these individuals are likely to: (1) actively search for meaning following a traumatic event, due to the, often, large-scale devastation of war16; (2) embody characteristics such as emotional instability — the majority of those in the forces come from a relatively deprived background and instability goes hand in hand with low income and socioeconomic status53; and (3) be young in age, particularly in comparison with those suffering from ill-health related traumas, which increase with age.54 All of these factors are related to lower levels of growth, which could explain the higher PTG scores reported by non-military, compared with military, trauma victims. Moreover, perhaps the experienced traumas themselves are qualitatively different depending on military status, which in turn may elicit varying responses in armed forces personnel versus civilians. These concepts will be explored more in the section below.

Associated factors

The consistent associations found between ethnicity and PTG19 37 39 42 44 47 are in line with past research.55 Minority groups may be more likely to be socioeconomically disadvantaged and to subsequently experience multiple additional stressors in their daily lives. The prior confrontation with differing worries may enable them to develop the skill of growing from hardships,56 including armed forces deployments. Alternatively, the greater significance of spirituality among ethnic minorities may increase growth. An emerging body of evidence supports the fact that religion and spirituality may provide beneficial ways for trauma survivors to understand their traumatic experiences,13 for example, by increasing personal strength and appreciation of life. Indeed, there was some support for the association between spirituality and PTG within this review (in two42 47 out of a possible four studies). Although the observed associations between both ethnicity and spirituality and growth were expected, it is interesting to note that all of the studies endorsing such links employed representative military samples, as opposed to medical or clinical groups.

A positive association was found between social support and PTG.22 36 49 Being cared for by others fosters an environment in which an individual can create meaning from their experience and can subsequently improve their ability to cope.3 However, findings linking growth with more specific operationalisations of social support in this review were mixed. Three papers included found no association between PTG and unit social support,44 post-deployment social support34 and social connectedness,47 respectively. Two44 47 out of these three34 44 47 opposing papers, showing no association between these specific forms of social support and PTG, employed a longitudinal research design and recruited samples accessing healthcare services for medical problems. In contrast, two22 36 out of the three22 36 49 papers that endorsed the link between greater social support and greater PTG were cross-sectional in nature and focused on representative military samples. First, these methodological differences show that the two44 47 studies reporting a non-significant link between the target variables were more robust and empirically valid. Second, the results suggest that the relationship between PTG and social support is weaker for individuals with less severe and complex military-related needs. Overall, these inconsistent conclusions highlight the need for both high-quality studies and qualitative research into these specific relationships.

The finding linking rumination22 40 41 with PTG was predictable, because this link is already well-established within the literature.57 Indeed, rumination is listed as a key element of the PTG model, originally proposed by Tedeschi and Calhoun.3 According to these authors,3 PTG does not emerge as a direct result of trauma; rather growth is a consequence of an individual’s struggle with, and development of, a new reality following a distressing event. In line with this definition, the closely related construct of challenges to core beliefs was found to correlate positively with growth here.40 41 Importantly, all three of the studies22 40 41 reporting significant associations between rumination, challenges to core beliefs and growth focused on veterans and were cross-sectional in nature. Such a relationship needs clarifying in longitudinal studies of active duty and reservist personnel, because it may be that study design and military role impact on the link between these constructs.

In keeping with previous studies,54 combat exposure and PTG were positively associated.19 37 42 43 Three19 37 43 (out of a possible four19 37 42 43) papers that endorsed this association included active duty service members. It has been suggested that individuals may gradually build up a tolerance to stress and trauma on repeated exposure and consequently may develop coping skills to deal with such situations.58 In the context of combat then, active military personnel who are repeatedly exposed to conflict may be better equipped to handle the effects of traumatic ordeals. Thus, these individuals may be more likely to experience positive growth from these distressing experiences. Contrastingly, when veterans and, in one paper, reservists were studied, this review found that combat experience was non-significantly associated with PTG.34 36 38 39 44 This seems logical and in line with Schnurr and colleagues’59 proposal, as ex-serving personnel are less likely to have been subjected to recent recurring traumas than active duty individuals. Alternatively, perhaps it is simply the passing of time since the trauma that results in less PTG for veterans and reservists who, we could argue, are more distanced from the immediate impact of the military than those on active duty.

