Deployment-Related Traumatic Events and Suicidal Behaviours in a Nationally Representative Sample of Canadian Armed Forces Personnel

Abstract

French

Objective:

Worldwide, there has been substantial controversy with respect to whether military deployment is a risk factor for suicidal behaviour. The present study examined the relationship between lifetime exposure to deployment and deployment-related traumatic events (DRTEs) and past-year suicidal ideation (SI), suicidal plans (SP), and suicidal attempts (SA).

Method:

Data were analysed from the 2013 Canadian Forces Mental Health Survey (8161 respondents; response rate, 79.8%; aged 18-60 years). A total of 12 individual items assessed exposure to DRTEs (e.g., combat, witnessing human atrocities, feeling responsible for the death of Canadian or ally personnel, knowing someone who was injured or killed). We examined each individual DRTE type as well as the number of types of DRTEs in relation to suicidal behaviour.

Results:

Lifetime deployment was not significantly associated with suicidal behaviour. In models adjusted for sociodemographic variables, most of the individual DRTE items and the DRTE count variable were significantly associated with suicidal behaviours (adjusted odds ratio ranged between 1.10 and 5.32). When further adjusting for child abuse exposure, these associations were minimally attenuated, and some became nonsignificant. In models adjusting for mental disorders and child abuse, most DRTEs and number of types of DRTEs became nonsignificant in relation to SI, SP, and SA.

Conclusions:

Active military personnel exposed to increasing number of DRTEs are at increased risk for SI, SP, and SA. However, most of the association between DRTEs and suicidal behaviour is accounted for by child abuse exposure and mental disorders.

Keywords

deployment trauma military suicidal behaviour Canada

Introduction

Some nations have seen substantial increases in the prevalence of suicidal behaviour in active duty military and veterans.^1,2 Suicidal ideation (SI), suicidal plans (SP), and suicidal attempts (SA) are strong risk factors for death by suicide.³ Recent work has demonstrated a substantial lifetime prevalence of SI (15.4%), SP (6.1%), and SA (3.0%) in the 2013 nationally representative sample of Canadian active military personnel.⁴ These rates are very similar to the Canadian general population with lifetime SI (13.3%), SP (4.6%), and SA (3.5%).⁴

Although there are many theoretical models that explain suicidal behaviour (Durkheim,⁵ Mann et al.,⁶ Hawton et al.,⁷ and Zamorski⁸), Joiner’s interpersonal theory of suicide⁹ is particularly relevant among military personnel. Joiner’s theory posits that an individual will die by suicide if he or she has both the desire to die and the acquired ability to carry out suicidal behaviour. He posits that the desire of death develops when the person perceives himself or herself to be a burden on others and/or experiences a sense of loneliness. The acquired ability to attempt suicide occurs through habituation of painful experiences such as previous self-harm behaviour, physical abuse, physical fights, and indirect exposure to death and loss (e.g., physicians and soldiers).

Empirical literature on risk factors for suicidal behaviour among military personnel and veterans can be understood in relation to sociodemographic and military factors,^1,10 predeployment factors (e.g., child abuse exposure),¹¹ deployment-related trauma (e.g., combat exposure),^12
–14 and postdeployment factors (e.g., mental disorders, poor social supports, and separation from service).^15

–18

There has been substantial controversy with respect to the association between deployment to combat and peacekeeping missions and suicidal behaviour. Many studies have examined an overall deployment variable (e.g., ever deployed vs. never deployed) in relation to suicide risk. With this approach, most studies have found no relationship between deployment and suicidal behaviour.^18,19 A recent meta-analysis of 22 studies found a modest effect, 25% increase in risk, of deployment associated with suicidal behaviour.²⁰ However, many argue that exposure to traumatic events during deployment (which varies substantially from deployment to deployment and from person to person) is more important to study in relation to suicidal behaviour than a dichotomous (ever vs. never) deployment variable.²¹ A smaller number of studies examining exposure to specific events during deployment (e.g., witnessing atrocities and dead bodies) have found a larger association between traumatic events and suicidal behaviour.^20,22,23 In summary, these studies suggest the need to examine in greater detail the types of events experienced by military personnel during deployment to better understand their association with suicidal behaviour.

There are several limitations of the current literature on deployment and suicide risk. First, many studies have not captured details of exposure to specific traumatic experiences during deployment.^1,19 To our knowledge, previous work has not examined deployment history, exposure to specific traumatic events, and increasing number of traumatic events during deployment in relation to suicidal behaviour in the same population. Second, many studies did not assess or adjust for child abuse exposure, which is emerging as an important premilitary risk factor for suicidal behaviour.¹¹ Third, while some studies have adjusted for mental disorders when examining the association between deployment-related traumatic events (DRTEs) and suicide risk,^2,19 others have not.¹² One could argue that controlling for mental disorders in these models is adjusting for a variable that is part of the causal pathway. For example, DRTEs may increase the likelihood of developing a mental disorder, which in turn increases the risk for suicidal behaviour. However, due to the cross-sectional nature of most studies in the area, it is difficult to test this hypothesis. On the other hand, from a public health perspective, it is also important to know whether DRTEs have an independent effect on suicidal behaviour after adjusting for mental disorders. Such a relationship would suggest the need for postdeployment screening and intervention methods to go beyond screening for mental disorders and also include exposure to DRTEs. Furthermore, it is also possible that DRTEs and mental disorders have a synergistic effect on suicidal behaviour. To our knowledge, no study has examined the interaction between DRTEs and mental disorders in relation to suicidal behaviour.

