ABSTRACT
Background Self-harm and suicide behaviours are a major public health concern. Several factors are associated with these behaviours among military communities. Identifying these factors may have important implications for policy and clinical services.
Aims To review the risk and protective factors associated with self-harm and suicide behaviours among serving and ex-serving personnel of the United Kingdom Armed Forces, Canadian Armed Forces, Australian Defence Force and New Zealand Defence Force.
Method A systematic search of seven online databases (PubMed, Web of Science, Embase, Global Health, PsycINFO, PTSDpubs and CINAHL) was conducted alongside cross-referencing, in October 2022. Following an a priori PROSPERO approved protocol (CRD42022348867), papers were independently screened and assessed for quality. Data were synthesised using a narrative approach.
Results Overall, 28 papers were included: 13 from Canada, 10 from the United Kingdom, five from Australia and none from New Zealand. Identified risk factors included being single/ex-relationship, early service leavers, shorter length of service, junior ranks, exposure to deployment-related traumatic events, physical and mental health diagnoses, and experience of childhood adversity. Protective included being married/in a relationship, higher educational attainment, employment, senior ranks, and higher levels of perceived social support.
Conclusions Adequate care and support are a necessity for the military community. Prevention and intervention strategies for self-harm and suicide behaviours should promote social networks as a key source of support. More peer reviewed research is needed, particularly among the New Zealand Defence Force and regarding modifiable risk and protective factors.
1. INTRODUCTION
Self-harm and suicide behaviours are a major public health concern, with over 700,000 people dying by suicide every year worldwide and an estimated 14.6 million people engage in self-harm each year globally.1,2 The aetiology and onset of self-harm and suicide behaviours is complex and multifaceted; prevalence and risk appear to be influenced by several factors and vary by age, sex, ethnicity, geographical region and occupation.3–7 Military populations are potentially an at-risk group as military service can impact on the health and wellbeing of personnel both during and after service.8,9 Self-harm and death by suicide appear to be relatively rare among the military community, although rates of both behaviours have increased in recent years, they remain either lower or comparable to the general population.10,11,20,21,12–19 Suicide behaviours can present as ideation (i.e. thoughts) and attempts which are also prevalent among serving and ex-serving personnel, with a reported 11% global prevalence of these behaviours in the entire military.22
Previous research has reported that serving and ex-serving personnel are at risk of self-harm and suicide behaviours if they have mental health diagnoses, for example post-traumatic stress disorder (PTSD).18,19,23–26 Another specific at-risk group includes those separated from the military18,27–30, especially if separation occurred involuntarily (e.g. medical discharge).31,32 Additionally, younger age is a risk factor16,31,33, with one recent study from the United Kingdom (UK) reporting that suicide risk was two to three times higher in men and women under 25 years old who had left the UK Armed Forces (compared to the general population).16 The study also reported that ex-serving personnel with a shorter length of service (<10 years of service) were at increased risk.16 Protective factors for self-harm and suicide behaviours among the military community have been less frequently explored than risk factors, but include social support18,34–36, higher educational attainment37, employment38 and holding a more senior rank during service (e.g. officer)18,19,39.
There remains a paucity of research exploring the risk and protective factors for self-harm and suicide behaviours among several military populations, with available literature focussing on a limited number of countries. This review focused on military populations from four nations from the Five Eyes Alliance: namely the UK, Canada, Australia and New Zealand. The Five Eyes nations (which also includes the United States (US)) have a combined military population estimated at 2.6 million, are all developed countries which share a common language, have similarities in society and culture, broadly similar military involvement, and well-resourced military healthcare systems.40 Of the limited systematic reviews in the field, studies on US military populations make up all or the majority of included papers.35,41,42 To address this gap in the literature, the current review focuses on the remaining four Five Eyes nations where there has been limited attention in the literature previously. Additionally, the US was excluded due to the differences in firearms culture, including access, licensing and laws which affects access to means of suicide. For instance, a recent systematic review in the US highlighted that US veterans are at a substantially increased risk of firearm suicide and have higher rates of firearm ownership than the US general population.43
The current systematic review aimed to address these gaps in the literature by identifying risk and protective factors associated with self-harm and suicide behaviours among serving and ex-serving personnel of the UK Armed Forces, Canadian Armed Forces, Australian Defence Force and New Zealand Defence Force.