The current review highlights mixed results for the association between PTG and PTSD.19 33 34 36 38–44 48 49 This corresponds with past literature showing that the relationship between these outcomes is unclear and potentially complex. The inconsistent associations reported may be accounted for by a third mediating variable, such as resilience or cognitive appraisal post-deployment. While there has been controversy regarding the role of resilience, some research has suggested that those who develop negative post-deployment outcomes, like PTSD, may have both more resilient personality characteristics60 and a more positive appraisal of their distressing experience,61 which could make them more likely to experience growth following a potentially traumatic event.

Alternatively, two previous studies investigating distress following both a terrorist attack62 and severe breast cancer63 have demonstrated a curvilinear association between PTSD and PTG, whereby the relationship follows an inverted ‘U’-shaped curve. Similar outcomes have been drawn in military samples64 and may be at play here. Such non-linearity would indicate that both high and low levels of PTSD symptoms are linked to low levels of growth and that mid-levels are linked to higher levels of growth. It is also worth noting that low levels of growth and low levels of PTSD occurring together, in tandem, may simply be indicative of the fact that the traumatic event in question was not as traumatic to the participant as would be expected.

Strengths and limitations

This comprehensive, multidatabase systematic search and review into quantitatively measured PTG in the military used robust and well-established methodology and quality guidelines. However, taking into account the variability in PTG scores across studies, our use of means as summary statistics is an important limitation. Considering the individual studies included in the systematic review, only one was conducted outside of the USA and four recruited fewer than four women participants. Furthermore, all studies made use of self-report questionnaires. While the full PTGI scale, employed by the majority of papers here, has demonstrated reliability and validity in multiple populations,45 rater bias is considered a risk for surveys that require completion by participants themselves.65 This is likely to be an especially prominent problem for poorly validated measures, such as the six-item PTGI (used here by Pietrzak et al 34) and the four-item Positive Benefits of Deployment Scale (used here by Scott et al).35 Perhaps reflecting the weaknesses addressed here, only five out of the 21 target papers were rated as ‘good’ quality. Employing qualitative, researcher-led interviews may help to discern differential reasons for growth, while simultaneously reducing rater bias.

Implications

The evidence available to date indicates that military populations experience moderate PTG following deployment and that this growth is not necessarily related to symptoms of PTSD. As well as focusing on the negative outcomes of trauma in this context, care provisions for armed forces members, post-deployment, can usefully encourage more positive consequences. Indeed, this paper’s findings indicate that interventions aimed at helping current and former armed forces personnel to identify and promote PTG may be beneficial for their psychological well-being. In line with the salient associated factors identified here, clinicians should be advised to encourage and provide additional social support to those returning from Iraq and Afghanistan and other conflict zones who identify as Caucasian and who report low levels of spirituality and rumination.

Conclusions

This systematic review adds to the evidence base on PTG, indicating that growth exists in military personnel and that negative reactions to trauma are not the only consequences following deployment. Therefore, interventions aimed at helping past and present military members to identify and promote more positive outcomes, particularly post-conflict, may be beneficial.

References

Footnotes

  • Contributors Study concept and design: SAMS, JC and NTF. Acquisition, analysis or interpretation of data: KMM and JC. Drafting of the manuscript: KMM and JC. Critical revision of the manuscript for important intellectual content: all authors. Statistical analysis: KMM. Administrative, technical or material support: SAMS. Study supervision: SAMS and NTF.

  • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

  • Competing interests NTF is a trustee of a veterans’ charity.

  • Patient consent Not required.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Data sharing statement There are no unpublished data from the study.