To address these important gaps in the literature, we analysed data from a nationally representative sample of active military personnel (the 2013 Canadian Forces Mental Health Survey).⁴ This in-person interview survey assessed 12 types of potentially traumatic events during deployment, past child abuse experiences, mental disorders, and SI, SP, and SA.

In this study, we examined lifetime deployment, exposure to individual, and increasing number of types of DRTEs in relation to SI, SP, and SA. We hypothesise that certain DRTEs (e.g., witnessing atrocities and death)¹³ will remain independently associated with suicidal behaviour even after adjustment for sociodemographic variables, mental disorders, and child abuse exposure. We also hypothesised that there will be a significant interaction between DRTEs and mental disorders, elevating the risk for suicidal behaviour.

Methods

Survey

Data were drawn from the 2013 Canadian Forces Mental Health Survey (CFMHS) (N = 8161, response rate = 79.8%). Detailed methodology of this national survey has been published elsewhere.^4,11,24 The CFMHS comprised a representative sample of the Regular Force military personnel (n = 6692, approximately 46% of whom had deployed in support of the mission in Afghanistan) and a sample of Reserve Force personnel (n = 1469) who deployed in support of the mission in Afghanistan.²⁴ Reservists who had not deployed in support of the mission in Afghanistan were not included in the survey. The survey was voluntary, and privacy and confidentiality of the respondents are ensured under the Statistics Act.²⁵ Respondents provided informed consent. The authors obtained approval from Statistics Canada to conduct this analysis.

Measurements

Suicidal behaviours

Suicide-related behaviours consisted of SI, SP, and SA, which were assessed based on past-year and lifetime time frames. Respondents were asked if they had 1) seriously thought about committing suicide or taking their own life, 2) made a plan for committing suicide, or 3) attempted suicide or tried to take their own life. Age of onset data were not available for suicidal behaviour or DRTEs. To address this limitation, and consistent with previous literature, we examined the impact of lifetime DRTEs on past-year suicidal behaviour.⁶

Deployment and DRTEs

Lifetime deployment was based on whether the respondent endorsed any deployment (yes or no). There were 3 sections in the survey where participants were asked about DRTEs:

Deployment-specific CFMHS 2013 items

DRTEs included the following: 1) known someone who was seriously injured or killed; 2) found yourself in a life-threatening situation where you were unable to respond because of rules of engagement; 3) ever been injured; 4) ever seen ill or injured women or children who you were unable to help; 5) ever received incoming artillery, rocket, or mortar fire; 6) ever felt responsible for the death of Canadian or ally personnel; 7) ever had a close call (e.g., shot or hit but protective gear saved you); and 8) ever had difficulties distinguishing combatants from noncombatants. These items were selected from a longer combat exposure inventory,^26,27 based on conceptual and empirical considerations.²⁶ Personnel indicating that they had not ever been deployed were coded as having zero DRTEs.

Lifetime posttraumatic stress items

We also examined items from the traumatic events section of the posttraumatic stress disorder (PTSD) Composite International Diagnostic Interview (CIDI) trauma inventory²⁸ that were likely to have occurred during deployment: 1) participated in combat, 2) peacekeeping or relief worker in a war zone, and 3) witnessing atrocities or massacres such as dead bodies. These items have been used previously.¹²

Deployment-related sexual assault

Lifetime sexual assault was assessed using 2 items also on the PTSD CIDI trauma inventory relating to ever having experienced 1) forced sexual activity or 2) unwanted sexual touching. Those endorsing each of these items were asked if any of the underlying events had occurred while deployed. Endorsing either item was coded as deployment-related sexual assault (yes or no).

We created a count variable (0-12) to examine the impact of increasing number of types of DRTE on the risk for suicidal behaviour. Since this a unique combination of DRTE items, we tested the reliability of the DRTE scale. We found high reliability with an α of >0.8. We also did sensitivity analyses with an 8-item DRTE scale that did not include the 3 PTSD module items and sexual trauma. The results of the 12-item and 8-item scales were consistent.