2. METHOD
2.1. Design
This systematic review was conducted following Cochrane methodology and Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.44 Prior to commencing the review, the review protocol was registered with PROSPERO (CRD42022348867), an online database of systematic review protocols submitted prospectively to maintain research integrity.
2.2. Search Strategy
Seven electronic databases were searched for relevant literature in October 2022: PubMed (including MEDLINE and PubMed Central), Web of Science, Embase (via OVID), Global Health (via OVID), PsycINFO (via OVID), PTSDpubs (formerly PILOTS) and CINAHL. All databases were searched using a combination of pre-defined terms related to: (1) self-harm and suicide behaviours, (2) the military, and (3) geographical locations. The syntax used for the searches is reported in Supplementary 1.
The search included all original, peer-reviewed work, published in English, that identified risk and/or protective factors associated with self-harm and/or suicide behaviours both during (serving personnel) and after (ex-serving personnel/veterans) military service. Restrictions were placed on publication dates from 1st January 2001 to 30th September 2022 to allow for the coverage of the start of the Iraq and Afghanistan conflicts. Additionally, limits of the year 2001 onwards were chosen due to better availability and quality of literature in the field of military medicine.45
PROSPERO was searched to identify completed, ongoing or planned systematic reviews and meta-analyses of relevance. The reference lists of all included studies and other relevant studies were also searched (cross-referencing). At least one expert from each nation working in the field of military self-harm and suicide research was contacted to ensure the key papers from each nation had been identified. If the full text of a paper was not available online, authors were contacted to request access.
2.3. Study Selection Criteria
Eligibility was determined using the following inclusion criteria:
Published in the English language
Original, peer-reviewed work
Published between 1st January 2001 to 30th September 2022 (inclusive)
Studies that reported risk and/or protective factors for outcomes of self-harm, suicidal ideation, suicide attempts and/or completed suicide
Full or part of the reported population sample made up of serving and ex-serving personnel. If military only make up part of the sample, only included papers that reported on outcomes for serving/ex-serving personnel separate from the other population (e.g., general population). Various military populations were explored (if reported) including serving and exserving personnel (Regulars and Reservists) from the Navy, Army, Air Force, and Marines
Population comprising of serving and/or ex-serving personnel from the UK Armed Forces, Canadian Armed Forces, Australian Defence Force or New Zealand Defence Force
Exclusion criteria included:
Study design: qualitative, systematic reviews/meta-analyses, pilot studies, case-control studies, case reports and series, study protocols, grey literature, conference abstracts/papers, dissertations/theses, and editorials
Outcomes of assisted suicide (i.e. the act of deliberately assisting another person to kill themselves46) and suicide bombing (i.e. a terrorist bomb attack in which the perpetrator expects to die whilst killing a number of other people)
Studies that only included a population sample of conscripts (i.e., people compulsorily enlisted into the military), cadets or officer students
For the purpose of this review ex-serving personnel were defined as those who completed a minimum of one day paid employment in the military but no longer serve (following the UK definition).47 Self-harm (also termed as non-suicidal self-injury, self-inflicted injury) was considered to mean intentional self-inflicted injury to the body with the expectation that the injury will lead only to minor physical harm.48 Suicidal ideation was defined as thoughts, ideas, or ruminations about the possibility of ending one’s own life.49 A suicide attempt was considered as a specific episode of self-harming behaviour undertaken with the conscious intention of ending one’s own life.50 Completed suicide was defined as intentional self-harm leading to death.51
2.4. Screening and Data Extraction
After the initial search was conducted, all identified studies were imported into Endnote 20 and screened for duplicates. Where duplicates were found, they were removed. CW independently reviewed the titles and abstracts of all papers. Subsequently, the full papers for the remaining studies identified as relevant were then reviewed. The reference lists of all included papers were manually searched for any additional papers of relevance (i.e., cross-referencing). BC independently performed second reviewer screening on a sample of studies (10% at title/abstract screening stage and 20% at full text screening stage). The reviewers (CW and BC) independently decided which studies met the eligibility criteria to be included in the review and, at full text screening stage, noted any reasons for exclusion. Any discrepancies were resolved through discussion. Interrater reliability was calculated at each screening stage. At the title/abstract screening stage, agreement between reviewers was 99% and at full-text screening stage, agreement was 100%.
