The Canadian Journal of Psychiatry / La Revue Canadienne de Psychiatrie
2023, Vol. 68(9) 682‐690© The Author(s) 2022
Article reuse guidelines:
sagepub.com/journals-permissionsDOI: 10.1177/07067437221125292TheCJP.ca | LaRCP.ca
Introduction: Military sexual trauma (MST) is an ongoing problem. We used a 2002 population-based sample, followed up in 2018, to examine: (1) the prevalence of MST and non-MST in male and female currently serving members and veterans of the Canadian Armed Forces, and (2) demographic and military correlates of MST and non-MST.
Methods: Data came from the 2018 Canadian Armed Forces Members and Veterans Mental Health Follow-up Survey (n = 2,941, ages 33 years + ). Individuals endorsing sexual trauma were stratified into MST and non-MST and compared to individuals with no sexual trauma. The prevalence of lifetime MST was computed, and correlates of sexual trauma were examined using multinomial regression analyses.
Results: The overall prevalence of MST was 44.6% in females and 4.8% in males. Estimates were comparable between currently serving members and veterans. In adjusted models in both sexes, MST was more likely among younger individuals (i.e., 33–49 years), and MST and non-MST were more likely in those reporting more non-sexual traumatic events. Among females, MST and non-MST were more likely in those reporting lower household income, non-MST was less likely among Officers, and MSTwas more likely among those with a deployment history and serving in an air environment. Unwanted sexual touching by a Canadian military member or employee was the most prevalent type and context of MST.
Interpretation: A high prevalence of MST was observed in a follow-up sample of Canadian Armed Forces members and veterans. Results may inform further research as well as MST prevention efforts.
Introduction: Le traumatisme sexuel militaire (TSM) est un problème constant. Nous avons utilisé un échantillon de 2002 dans la population, suivi en 2018, pour examiner : 1) la prévalence du TSM et du non-TSM chez les hommes et les femmes en service dans les Forces armées et les anciens combattants des Forces armées canadiennes, et 2) les corrélats démographiques et militaires du TSM et du non-TSM.
Méthodes: Les données provenaient de l‘Enquête de suivi sur la santé mentale des membres et des anciens combattants des Forces armées canadiennes de 2018 (n = 2 941, âgés de 33 ans + ). Les personnes cautionnant le traumatisme sexuel ont été stratifiées en TSM et non-TSM et comparées aux personnes sans traumatisme sexuel. La prévalence de TSM de durée de vie a été calculée, et les corrélats des traumatismes sexuels ont été examinés à l‘aide des analyses de régression multinomiales.
Résultats: La prévalence globale du TSM était de 44,6% chez les femmes et de 4,8% chez les hommes. Les estimations étaient comparables entre les membres présentement en service et les anciens combattants. Dans les modèles ajustés chez les deux sexes, le TSM était plus probable chez les personnes plus jeunes (c.-à-d., les 33–49 ans), et le TSM et non-TSM étaient plus probables chez ceux déclarant plus d‘événements traumatiques non-sexuels. Chez les femmes, le TSM et le non-TSM étaient plus probables chez ceux déclarant un revenu du ménage plus faible, le non-TSM était moins probable chez les Officiers, et le TSM était plus probable chez ceux ayant des antécédents de déploiements et servant dans un environnement aérien. Un toucher sexuel non consenti par un membre ou un employé militaire canadien était le type et le contexte le plus prévalent de TSM.
Interprétation: Une prévalence élevée de TSM a été observée dans un échantillon de suivi des membres et des anciens combattants des Forces armées canadiennes. Les résultats peuvent éclairer la future recherche ainsi que les initiatives de prévention du TSM.
Keywordsmilitary sexual trauma, Canadian armed forces, veterans, epidemiology
Military sexual trauma (MST), which has varying operationalizations but unanimously includes unwanted sexual experiences during military service, is a persistent problem.1 MST has been associated with negative sequelae including mental disorders and chronic pain.2–4 With the recent dissolution of Operation HONOUR, representing efforts by the Canadian Armed Forces to more effectively respond to MST, and the military‘s more recent efforts to address this issue,5,6 it is imperative to understand the burden of MST. It is also important to identify sub-populations at greater likelihood of MST to more definitively address this problem.