Child abuse experiences

Items from the Childhood Experiences of Violence Questionnaire (CEVQ)²⁹ were used to measure physical abuse and exposure to intimate partner violence (IPV). Childhood abuse items pertained to abuse that occurred before the age of 16 years. Binary classifications of child abuse (present/absent) were made based on the CEVQ recommendations: physical abuse was present if 1 or more of 3 variables met the following threshold criteria: 1) being slapped on the face, head, or ears or hit or spanked with something hard 3 times or more; 2) being pushed, grabbed, shoved, or having something thrown at the respondent to hurt him or her 3 times or more; and 3) being kicked, bit, punched, choked, burned, or physically attacked 1 time or more. Exposure to IPV was measured by asking the respondent if he or she had seen or heard parents, stepparents, or guardians hitting each other or another adult in the home 3 times or more. Similar to previous work,³⁰ sexual abuse was assessed using 2 items and was classified as present if either of the following occurred 1 time or more: 1) attempted or being forced into unwanted sexual activity by being threatened, held down, or hurt in some way and/or 2) sexually touched, meaning unwanted touching or grabbing, kissing, or fondling against the respondent’s will. A respondent was classified as having experienced any child abuse if he or she met the threshold criteria for at least 1 of the 3 types of abuse.¹¹ A count variable was computed based on the total number of types of abuse experienced (ranging from 0 to 3).³⁰

Mental disorders

The World Mental Health Survey CIDI²⁸ for DSM-IV was used to assess the presence of the following lifetime and past-year disorders: depression, generalised anxiety disorder, panic disorder and/or panic attacks, posttraumatic stress disorder, and alcohol abuse or dependence. For this article, we combined all disorders together to create a dichotomous variable to assess the presence of any past-year mental disorder (yes/no).

Sociodemographic covariates

The sociodemographic covariates included age, sex, race, education, household income, marital status, military force type (Regular or Reserve), and military rank category (junior noncommissioned member [NCM], senior NCM, and officer).

Statistical Analysis

To ensure population representation, statistical weights were applied. Bootstrapping was performed for variance estimation to account for the complex sample. First, descriptive statistics were calculated for each variable used in the analysis. Second, based on previous literature suggesting a differential response between sex and DRTE exposure,^31,32 we tested for effect modification between sex and DRTEs. We did not find any of these interaction terms significant in the prediction of suicidal behaviour. Third, a series of separate multivariable logistic regression models were computed to determine if individual DRTEs and number of types of DRTEs increased the likelihood of suicide-related outcomes. Separate regressions were conducted for lifetime deployment, each DRTE, and the DRTE count variable. In the first adjusted models (adjusted odds ratio [AOR]–1), we included sociodemographic variables, including age, sex, race, education, household income, marital status, military force type, and military rank. In the second adjusted models (AOR-2), we included all variables in AOR-1 and added a count variable of child abuse experiences. In the third adjusted models (AOR-3), we included all variables in AOR-1 and AOR-2 and added the dichotomous any past-year mental disorder variable. Finally, we tested for interactions between the DRTE count variable and mental disorders in relation to suicidal behaviour.

Results

Table 1 shows the prevalence of all the variables in the sample, including sociodemographic variables, DRTEs, child abuse experience, and mental disorders. Among the DRTEs, exposure to combat, known someone seriously injured or killed, and received incoming artillery, rocket, or mortar fire were the 3 most common DRTEs, while sexual trauma during deployment was the least common DRTE.

Table 1.

Prevalence and Means for the Independent and Dependent Variables in the Sample (N = 8161).^a

Study Variables	Prevalence (%) or Mean	95% CI
Dependent variables
Past-year suicide ideation	4.4	3.9-4.9
Past-year suicide plan	1.7	1.4-2.1
Past-year suicide attempt	0.35	0.2-0.6
Lifetime deployment
Deployment-related traumatic events: PTS items (lifetime)
Combat exposure	37.1	36.2-38.0
Peacekeeping or relief worker in war zone	35.4	34.5-36.3
Witness atrocities	15.2	14.4-16.0
Deployment-related traumatic events: CF 2013 items (deployment specific)
Known someone seriously injured or killed	39.3	38.4-40.2
In a life-threatening situation but were unable to respond due to rules of engagement	18.1	17.4-18.9
Ever been injured	15.0	14.3-15.8
Seen ill or injured women or children that you were unable to help	24.9	24.0-25.8
Received incoming artillery, rocket, or mortar fire	36.8	36.1-37.6
Felt responsible for the death of Canadian or ally personnel	3.8	3.4-4.2
Had a close call (e.g., shot or hit but protective gear saved you)	13.9	13.2-14.7
Had difficulty distinguishing combatants from noncombatants	20.7	19.9-21.5
Military sexual trauma
Military sexual trauma during CF deployment	0.7	0.5-0.9
Total number of DRTEs
Total DRTEs 12 items: lifetime, deployment specific, and military sexual trauma (mean)	2.61	2.56-2.66
Covariates
Age (mean)	35.5	35.3-35.7
Sex
Male	86.4	85.6-87.2
Female	13.6	12.8-14.4
Race
White	90.1	89.3-90.9
Other	9.9	9.2-10.7
Marital status
Married or common-law	65.2	64.1-66.3
Single/separated/divorced/widowed	34.8	33.7-35.9
Education
High school or less	29.3	28.3-30.4
Some postsecondary school	49.2	48.1-50.4
University degree (bachelor’s or higher)	21.4	20.9-22.0
Income
Less than $80,000	35.5	34.3-36.6
$80,000 or more	64.5	63.4-65.6
Military rank
Junior NCM	54.6	54.4-54.8
Senior NCM	24.3	24.1-24.6
Officer	21.0	20.9-21.1
Military force type
Regular Forces	93.5	93.5-93.5
Reserve Forces	6.5	6.5-6.5
Any past-year mental disorder
No	76.9	75.8-77.9
Yes	23.1	22.1-24.2
Number of types of child abuse experienced (mean)	0.62	0.60-0.63

^aCF, Canadian Forces; CI, confidence interval; DRTE, deployment-related traumatic event; NCM, noncommissioned member; PTS, Post-traumatic stress.