Data were extracted independently by CW for all papers deemed relevant for inclusion in the review. The following data were extracted for each included paper:
General information: title, lead author, publication date, journal title, location/country of study
Study characteristics: study aim, study design and methods, response rate (where relevant), sample size, data collection date
Sample characteristics: age (in years), sex distribution, ethnicity, population type (e.g., clinical/non-clinical)
Military characteristics:serving status (serving/ex-serving), engagement type (regular/reserve), service branch, rank, deployment, era of service
Outcomes: type (self-harm, suicidal ideation, suicide attempts, completed suicide), definition and measures
Associated factors
Study findings: conclusions, limitations, future research
Statistical findings are reported in Supplementary 2. Where available, adjusted odds ratios or effect estimates have been presented. If this information was not available, unadjusted results have been reported and clearly identified.
2.5. Quality Assessment
The quality of each included paper was assessed independently by CW using the National Heart, Blood and Lung Institute (NHBLI) tailored quality assessment tools.52 BC independently performed second reviewer quality assessment on a sample of papers (20%). Any discrepancies were resolved through discussion. Papers were not excluded based on their quality, instead, results from the assessment provided additional insight into the quality of research in the field.
3. RESULTS
Database searches returned a total of 4,576 papers, of which 497 were duplicates and were removed (Supplementary 3). In total, the title and abstracts of 4,079 papers were screened, leaving 94 papers to be screened at full-text stage. Experts in the field identified no additional peer-reviewed papers and cross-referencing identified one additional paper of relevance to the review. Overall, 28 papers met the inclusion criteria (Supplementary 4).
3.1. Study Characteristics
Included studies used a range of study designs; the majority were cohort studies (n=23)18,19,32,36–38,53–69 and the remaining papers utilised a retrospective cohort study design (n=5)15,28,70–72 (Table 1). Of the 28 papers included in the review, the majority were from Canada (n=13)55–58,61–63,66–68,71,72 and the UK (n=10)15,18,19,28,53,54,64,65,69,70. The rest of the papers were from Australia (n=5)32,37,38,59,60, but none of the included papers were from New Zealand. Several papers explored samples of ex-serving personnel only (n=12)56–59,61,66–68, followed by serving personnel only (n=8)15,28,37,38,60,62–65,69,70,72 and mixed samples of serving and ex-serving personnel (n=8)18,19,32,36,53,55,71,73. Of the 28 included papers, 24 included mixed male and female samples, however the percentage of male participants typically ranged from around 85% to 99%.15,18,19,28,32,38,53–58,61–72 Of the last four papers, three used all male samples37,59,60 and one did not explicitly report on the sex of the sample directly within the paper.36
3.2. Quality Assessment
Of the 28 papers included in the review, most papers received a quality assessment score of ‘fair’ (n=21), and the remaining papers scored ‘good’ (n=7) (Supplementary 2). Some of the common reasons for scoring ‘fair’ in the quality assessment included lacking a sample size justification or power calculation, not assessing the exposure variables more than once over time, and not using outcome measures which were clearly defined, valid and reliable.