Most studies examining the prevalence and correlates of MST have been conducted in the United States, and have consistently found higher estimates of MST in women and females than in males.7 A meta-analysis of 69 studies identified a lifetime prevalence of MST (including harassment and assault) of 38.4% in women and 3.9% in men among service members and veterans.8 These estimates are relatively comparable with those found in a nationally representative sample of US veterans, where the prevalence of lifetime MST was 32.4% in females and 4.8% in males,4 and a national study of 60,000 OEF/OIF era veterans that found that 41% of women and 4% of men reported its occurrence.9 Few population-based studies of MST have been conducted in Canada. However, in a 2013 cross-sectional sample of Regular Force members, a lifetime MST prevalence of 15.5% for females and 0.8% for males was observed,10 when MST was defined as unwanted sexual assault or sexual touching. This sample only allowed for an examination of correlates of MST in females, which included being single, higher rank (perhaps due to longer length of service), and more non-sexual traumatic exposures. This study did not examine MST in veterans, however. In a 2018 representative sample of current Canadian Armed Forces members, 4.4% of women and 1.1% of men Regular Force members reported past-year MST, defined as sexual assault in a military-related setting or involving military members.11 These estimates were comparable to those of a 2016 survey of Canadian Armed Forces members.12 It should be noted that some of these studies assessed gender and some assessed biological sex, which are different constructs and could impact prevalence estimates.
With regards to demographic correlates, a higher prevalence of MST has been positively associated with younger age, unemployment, being nonwhite, having enlisted, noncollege completion, and serving in the Navy, and negatively associated with being married/co-habiting.4,13,14 Few studies have examined sex- or gender-specific correlates of MST.9,15 In the Canadian study by Watkins et al.,10 demographic and military variables associated with MST included being single, residing in the Western region, higher rank, and a greater burden of non-sexual trauma exposure. However, these analyses were conducted only among females.
Limitations of previous research examining MST include studies being conducted in mostly American samples and not being able to examine sex- or gender-specific correlates. Updated studies of MST prevalence are needed in light of recent initiatives within the Canadian military (e.g., Heyder-Beattie class action settlement agreement; 16), and major sociopolitical movements (i.e., “#MeToo”) that may have impacted reporting.17 Finally, there are no estimates of MST among Canadian Armed Forces veterans, an important gap given that MST experiences may represent one factor for early military departure.18
To address knowledge gaps, we used the 2018 Canadian Armed Forces Members and Veterans Mental Health Follow-up Survey to examine: (1) the prevalence of MST among males and females, and (2) correlates of MST relative to non-military sexual trauma (NMST) and no sexual trauma. This survey provides the most contemporary lifetime estimates of MST in Canada to date. It is also the first, population-based Canadian survey to include both actively serving members and veterans in the same dataset, in order to provide MST estimates in veterans and compare between groups.
Sample. The Canadian Armed Forces Members and Veterans Mental Health Follow-up Survey is a survey of 2,941 participants (34% Regular Force members, 66% veterans) conducted in 2018. These individuals were re-interviewed following their participation in the nationally representative Canadian Community Health Survey on Mental Health and Wellbeing: Canadian Forces Supplement conducted in 2002 (response rate: 79.5%), and comprised 68% of the original sample of 5,155 actively serving Regular Force members who were eligible to participate in this survey. Further information related to participant recruitment and survey methodology is located elsewhere.19
Sociodemographic and military variables. Demographic variables were operationalized in accordance with previous research,20,21 and in order to ensure adequate cell size. Variables included: sex (male, female), age, household income, marital status, education, member status, current rank for still serving members, operational environment, lifetime deployment exposure and length of service.
Military sexual trauma and other trauma. Two items from the traumatic events module within the posttraumatic stress disorder module of the World Health Organization Composite International Interview assessed lifetime sexual trauma: “We define sexual assault as anyone forcing you or attempting to force you into any unwanted sexual activity by threatening you, holding you down, or hurting you in some way. Has this ever happened to you?” (sexual assault) and, “Has anyone ever touched you against your will in any sexual way? By this I mean unwanted touching or grabbing, to kissing or fondling” (unwanted sexual touching). An affirmative response to either question was indicative of sexual trauma. Participants who endorsed either of these items were further asked three questions that were used to determine endorsement of MST: (i) whether the events had taken place while deployed as part of a Canadian Forces operation, (ii) whether the events had occurred in the Canadian Forces workplace and (iii) whether the individuals who perpetrated the acts were a member of the Canadian Forces or a civilian employee with the Department of National Defence when the events occurred. Responses to the sexual trauma and MST-specific items were used to create a three-level variable of lifetime experiences in accordance with previous work10: MST, NMST and no sexual trauma. Individuals endorsing both MST and NMST were placed in the MST group.