Table 2 presents the regression models for lifetime deployment and individual DRTEs in relation to past-year SI. Lifetime deployment was not significantly associated with SI in any of the 3 models. In models adjusting for sociodemographic variables (AOR-1), all the individual DRTEs, with the exception of sexual trauma, were independently associated with SI (AOR between 1.31 and 2.92). In models that added child abuse exposure (AOR-2), 6 of the 12 DRTEs items became nonsignificant (AOR between 1.40 and 2.51). In AOR-3, adjusting for any mental disorder, all the DRTEs with the exception of one (“felt responsible for death of Canadian or ally personnel”) became nonsignificantly associated with SI. The DRTE count variable was significantly associated with SI in AOR-1 and AOR-2 but became nonsignificantly associated with SI in AOR-3.

Table 2.

Lifetime Deployment and Deployment-Related Traumatic Events and Past-Year Suicide Ideation.^a

	Model 1: SES and Military Covariates	Model 2: Add Number of Types of Child Abuse	Model 3: Add Any Past-Year CIDI Disorder
	AOR-1 (95% CI)	AOR-2 (95% CI)	AOR-3 (95% CI)
Lifetime deployment	1.02 (0.74-1.40)	1.02 (0.74-1.41)	0.84 (0.60-1.16)
Deployment-related traumatic events
Combat exposure (lifetime)	1.50 (1.15-1.95)	1.40 (1.07-1.82)	1.08 (0.82-1.43)
Peacekeeping (lifetime)	1.33 (1.00-1.76)	1.29 (0.98-1.71)	0.93 (0.69-1.25)
Witnessed atrocities (lifetime)	1.48 (1.08-2.05)	1.33 (0.95-1.85)	1.01 (0.72-1.43)
Known someone injured or killed (CF deployment)	1.55 (1.18-2.03)	1.46 (1.11-1.92)	1.08 (0.80-1.44)
Ever in life-threatening situation and unable to respond because of rules of engagement (CF deployment)	2.05 (1.53-2.73)	1.90 (1.42-2.56)	1.30 (0.95-1.78)
Ever been injured (CF deployment)	1.84 (1.36-2.49)	1.65 (1.20-2.25)	1.08 (0.79-1.49)
Seen ill or injured women or children (CF deployment)	1.38 (1.07-1.79)	1.25 (0.96-1.63)	0.86 (0.65-1.13)
Received incoming artillery, rocket, or mortar fire (CF deployment)	1.31 (1.00-1.72)	1.27 (0.96-1.66)	0.97 (0.73-1.29)
Felt responsible for death of Canadian or ally personnel (CF deployment)	2.92 (1.99-4.28)	2.51 (1.68-3.75)	1.64 (1.08-2.51)
Had a close call but protective gear saved you (CF deployment)	1.50 (1.09-2.06)	1.34 (0.97-1.85)	0.95 (0.68-1.31)
Had difficulty distinguishing combatants from noncombatants (CF deployment)	1.51 (1.16-1.97)	1.42 (1.08-1.86)	0.98 (0.73-1.31)
MST during a CF deployment	3.01 (0.91-9.94)	2.57 (0.76-8.75)	1.96 (0.57-6.73)
Total number of DRTEs (12 items: 8 CF deployment, 3 PTS, and MST)	1.10 (1.05-1.15)	1.08 (1.03-1.13)	1.01 (0.96-1.06)

^aAOR-1 models adjust for age, sex, race, marital status, education, income, military force type, and military rank. AOR-2 models adjust for all AOR-1 covariates and number of types of child abuse experienced. AOR-3 models adjust for all AOR-1 covariates, number of types of child abuse experienced, and any past-year mental disorder. AOR, adjusted odds ratio; CF, Canadian Forces; CI, confidence interval; CIDI, Composite International Diagnostic Interview; DRTE, deployment-related traumatic event; MST, military sexual trauma; PTS, Post-traumatic stress; SES, sociodemographic covariates.

Table 3 presents the relationship between lifetime deployment and individual DRTEs and past-year SP. Lifetime deployment was not significantly associated with SP in any of the 3 adjusted models. There was insufficient power to test the relationship between deployment-related sexual assault in relation to SP. In AOR-1 models, all the individual DRTEs were significantly associated with SP (AOR between 1.76 and 3.59). In AOR-2 models, where child abuse exposure was included, most of the DRTEs remained significantly associated with SP (AOR between 1.54 and 2.81). However, in AOR-3 models, where any mental disorder was included, all the DRTEs became nonsignificant. The DRTE count variable was significantly associated with SP in AOR-1 and AOR-2 but became nonsignificantly associated with SP in AOR-3.