3.3. Reported Self-Harm and Suicide Behaviour Outcomes
Included papers reported on a range of self-harm and suicide behaviour outcomes outlined in Table 1. Four papers reported on self-harm only19,53,54,70, nine reported on suicidal ideation only36,55,56,62–64,68,69,71, two on suicide attempts only38,67 and three on completed suicides only15,28,72. The remaining 10 papers reported on a mix of these outcomes, typically referred to as ‘suicidality’ (often including outcomes of suicidal ideation, suicide plans and suicide attempts)18,32,37,57–61,65,66.
3.3.1. Definitions of Self-Harm and Suicide Behaviour Outcomes
Only six papers provided a definition for the outcomes of interest (Table 1), with the remaining papers failing to report why these terms were not defined. O’Toole et al defined suicidality as “any self-initiated behaviour occurring on a continuum ranging from suicidal ideations, to making a suicide plan, through to a suicide attempts74”.37 Pinder et al defined self-harm in line with the National Clinical Practice Guidelines, “self-harm is self-poisoning or self-injury irrespective of the apparent purpose of the act75”.19 The four remaining papers that provided an outcome definition utilised International Classification of Diseases (ICD) codes (versions 8, 9 and 10) to define and measure the outcome of interest.15,28,70,72
3.3.2. Measurement of Self-Harm and Suicide Behaviour Outcomes
The majority of papers used self-report surveys (n=17)19,32,36,38,53–58,61,64,66–68,71. Two papers collected data via structured clinical interviews62,63 and four papers utilised a mix of self-report surveys and structured clinical interviews18,37,59,60. One study measured the outcomes of interest via a clinical assessment69. The remaining papers used ICD codes to define and measure the outcome of interest15,28,70,72.
A variety of outcome measures were used to collect data on the outcomes of self-harm and suicidal behaviours (Table 1). For instance, ICD codes15,28,70,72, Patient Health Questionnaire-955,71, Clinical Interview Schedule-Revised18,19 and Suicidal Behaviors Questionnaire-Revised64.
Risk Factors
Numerous risk factors for self-harm and suicide behaviours among serving and ex-serving personnel were identified in the review, including factors relating to sociodemographic characteristics, military characteristics, and health [Table 2].
3.3.3. Risk Factors for Self-Harm
Pre-enlistment factors increased the risk of self-harm, including having experienced childhood adversity and abuse than those without these experiences (approximately doubling the risk).19,53 Certain demographic groups were also found to be at more at risk including younger age groups19,53,54 compared to middle/older age groups and those no longer in a relationship (i.e., separated/divorced)53 compared to being in a relationship. There were discrepancies between studies regarding the impact of sex, with two studies reporting the risk of self-harm was increased for females18,53 but one study stated increased risk for males70. Military characteristics associated with increased risk of self-harm included having left service, with ex-serving personnel reported to be as much as three times more at risk of self-harm than serving personnel or non-military population depending on sample explored.18,19,53,70 Other military characteristics that increased the risk of self-harm included being an early service leaver, defined within the relevant papers as <4 years65 and ≤2.5 years70 of service (e.g., unadjusted odds of around 1265).
Several health-related factors also increased risk of self-harm among military serving and ex-serving personnel. The factors with some of the largest effect sizes (ranging from approximately two to eight times more likely) included clinical diagnosis or probable diagnosis of PTSD18,19,70, depression18,19, and anxiety18,70, as well as history of suicidal ideation18. Further, other risk factors for self-harm included a lack of social support53, mental health-related stigmatisation18, perceived practical barriers to care18, negative attitudes to mental illness18, and formal medical help-seeking18 (all approximately tripling risk).