Endorsement of non-sexual traumatic events was combined into a “number of non-sexual traumatic events” count variable.
STATA/MP (Version 16; 22) was used to conduct all analyses. First, cross-tabulations examined the prevalence of any MST and NMST in male and female currently serving members and veterans. Second, cross-tabulations examined the prevalence of each type and context of MST among males and females reporting any MST. Third, crosstabulations examined sex-stratified correlates of the threelevel sexual trauma variable. Fourth, associations between demographic and military characteristics and the three-level sexual trauma variable were examined using unadjusted, multinomial logistic regression analyses (available upon request). A model adjusted for those demographic correlates emerging as statistically significant in the unadjusted model for both males and females—age and income—was then computed. All analyses were stratified by sex, and sex X correlate interactions were also tested. Statistics Canada calculated sampling weights that were applied to all inferential analyses to ensure that the current sample remained representative of the original 2002 cohort. The weights accounted for attrition and out-of-scope units in 2018. To account for the complex sampling design of the data, bootstrapping was used as a standard error estimation technique. In accordance with Research Data Centre policy, most unweighted n’s (with the exception of some descriptives, rounded to base 5) and cells with n < 5 were not releasable.
The sociodemographic profile of the sample can be found in Table 1. The prevalence of MST, in comparison to NMST only and no sexual trauma, is displayed in Table 2. The prevalence of lifetime MST was 44.6% among females and 4.8% among males (9.6% in the total sample). For females, this prevalence was much higher than that of NMST (16.2%), while for males, a higher prevalence of NMST than MST was reported (8.8%). The prevalence of MST did not differ significantly by serving status (i.e., currently serving vs. veteran) across both sexes (male currently serving 5.7%; male veteran 4.3%; female currently serving 48.3%; female veteran 42.6%). On account of comparable estimates between currently serving members and veterans, we present other results stratified by sex only.
The prevalence of each type and context of MST among participants reporting any MST is shown in Table 3. A higher proportion of females than males reported experiencing all types of MST in all contexts, with the exception of unwanted sexual touching while deployed on a CAF operation, where the difference between males and females was not statistically significant (38.0% and 29.4%, respectively). Unwanted sexual touching by a military member or civilian employee of the Department of National Defence had the highest prevalence for both males (59.9%) and females (91.8%).
Table 4 presents the prevalence of sociodemographic and military correlates of MST among males and females, with the prevalence of NMST and no sexual trauma included for comparison, while Table 5 displays adjusted associations between demographic and military characteristics and the 3-level sexual trauma variable. In males and females, relative to no sexual trauma, individuals 33 to 49 years of age were more likely to report MST (relative risk ratio [RRR]s 2.06 and 1.67, respectively) relative to those 60 years and older. MST and NMST were associated with a higher number of non-sexual traumatic events in both sexes (RRR range 1.12–1.36). In females, being in the ≤$49,999 income bracket relative to ≥$50,000 was associated with a higher likelihood of NMST and MST (RRRs 2.13 and 1.51, respectively), lifetime deployment (RRR 1.48) and serving in an air environment (RRR 1.45) were associated with a higher likelihood of MST, and Officer rank (relative to Junior/Senior Non-Commissioned ranks) was associated with a lower likelihood of NMST (RRR 0.48).
Three correlate-by-sex interactions were detected: (1) Officer rank by sex on MST (RRR 0.38, 95% CI [0.16–0.94], P = 0.036); (2) number of non-sexual traumas by sex on NMST (RRR = 1.11, 95% CI [1.02–1.22], P = 0.016); and (3) number of non-sexual traumas by sex on MST (RRR = 1.12, 95% CI [1.03–1.21], P = 0.006).