Table 3.

Deployment and Deployment-Related Traumatic Events and Past-Year Suicide Plans.^a

	Model 1: SES and Military Covariates	Model 2: Add Number of Types of Child Abuse	Model 3: Add Any Past-Year CIDI Disorder
	AOR-1 (95% CI)	AOR-2 (95% CI)	AOR-3 (95% CI)
Lifetime deployment	1.07 (0.68-1.69)	1.09 (0.69-1.73)	0.87 (0.54-1.39)
Deployment-related traumatic events
Combat exposure (lifetime)	1.50 (0.996-2.27)	1.36 (0.88-2.08)	0.98 (0.63-1.52)
Peacekeeping (lifetime)	2.08 (1.34-3.22)	1.99 (1.28-3.10)	1.41 (0.89-2.25)
Witnessed atrocities (lifetime)	2.04 (1.27-3.30)	1.74 (1.03-2.92)	1.39 (0.83-2.30)
Known someone injured or killed (CF deployment)	1.80 (1.20-2.71)	1.67 (1.10-2.53)	1.15 (0.74-1.77)
Ever in life-threatening situation and unable to respond because of rules of engagement (CF deployment)	2.49 (1.60-3.87)	2.21 (1.39-3.50)	1.42 (0.87-2.30)
Ever been injured (CF deployment)	1.79 (1.07-2.97)	1.47 (0.85-2.53)	0.96 (0.56-1.64)
Seen ill or injured women or children (CF deployment)	1.76 (1.19-2.60)	1.54 (1.02-2.33)	1.01 (0.66-1.53)
Received incoming artillery, rocket, or mortar fire (CF deployment)	1.78 (1.21-2.62)	1.71 (1.15-2.54)	1.26 (0.84-1.90)
Felt responsible for death of Canadian or ally personnel (CF deployment)	3.59 (1.97-6.54)	2.81 (1.44-5.46)	1.78 (0.92-3.44)
Had a close call but protective gear saved you (CF deployment)	2.17 (1.34-3.51)	1.83 (1.11-3.04)	1.33 (0.81-2.19)
Had difficulty distinguishing combatants from noncombatants (CF deployment)	1.94 (1.31-2.89)	1.80 (1.20-2.71)	1.18 (0.77-1.80)
MST during a CF deployment	NR	NR	NR
Total number of DRTEs (12 items: 8 CF deployment, 3 PTS, and MST)	1.15 (1.07-1.23)	1.12 (1.04-1.20)	1.04 (0.97-1.11)

Table 4 presents the relationship between lifetime deployment and individual DRTEs and past-year SA. Lifetime deployment was not associated with SA in any of the 3 models. In AOR-1, 7 of the 12 DRTE items were independently associated with SA (AOR between 1.92 and 5.32). In AOR-2, only 4 of the 12 items were significantly associated with SA (AOR between 2.81 and 5.07). In AOR-3, only “being unable to respond to a life-threatening situation because of rules of engagement” remained significantly associated with SA. The DRTE count variable was significantly associated with SA in AOR-1 and AOR-2 but became nonsignificantly associated with SA in AOR-3.

Table 4.

Deployment-Related Traumatic Events and Past-Year Suicide Attempts.^a

	Model 1: SES and Military Covariates	Model 2: Add Number of Types of Child Abuse	Model 3: Add Any Past-Year CIDI Disorder
	AOR-1 (95% CI)	AOR-2 (95% CI)	AOR-3 (95% CI)
Lifetime deployment	1.09 (0.40-2.97)	1.10 (0.41-3.00)	0.89 (0.33-2.40)
Deployment-related traumatic events
Combat exposure (lifetime)	3.16 (1.22-8.12)	3.02 (1.14-7.96)	2.29 (0.85-6.18)
Peacekeeping (lifetime)	2.92 (1.21-7.08)	2.81 (1.16-6.81)	2.03 (0.81-5.07)
Witnessed atrocities (lifetime)	3.12 (1.12-8.71)	2.92 (0.99-8.59)	2.13 (0.76-5.98)
Known someone injured or killed (CF deployment)	2.25 (0.92-5.52)	2.18 (0.87-5.47)	1.58 (0.62-4.02)
Ever in life-threatening situation and unable to respond because of rules of engagement (CF deployment)	5.32 (2.13-13.32)	5.07 (1.96-13.12)	3.34 (1.27-8.78)
Ever been injured (CF deployment)	3.14 (1.04-9.46)	2.89 (0.92-9.02)	1.78 (0.56-5.64)
Seen ill or injured women or children (CF deployment)	2.07 (0.78-5.46)	1.95 (0.71-5.39)	1.29 (0.47-3.50)
Received incoming artillery, rocket, or mortar fire (CF deployment)	1.95 (0.84-4.57)	1.93 (0.82-4.53)	1.43 (0.62-3.29)
Felt responsible for death of Canadian or ally personnel (CF deployment)	NR	NR	NR
Had a close call but protective gear saved you (CF deployment)	4.48 (1.58-12.72)	4.17 (1.40-12.39)	2.77 (0.95-8.06)
Had difficulty distinguishing combatants from noncombatants (CF deployment)	1.92 (1.30-2.83)	2.15 (0.73-6.36)	1.41 (0.48-4.11)
MST during a CF deployment	NR	NR	NR
Total number of DRTEs (12 items: 8 CF deployment, 3 PTS, and MST)	1.25 (1.08-1.45)	1.23 (1.05-1.44)	1.14 (0.99-1.32)

We also tested interaction terms between mental disorders and the DRTE count variable for each suicidal behaviour. There was no significant interaction between number of types of DRTEs and mental disorders for the outcomes of SI and SP. There was not enough power to test the interactions between number of types of DRTEs and mental disorders for SA.