3.3.4. Risk Factors for Suicide Behaviours
Similar risk factors were identified for suicide behaviours (i.e., suicidal ideation, suicide attempts and completed suicide). Pre-enlistment factors, such as experience of childhood adversity and abuse were found to increase the risk of suicide when compared to those without these experiences (around two to seven times more likely).59–61,66,69 Certain demographic characteristics also increased the risk of suicide behaviours, including being single or being separated, divorced, or widowed32,56,59,62, having lower educational attainment (less than university degree level)59, and being unemployed38,69. Another risk factor for suicide behaviours included a lack of social support.55 The influence of age remains unclear, as studies reported a variety of age groups to be at increased risk of suicide behaviours15,28,55,59,72; for instance younger age groups (e.g., 16 to 24 years)28, younger/middle age groups (e.g., <25 to 44 years)72, and a mixture of age groups (i.e., 25 to 45+ years)59.
Numerous military characteristics increased the risk of suicide behaviours, including being medically discharged from service32, having a shorter length of service (i.e., one to four years28)28,55, holding a more junior rank during service59,60,72, serving as a Reservist (vs Regular)18, and shorter intervals since return from last Afghanistan deployment (i.e., <4 years)68. Being an early service leaver was another factor that increased risk of suicide behaviours, defined within the relevant papers as ≤3 years15 and <4 years65,69 of service (e.g., increased risk by as much as eight-fold69). Experience of deployment-related traumatic events36,57,58,61,66–68 and exposure to trauma36,59,60,64,67 were other factors associated with increased risk of suicide behaviours.
Several health-related factors were positively associated with suicide behaviours among military serving and ex-serving personnel. These included number of physical health disorders (small effect size of just over one)62 and number of mental health disorders (odds ranging from around two to around 20 times more likely)56,62,63. More specifically, clinical diagnosis or probable diagnosis of PTSD18,32,36–38,56,61, depression18,32,36,37,56,59–62,71, anxiety18,32,37,56,60,61, alcohol use disorder32,37,56,61, cannabis use37, mood disorder62, insomnia56, or panic attacks61 (ranging from just over one to 15 times more likely). Further, other risk factors for suicide behaviours included a higher number of perceived practical barriers to care, negative attitudes to mental illness, and formal medical help-seeking.18 Lifetime suicidal ideation was reported as a risk factor for suicide attempts (up to 12 times more likely).18
3.4. Protective Factors
Several protective factors were identified in the included studies [Table 2].
3.4.1. Protective Factors for Self-Harm
Several factors were identified that decreased the risk of self-harm, including middle/older age groups18,53,70 (e.g. 35 to 49+ years18) compared to other age groups, and experience of less adversity during childhood (i.e., less than three factors)19. Certain military characteristics reduced the likelihood of self-harm, for example holding a more senior rank compared to junior ranks18,19, being a Reservist compared to Regular53, and having a longer length of service19,70. Further, a higher level of perceived social support was associated with a lower likelihood of self-harm.18,53
3.4.2. Protective Factors for Suicide Behaviours
When considering suicide behaviours (i.e., suicidal ideation, suicide attempts and completed suicide), there were several factors that decreased the risk of suicide behaviours. Certain demographics were significantly associated with reduced risk of suicide behaviours including being female28, middle age groups (e.g. 35 to 54 years compared to those <35 years)69, having a current significant relationship (i.e., being married or in a relationship approximately half as likely than those in ex-relationships)28,59,60, higher educational attainment59,60, and being employed38.
Several military characteristics decreased the risk of suicide behaviours, including holding a more senior rank compared to junior ranks18,28, service in the Royal Airt Force18,28 or Naval Services28 compared to service in the Army, and certain deployment-related factors such as higher cumulative duration of Afghanistan-related deployments (i.e., ≥361 days)68, and less time away on deployment in the past three years (i.e., up to two years)68 (all approximately half as likely). Higher levels of perceived social support18,36 and taking longer than five years to seek support (compared to taking less than five years)69 were also identified factors that reduced the likelihood of suicide behaviours.
4. DISCUSSION
This review identified 28 papers reporting on a wide range of risk and protective factors for self-harm and suicide behaviours among serving and ex-serving personnel of the UK Armed Forces, Canadian Armed Forces, Australian Defence Force and New Zealand Defence Force.