This study examined the prevalence and correlates of MST in a large and contemporary (2018) sample of Regular Force Canadian Armed Forces members and veterans. The prevalence of lifetime MST was 44.6% in females and 4.8% in males (9.6% overall). Further, while the overall prevalence of NMST was comparable to that of MST (which may also include NMST), at 9.7%, females were approximately nine times more likely to experience MST than males, whereas they were twice as likely to experience NMST than males.
Understanding MST trends over time is not possible with this study in the absence of repeated cohort data, and MST was not assessed in the original 2002 survey. However, in the 2013 nationally representative sample of Regular Force members, the prevalence of MST was 15.5% in females and <1% in males, when MST was correspondingly operationalized.10 There are various possibilities for the difference in estimates between studies. First, our study was conducted using an older sample of serving members and veterans (representative of the Canadian Armed Forces in 2002), with many participants serving prior to more recent organizational efforts to prevent MST. Second, it is possible that those who re-participated in the 2018 survey were more heavily affected by MST. Third, in 2017, the world saw the rise of the #MeToo movement, and it is plausible that more individuals were empowered to disclose their experiences. As such, the observed prevalence of MST in the current study may be closer to its actual burden. It is unclear why reporting of NMST appears lower in the present study than in the 2013 study, although there have been several initiatives specific to addressing MST in Canada that may have increased reporting (e.g., Operation HONOUR; 17).
Among individuals reporting any MST, 96.7% of females and 71.5% of males reported at least one assault perpetrated by a military member or employee of the Department of National Defence, supporting the need for systemic efforts to prevent MST within the military. Further, a higher prevalence of most types and contexts of MST were reported for females than males. An exception was a comparable prevalence of unwanted sexual touching during deployment (38.0% and 29.4% for males and females, respectively). This type of MST context may be especially salient for males.
In adjusted models, correlates of MST and NMST common to both sexes included a higher number of non-sexual traumas, consistent with previous research,10 and being in the youngest age bracket for MST only. Females in the lower income category were more likely to report any sexual trauma, Officers were less likely to report NMST, and females who had deployed or served in the Air Force were more likely to report MST. We speculate that younger individuals (33–49 year olds) may be more willing to disclose their experiences due to a stronger influence of recent cultural and socio-political shifts to support MST disclosure. This age category may also be comprised of more active members who have had more recent exposure to MST. However, the current study did not include those younger than 33 years, which may have impacted MST estimates. The deployment environment may also increase MST vulnerability for females. Contrary to some prior work,10,14 rank and marital status were not associated with MST in the current study. It is possible that our dichotomous categorizations of these variables, due to restrictions of releasable cell sizes, account for this difference.
This study is the first to examine estimates of MST among Canadian Armed Forces veterans. A comparable prevalence of MST between currently serving members and veterans was identified. Females are more likely to medically release from the military,18 and future research should study reasons for early release, as MST could be a relevant contributor. This study is also among the first to examine sex-specific correlates of MST, and the current findings identify sub-groups of both males and females for whom it may be particularly important to focus efforts related to MST screening and treatment access.
The current findings should be made in the context of the following limitations. First, the 2018 survey represents a 16-year follow-up of a nationally representative survey of serving members in 2002. As such, this 2018 sample is not representative of the current composition of service members and veterans in Canada. Second, we examined sex differences in the burden of MST in the current study, however, we were unable to assess gender differences on account of the CAFVMHS assessing only biological sex and not gender identity. Future research should fill this critical knowledge gap. Third, because of the small sample size, we were also unable to examine estimates of MST within important subgroups (e.g., nuanced rank categories). Fourth, the self-report nature of variables introduces the possibility of recall bias, particularly for lifetime variables, that could lead to an underestimate or overestimate of prevalence. Fifth, MST was assessed using only two primary questions, and there may be other types of sexual trauma that were not captured by the survey items. Further, the way in which MST is assessed and operationalized can impact prevalence estimates.8 For example, the CAFVMHS began one MST item with the phrase, “We define sexual assault….” However, some MST survivors may reject the term “assault,” a phenomenon known as unacknowledged rape, and thus, answer “no” to such an item.23 Future research should continue to refine the assessment and operationalization of MST such that it is systematic across studies and measures the burden of MST in a maximally accurate way. Finally, only lifetime MST was assessed in this survey, and while it provides novel information about an older Canadian Armed Forces cohort, it is not possible to determine the recent burden of MST from these data for comparability with other estimates.11
To summarize, nearly half of female service members (44.6%) in this sample reported experiencing lifetime MST. It is also critical to emphasize that while females had a much higher prevalence of MST than males, almost 5% of males also reported MST, a number that could translate into thousands affected. There have been efforts in recent years by the Canadian Armed Forces to address military sexual mis-conduct,5 and recent recommendations to prevent MST made by the House of Commons Standing Committee on the Status of Women.24 A recent review additionally proposed recommendations for military leadership to mitigate the risk of sexual violence, including trauma-informed leadership and leadership support in identifying and managing risk factors for sexual violence, such as substance misuse.25 Finally, active efforts to reduce stigma related to MST disclosure among those who serve and universal screening of MST,26 may be key components of MST prevention and mitigation. Future studies should continue to examine MST research gaps in nationally representative samples of currently serving and veteran members. Such efforts can inform continuing efforts to eliminate the occurrence of MST and support those impacted.