Discussion

The present study provides important new information on the controversial relationship between deployment and suicidal behaviour. There are 4 key findings from this research. First, lifetime history of deployment was not significantly associated with past-year SI, SP, and SA. Second, we found a relationship between exposure to multiple individual DRTEs and number of types of DRTEs and SI, SP, and SA. Third, child abuse and DRTEs were significantly correlated, but the relationships between DRTEs and suicidality persisted even after adjustment for child abuse. Fourth, we demonstrated that most of the relationship between individual DRTEs and suicidal behaviour was accounted for by mental disorders. To our knowledge, this is the most comprehensive and in-depth examination of the impact of specific traumatic events during deployment in relation to SI, SP, and SA.

Our findings are consistent with previous literature in this area showing that exposure to traumatic events during deployment has greater explanatory value for suicidal behaviour than an omnibus “ever deployed” variable.²¹ Similar to the established dose-response relationship between deployment experiences and posttraumatic stress disorder and other mental disorders,^33,34 this study underscores that a similar dose-response relationship exists between deployment experiences and suicidal behaviour. We also found that certain DRTEs had stronger associations with suicidal behaviour than others. Military personnel who reported being responsible for the death of a Canadian ally or being unable to respond to a life-threatening situation because of rules of engagement were at substantially higher odds of reporting suicidal behaviour. These particular events might be especially stressful in creating anxiety and guilt that might increase the risk of SI, SP and SA. These findings provide empirical support for Joiner’s interpersonal theory of suicide where exposure to death may be an important part of the development of suicidal behaviour.³⁵

In this study, we also demonstrated the importance of adjusting for the effects of child abuse experiences in examining the relationship between DRTE and suicidal behaviour. Early experiences of trauma are well known to increase the vulnerability to both mental disorders and suicidal behaviour.^34,36 Our previous work has shown that military personnel have a high prevalence of child abuse exposure, which is strongly associated with suicidal behaviour.¹¹ It is important to underscore that many of the specific DRTEs and the DRTE count variable remained significantly associated with suicidal behaviour in models adjusted for child abuse exposure. Child abuse exposure also remained significantly associated with suicidal behaviour. These findings support the importance of a dose-response relationship between DRTEs and suicidal behaviour among military personnel. However, the association between number of DRTEs and suicidal behaviour became nonsignificant when adjusting for mental disorders.

Previous studies in the literature have sometimes adjusted for the effects of mental disorders,¹⁹ while others have not.¹² Thus, our study is consistent with earlier work that has found a significant relationship between DRTEs and suicidal behaviour without accounting for the effects of mental disorders. Furthermore, most studies that have adjusted for the effects of mental disorders have found small or nonsignificant relationships between deployment and suicide. Future studies could consider examining whether specific mental disorders mediate the relationship between DRTEs and suicidal behaviour.

The findings of this study need to be considered in the context of several strengths and limitations. The strengths include the use of 1) a large, contemporary, representative sample; 2) state-of-the-art epidemiological methods using the World Health Organization CIDI²⁸; and 3) use of a broad list of DRTE items.

There are 5 major limitations of the current study. First, the survey was a cross-sectional design that did not assess information related to the age of onset of the deployment, mental disorders, and suicidal behaviour. To address this issue, similar to previous work,¹² we used lifetime DRTEs with past-year suicide-related outcomes. We also adjusted for mental disorders during the past year. It would have been ideal to have age of onset of mental disorders, traumatic events, and suicidal behaviour such that we could explore the temporal relationships between these 3 variables. Our current study is limited by the fact that mental disorders and suicidal behaviour could have had their onset prior to the deployment. Second, the current survey used a single-item measure of suicide attempt. Some of the analysis was limited by power, especially with the SA outcomes. We did not have detailed information about the number and lethality of SA. Although the item used in the current survey is part of the CIDI used worldwide,³⁷ recent work is showing that a single-item measure of suicide attempts may have limited reliability and validity.^38,39 For example, in 1 study, only 60% of people who endorsed a single item of “Have you ever attempted suicide?” were classified as having a history of suicide attempt based on follow-up questions.³⁸ Third, the composite DRTE measure was uniquely created from several items assessed in the survey. Although we found high reliability of the scale and the items have face validity, the items do not measure severity and perceived threat during exposure to the DRTE. Perceived threat during exposure to trauma has been strongly associated with subsequent mental health problems.⁴⁰ Fourth, we also were not able to account for the effects of other adult traumatic events because it was not clear whether the traumatic event reported occurred during deployment. Finally, the findings of this study may not be generalizable to deaths by suicide. We want to highlight that in our study, and in other studies, SA and suicide deaths have a low prevalence. So, although the AORs in this study were not significant, many AORs were greater than 2.0, showing that the data may not have enough power to accurately examine these outcomes. To address these limitations, there is a need for longitudinal mental health survey research with more detailed assessment of DRTEs, perceived threat during exposure to DRTEs, and temporal information about the onset of mental disorders, DRTEs, and suicidal behaviour.