A variety of definitions and measurements were used for the self-harm and suicide behaviour outcomes reported in the included papers of this review. Definition is an important aspect of academic research and clinical practice, yet precise definitions have been contested, and current terminology varies across nations.76 For instance, there are several alternative terms that may be used for self-harm, including non-suicidal self-injury, deliberate self-harm and self-inflicted violence. In the UK, the term deliberate self-harm has been removed from use due to it appearing judgemental and placing blame on the individual.77,78 In the US however, the term non-suicidal self-injury is commonly used. Definitions of mental health conditions and the use of consistent language are an important starting point within good quality research papers and are also important for reducing stigma and encouraging individuals to seek help, particularly relating to self-harm and suicide behaviours.79
Several of the identified risk factors reflected those found among the general population and other occupational groups with an increased risk of traumatic exposures in the line of duty including emergency responders3,80,81, and were consistent across the included geographical regions (UK, Canada, and Australia). Although no relevant peer-reviewed papers from New Zealand were identified, suicide prevention work is being conducted.82 In line with the findings of this review, risk factors for suicide among the New Zealand Defence Force have been reported to include current mental health concerns, acute life stressors and negative attitudes towards help-seeking, and protective factors include positive mental health and social support.82 However, it is important to recognise that some risk and protective factors were not consistent across included studies in this review, and the influence of age, sex, certain military service characteristics, and certain health diagnoses remains unclear. For instance, having served as a Reservist increased the likelihood of suicide behaviours18 but decreased the likelihood of self-harm53, highlighting the need for a holistic approach when supporting the military community. It is important to develop an enhanced understanding about motivations for engaging in self-harm and suicide behaviours, as for some self-harm can be used as a coping mechanism to regulate internal emotions, whereas suicide behaviours may become more apparent when the individual feels they can no longer cope.83
One key risk factor for self-harm was having left service. Potentially the transition from serving to ex-serving is a period of risk and support should be in place during this time. It might be that after leaving service, ex-serving personnel no longer feel like part of the military ‘family’, experience a shift in their sense of self and have difficulty connecting to civilian life84. Alternatively, it could potentially reflect the influence of time as ex-serving personnel are typically older, and therefore, if considering lifetime prevalence and risk, this would capture a longer period of time. Additionally, there may be an underrepresentation among serving personnel as they are potentially more hesitant to report these behaviours due to stigma and fear of negative consequences to their career85; personnel with ill health (physical or mental) may be forced to leave service (i.e. medical discharge) and those who involuntarily separate from the military are at high risk of self-harm and suicide behaviours.31,32,86,87
Aligning with previous international research17,88, another key risk factor for self-harm and suicide behaviours was poor health, in particular probable or clinical diagnosis of PTSD, depression and anxiety. This highlights the importance of early detection of mental health problems, providing adequate care and support to military personnel throughout their military career, and providing continuity of care as they transition out of service. Interestingly, seeking help from formal medical sources was positively associated with lifetime suicide attempts18 and taking more than five years to seek help was negatively associated with suicidal ideation69. It is unlikely that seeking help from formal medical sources has a causal relationship with suicide behaviours. Instead, UK research has found that ex-serving personnel are known to present to clinical services at times of crisis, which might involve an active episode of self-harm or suicide behaviours, therefore, placing them more at risk89. Additionally, those who delay seeking help for longer may have discovered ways to cope with their difficulties on their own, whereas those who seek help sooner may have been in crisis and at higher risk of suicidal ideation69. Clearly, a key challenge relating to self-harm and suicide behaviour risk is prevention. Most developed nations already have suicide prevention strategies in place, including for serving and ex-serving personnel, or as part of the wider mental health strategy90–93. The findings of this review suggest that prevention and intervention strategies should focus on the broader context of improving health and wellbeing (physical, mental, and social health).