The initial multistage sampling frame used for the 2002 survey included five strata delineated by regular versus reserve member status, sex, rank, military environment and region.27 The sampling frame for the 2018 CAFVMHS was the sample of regular force members who participated in the 2002 survey. When examining baseline (2002) predictors of 2018 attrition in an unweighted sample, only younger age, serving in a land occupation, and reported a history of suicide attempts were linked with overall attrition in the CAFVMHS.28 Sampling weights were developed by Statistics Canada for the 2002 CCHS-CFS sample that produced estimates representative of the target population of regular forces in 2002, wherein each individual represents several other people not in the sample. Modifications were made to these original weights to render them longitudinal, but still representative of the original target sample. These modifications included accounting for attrition and suppression of out-of-scope units.19
Drs. Mota, Bolton, Enns, El-Gabalawy, Sareen, and Afifi and Ms. Sommer conceptualized the study and methods, edited the manuscript and interpreted the results. Dr. Mota drafted the manuscript, and Drs. Mota, Bolton and Afifi guided Ms. Sommer with regards to statistical analyses. Ms. Sommer conducted the statistical analyses. Ms. MacLean and Drs Hall, Sudom, Silins and Garber contributed to the interpretation of the results and intellectual editing of multiple drafts of the manuscript.
The authors declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Canadian Institutes of Health Research Foundation grant (#333252; PI: Sareen), True Patriot Love; Canadian Institutes of Military Veterans Health Research; University of Manitoba Rady Faculty of Health Sciences Research Start-up Funds (Mota & El-Gabalawy); and a Canadian Institute for Military and Veteran Health Research Mark Zamorski Award (Sommer).
Natalie Mota https://orcid.org/0000-0003-2832-2223
Jordana L. Sommer https://orcid.org/0000-0001-8311-1854
Shay-Lee Bolton https://orcid.org/0000-0002-0750-4266
Murray W. Enns https://orcid.org/0000-0002-8375-0151
Renée El-Gabalawy https://orcid.org/0000-0002-3445-5607
Stacey Silins https://orcid.org/0000-0002-4088-9161
Tracie O. Afifi https://orcid.org/0000-0002-3745-9785
1Departments of Clinical Health Psychology and Psychiatry, University of Manitoba, Winnipeg, Canada
2Departments of Psychology and Anesthesiology, Perioperative and Pain Medicine, University of Manitoba, Winnipeg, Canada
3Department of Psychiatry, University of Manitoba, Winnipeg, Canada
4 Departments of Anesthesiology, Perioperative and Pain Medicine, and Clinical Health Psychology, University of Manitoba, Winnipeg, Canada
5 Research Directorate, Veterans Affairs Canada, Charlottetown, Prince Edward Island, Canada
6 Canadian Forces Health Services Group, Department of National Defence, Ottawa, Ontario, Canada
7 Departments of Community Health Sciences and Psychiatry, University of Manitoba, Winnipeg, Canada
Corresponding author:Natalie Mota, PhD, Department of Clinical Health Psychology, University of Manitoba, Operational Stress Injury Clinic, 2109 Portage Avenue, Winnipeg, MB, Canada, R3J 0L3.Email: Natalie.Mota@umanitoba.ca