Clinical and Policy Implications

This study adds important new information on the relationship between deployment and DRTEs and suicidal behaviour. Although deployment itself is not associated with suicidal behaviour, increasing levels of exposure of DRTEs is associated with suicidal behaviour. Predeployment resilience training strategies that improve the capacity of military personnel to face DRTEs may reduce suicidal behaviour. Cognitive processing therapy and prolonged exposure therapy have been shown to be effective in reducing acute stress disorder symptoms⁴¹ and PTSD symptoms.⁴² Conceptually, these psychological interventions could be adapted for use during the deployment phase. However, there is no strong, empirical literature to support overall resilience training⁴³ or predeployment resilience training.⁴⁴ We suggest that early recognition of exposure to DRTEs during and after deployment may reduce the risk of postdeployment mental disorders and suicidal behaviour.

Conclusions

Serving military personnel exposed to increasing number of types of DRTEs are at increased risk for SI, SP, and SA. There is also variability of the relationship between some individual DRTEs and the risk for suicidal behaviour. These findings suggest the potential importance of early recognition and treatment of military personnel exposed to high levels of DRTEs and postdeployment mental disorders. Indeed, the Canadian Armed Forces has educational and screening programs that assess both deployment-related trauma and mental disorders in place for that very reason.

Footnotes

Acknowledgements

Statistics Canada collected and provided the data for academic purposes, but the analyses are the sole responsibility of the authors. The opinions expressed do not represent the views of the Department of National Defence (Canada), the Canadian Armed Forces, or Statistics Canada.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by funding from the Government of Canada and a CIHR Foundation Grant 333252 (Sareen).

References

Schoenbaum

Kessler

Gilman

. Predictors of suicide and accident death in the army study to assess risk and resilience in servicemembers (Army STARRS). JAMA Psychiatry. 2014;71(5):493.

Nock

Stein

Heeringa

. Prevalence and correlates of suicidal behavior among soldiers: results from the arm study to assess risk and resilience in servicemembers (Army STARRS). JAMA Psychiatry. 2014;71(5):514.

Nock

Borges

Bromet

. Suicide and suicidal behavior. Epidemiol Rev. 2008;30(1):133–154.

Sareen

Afifi

Taillieu

. Trends in suicidal behavior and use of mental health services in Canadian military and civilian populations. CMAJ. 2016;188(11):1–7.

Durkheim

. Suicide: a study in sociology. London (UK): Routledge & Keagan; 1952.

Mann

Apter

Bertolote

. Suicide prevention strategies. JAMA. 2005;294(16):2064.

Hawton

Saunders

KEA

O’Connor

. Self-harm and suicide in adolescents. Lancet. 2012;379(9834):2373–2382.

Zamorski

. Suicide prevention in military organizations. Int Rev Psychiatry. 2011;23(2):173–180.

Rudd

Berman

Joiner

. Warning signs for suicide: theory, research, and clinical applications. Suicide Life Threat Behav. 2006;36(3):255–262.

10.

Nock

Deming

Fullerton

. Suicide among soldiers: a review of psychosocial risk and protective factors. Psychiatry. 2013;76(2):97–125.

11.

Afifi

Taillieu

Zamorski

. Association of child abuse exposure with suicidal ideation, suicide plans, and suicide attempts in military personnel and the general population in Canada. JAMA Psychiatry. 2016;73(3):1–10.

12.

Sareen

Cox

Afifi

. Combat and peacekeeping operations in relation to prevalence of mental disorders and perceived need for mental health care. Arch Gen Psychiatry. 2007;64(7):843–852.

13.

Sareen

Belik

S-L

Afifi

. Canadian military personnel’s population attributable fractions of mental disorders and mental health service use associated with combat and peacekeeping operations. Am J Public Health. 2008;98(12):2191–2198.

14.

Kessler

Stein

Bliese

. Occupational differences in US Army suicide rates. Psychol Med. 2015;45(15):3293–3304.

15.

Ursano

Kessler

Stein

. Suicide attempts in the US army during the wars in Afghanistan and Iraq, 2004 to 2009. JAMA Psychiatry. 2015;72(9):917.

16.

Pinder

Iversen

Kapur

. Self-harm and attempted suicide among UK armed forces personnel: results of a cross-sectional survey. Int J Soc Psychiatry. 2012;58:433–439.

17.

Shen

Cunha

Williams

. Time-varying associations of suicide with deployments, mental health conditions, and stressful life events among current and former US military personnel: a retrospective multivariate analysis. Lancet Psychiatry. 2016;3(11):1039–1048.