Less research focussed on protective factors for self-harm and suicide behaviours among the included military populations. Despite this, one of the most common associations found was with higher levels of perceived social support18,36, indicating that this was likely a salient factor that could help prevent or mitigate the risk of self-harm and suicide behaviours, particularly due to its modifiable nature. This aligns with other international military populations, including the US94, as well as in the general population where social support has been recommended for use in suicide prevention strategies.95,96 One way to bolster social support among military populations is the use of peer support as a preventative strategy. For instance, a guide jointly developed by the UK Ministry of Defence and Samaritans (a suicide prevention charity in the UK) promoted the use of peer support among the UK Armed Forces97, and peer support has also been implemented successfully for US ex-serving personnel using a community-based model98.
Strengths and Limitations
A strength of this review was the search of seven literature databases using a broad search strategy outlined in an a priori PROSPERO approved review protocol. Generally, the protocol was followed but any changes made were reflected by updating the protocol (i.e., removing the limit around the age of the sample). Additionally, a proportion of the eligibility assessment and critical appraisal of papers was conducted by a second, independent reviewer with high inter-rater reliability which strengthened the review. Focus on the Five Eyes nations is also beneficial as this review collates evidence where attention has previously been limited.
However, there were some limitations to note. As with all systematic reviews, the findings of this review were subject to publication bias. Additionally, some potential risk and protective factors received less attention than others, however this review tried to provide the best synthesis of the evidence currently available. Due to the scope of the review, it was not appropriate to conduct a meta-analysis.
There were also several limitations relating to the included papers. First, papers each used different instruments and conceptualisations to define and measure self-harm and suicide behaviours which limited the possibility of comparing findings across studies. However, combining these papers in a review contributes to understanding. Second, there is no universally agreed definition of veterans (i.e., ex-serving personnel) which makes cross-cultural comparison difficult. For instance, in the UK a veteran is defined as someone who has served a minimum of one day paid employment in the UK Armed Forces but no longer serves.47 In Canada, a veteran is regarded to be a person who was on active service during World War I or II as a member of the naval, army or air force of Canada or of similar forces raised in Newfoundland99.
Third, studies did not always report on some military characteristics (such as service branch, rank, deployment experience, era of service and time since leaving service) and demographic characteristics (such as ethnicity and sexuality) which would have been useful for contextualisation and interpretation of the findings. There remains limited understanding of the impact of sex, experiences of ethnic minority and LGBTQ+ personnel. Fourth, included studies used a variety of sample sizes (range n= 144 to 233,803), study procedures, and sample characteristics. For example, the majority of papers relied on self-reported data which may have been subject to recall bias and social desirability bias. Finally, those still serving in the military may have been concerned about their career prospects if they admitted to experience of self-harm and suicide behaviours which may have led to underreporting of experiences.
4.1. Implications
This review has several important implications for policy, clinical practice, and research. The identification of risk and protective factors can be useful to inform military health services and policies including where to target suicide prevention policy to reduce the incidence and impact of suicide. In the UK, this is one of nine key health priority themes laid out in the Defence People Health and Wellbeing Strategy 2022-2027.100 In particular, one of the strategic objectives of the Armed Forces Suicide Prevention Strategy and Action Plan in the UK is to understand high risk groups within the Armed Forces.101
Identifying the risk and protective factors for self-harm and suicide behaviour outcomes is an important aspect of the development and implementation of effective prevention and intervention strategies to protect the mental health and wellbeing of military populations. Evidence on factors related to self-harm and suicide behaviours can inform healthcare practice and service provision for the Armed Forces community by highlighting several at-risk groups which may require additional support. This review suggests that additional support is required during the period of transition from military to civilian life but also highlights the importance of prevention early on in military service to reduce the impact on personnel as they transition out of service. The identified protective factors suggest that prevention and intervention strategies should focus on encouraging help-seeking for mental health problems early on before crisis events occur, as well as promoting social support networks and strengthening connections with family, friends and the community as a whole.
Stronger international research collaborations, such as the Five Eyes Mental Health Research and Innovation Collaborative102, is required to fully understand risk for self-harm and suicide behaviours among serving and ex-serving personnel. Additionally, focus should be placed on modifiable factors (i.e., factors that could reasonably be altered, such as psychiatric symptoms and social support). Although non-modifiable factors can help to identify level of risk, they are of less use during prevention and intervention as they cannot be changed to alter the level of risk. Therefore, particular focus on modifiable risk and protective factors is warranted as they are amenable to therapeutic intervention and can be key to addressing long-term risk due to their adaptive nature. However, it is still important to understand the role of non-modifiable factors as there may be an indirect effect, for example when considering military rank, it might be that officers are at lower risk of self-harm and suicide behaviours because they typically hold higher socio-economic status and higher educational attainment103 which are known protective factors.
This review highlighted a lack of original, peer-reviewed work conducted on this topic in New Zealand which warrants further investigation. Future studies should also focus on conducting longitudinal investigations which distinguish between pre-, peri- and post-service factors in order to identify pathways of self-harm and suicide behaviours, and to ensure support is in place at the right point in the military lifecycle.
4.2. Conclusions
This review highlighted several risk and protective factors for self-harm and suicide behaviours which warrant attention. Adequate care and support are necessary for serving and ex-serving personnel who may be at risk of experiencing self-harm or suicide behaviours. Particular focus should be placed on implementing prevention strategies early on in military service to reduce the impact on personnel as they transition out of service. The identified protective factors suggest that prevention and intervention strategies should promote social networks as a key source of support for military personnel. This review highlighted the lack of peer-reviewed research on self-harm and suicide behaviours among the New Zealand Defence Force. Research should continue to progress towards understanding and preventing self-harm and suicide behaviours among military populations.
Data Availability
Data availability is not applicable to this article as no new data were created or analysed in this study.
DECLARATIONS
Ethical approval and consent to participate
N/A
Consent for publication
N/A
Availability of data and materials
N/A.
Competing interests
CW is currently in receipt of a funded PhD studentship via Phase 4 of the King’s Centre for Military Health Research Health and Wellbeing Cohort Study funded by the Office for Veterans’ Affairs (OVA), Cabinet Office, UK Government. BC is currently in receipt of a funded PhD studentship from The Economic and Social Research Council (ESRC), UK Government. AS is a full-time member of the British Army seconded to King’s College London. NTF is partly funded by the United Kingdom’s Ministry of Defence (MOD). MLS is fully funded by a grant from the OVA. SAMS is supported by the National Institute for Health and Care Research (NIHR) Maudsley Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and the National Institute for Health and Care Research, NIHR Advanced Fellowship, Dr Sharon Stevelink, NIHR300592. The views expressed in this publication are those of the authors and not necessarily those of the OVA, the ESRC, the British Army, the MOD, the NHS, or the NIHR.
Funding
This work is part of a PhD nested within Phase 4 of the King’s Centre for Military Health Research Health and Wellbeing Cohort Study and funded by the OVA. For the purposes of open access, the author has applied a Creative Commons Attribution (CC BY) licence to any Accepted Author Manuscript version arising from this submission.
Authors’ Contributions
CW proposed the review. NTF and SAMS secured the funding for Phase 4 of the King’s Centre for Military Health Research Health and Wellbeing Cohort Study. CW, MLS and SAMS contributed to the design of the review. CW completed all stages of screening and data extraction. BC was the second rater for 10% of results at title/abstract screening and 20% at full text screening. CW completed the quality assessment for all included papers and BC performed second rater assessment for 20% of included papers. CW wrote the first draft, and all other authors (BC, AS, NTF, MLS, SAMS) contributed to each revision and approved the final manuscript.
Acknowledgements
We would like to thank Colonel Clare Bennett, Chief Mental Health Officer New Zealand Defence Force, for sharing relevant work being conducted in New Zealand.
Footnotes
↵^ Joint last authors
REFERENCES