18.

Reger

Smolenski

Skopp

. Risk of suicide among US military service members following Operation Enduring Freedom or Operation Iraqi Freedom deployment and separation from the US military. JAMA Psychiatry. 2015;72(6):561.

19.

LeardMann

Powell

Smith

. Risk factors associated with suicide in current and former US military personnel. JAMA. 2013;310(5):496.

20.

Bryan

Griffith

Pace

. Combat exposure and risk for suicidal thoughts and behaviors among military personnel and veterans: a systematic review and meta-analysis. Suicide Life Threat Behav. 2015;45(5):633–649.

21.

Bryan

Reger

LeardMann

. On deployment and military suicide risk. JAMA Psychiatry. 2015;72(9):949.

22.

Belik

Stein

Asmundson

. Relation between traumatic events and suicide attempts in Canadian military personnel. Can J Psychiatry. 2009;54(2):93–104.

23.

Sareen

Stein

Thoresen

. Is peacekeeping peaceful? A systematic review. Can J Psychiatry. 2010;55(7):464–472.

24.

Zamorski

Bennett

Boulos

. The 2013 Canadian Forces Mental Health Survey: background and methods. Can J Psychiatry. 2016;61(1 suppl):10S–25S.

25.

Statistics Canada. Canadian Forces Mental Health Survey Questionnaire. Ottawa (ON): Statistics Canada; 2013.

26.

Watkins

Sudom

Zamorski

. Association of combat experiences with post-traumatic stress disorder among Canadian military personnel deployed in support of the mission in Afghanistan. Mil Behav Heal. 2016;4(3):285–292.

27.

Sudom

Watkins

Born

. Stressors experienced during deployment among Canadian Armed Forces personnel: factor structure of two combat exposure scales. Mil Psychol. 2016;28(5):285–295.

28.

Haro

Arbabzadeh-Bouchez

Brugha

. Concordance of the Composite International Diagnostic Interview version 3.0 (CIDI 3.0) with standardized clinical assessments in the WHO world mental health surveys. Int J Methods Psychiatr Res. 2006;15(4):167–180.

29.

Walsh

MacMillan

Trocmé

. Measurement of victimization in adolescence: development and validation of the childhood experiences of violence questionnaire. Child Abuse Negl. 2008;32(11):1037–1057.

30.

Afifi

MacMillan

Boyle

. Child abuse and mental disorders in Canada. CMAJ. 2014;186(9):E324–E332.

31.

Street

Gilman

Rosellini

. Understanding the elevated suicide risk of female soldiers during deployments. Psychol Med. 2015;45(4):717–726.

32.

Mota

Medved

Wang

. Stress and mental disorders in female military personnel: comparisons between the sexes in a male dominated profession. J Psychiatr Res. 2012;46(2):159–167.

33.

Dohrenwend

Turner

Turse

. The psychological risks of Vietnam for U.S. veterans: a revisit with new data and methods. Science. 2006;313(5789):979–982.

34.

Sareen

Henriksen

Bolton

. Adverse childhood experiences in relation to mood and anxiety disorders in a population-based sample of active military personnel. Psychol Med. 2013;43(1):73–84.

35.

Van Orden

Witte

Gordon

. Suicidal desire and the capability for suicide: tests of the interpersonal-psychological theory of suicidal behavior among adults. J Consult Clin Psychol. 2008;76(1):72–83.

36.

Afifi

Enns

Cox

. Population attributable fractions of psychiatric disorders and suicide ideation and attempts associated with adverse childhood experiences. Am J Public Health. 2008;98(5):946–952.

37.

Borges

Nock

Haro Abad

. Twelve-month prevalence of and risk factors for suicide attempts in the World Health Organization world mental health surveys. J Clin Psychiatry. 2010;71(12):1617–1628.

38.

Hom

Joiner

Bernert

. Limitations of a single-item assessment of suicide attempt history: implications for standardized suicide risk assessment. Psychol Assess. 2016;28(8):1026–1030.

39.

Millner

Lee

Nock

. Single-item measurement of suicidal behaviors: validity and consequences of misclassification. PLoS One. 2015;10(10):e0141606.

40.

Holbrook

Hoyt

Stein

. Perceived threat to life predicts posttraumatic stress disorder after major trauma: risk factors and functional outcome. J Trauma. 2001;51(2):287–292.

41.

Shalev

Ankri

Gilad

. Long-term outcome of early interventions to prevent posttraumatic stress disorder. J Clin Psychiatry. 2016;77(5):e580–e587.

42.

Steenkamp

Litz

Hoge

. Psychotherapy for military-related PTSD. JAMA. 2015;314(5):489.

43.

Leppin

Bora

Tilburt

. The efficacy of resiliency training programs: a systematic review and meta-analysis of randomized trials. PLoS One. 2014;9(10):e111420.

44.

Adler

Williams

McGurk

. Resilience training with soldiers during basic combat training: randomisation by platoon. Appl Psychol Health Well Being. 2015;7(1):85–107.