Abstract
Prior research among military personnel has indicated that sexual harassment, stalking, and sexual assault during military service are related to negative health sequelae. However, research specific to LGBT U.S. service members is limited. The current study aimed to explore the health, service utilization, and service-related impact of stalking and sexual victimization experiences in a sample of active-duty LGBT U.S. service members (N = 248). Respondent-driven sampling was used to recruit study participants. U.S. service members were eligible to participate if they were 18 years or older and active-duty members of the U.S. Army, U.S. Navy, U.S. Marine Corps, or U.S. Air Force. This study included a sizeable portion of transgender service members (N = 58, 23.4%). Sociodemographic characteristics, characteristics of military service, health, and sexual and stalking victimization in the military were assessed. Regression was used to examine relationships between health and service outcomes and sexual and stalking victimization during military service. Final adjusted models showed that experiencing multiple forms of victimization in the military increased the odds of visiting a mental health clinician and having elevated somatic symptoms, posttraumatic stress disorder symptomatology, anxiety, and suicidality. Sexual and stalking victimization during U.S. military service was statistically significantly related to the mental and physical health of LGBT U.S. service members. Interventions to reduce victimization experiences and support LGBT U.S. service members who experience these types of violence are indicated. Research that examines the role of LGBT individuals’ experiences and organizational and peer factors, including social support, leadership characteristics, and institutional policies in the United States military is needed.
Introduction
Sexual harassment, stalking, and sexual assault during military service, and the related impact on U.S. service members who identify as lesbian, gay, bisexual, or transgender (LGBT) have been understudied in the years following the repeal of Don’t Ask, Don’t Tell (DADT; Castro & Goldbach, 2018). DADT was a policy in place between 1994 and 2011 that prohibited LGBT U.S. service members from serving openly. Prior research with active-duty samples suggested that sexual trauma experienced during military service is related to negative health sequelae among U.S. service members (Morgan et al., 2017; Rosellini et al., 2017; Shipherd et al., 2009); however, research specific to LGBT U.S. service members was heavily constrained during DADT, given that LGBT people could be removed from U.S military service if their identities were uncovered (Burks, 2011). Since the repeal, evidence has begun to emerge suggesting LGBT U.S. service members face an enhanced risk of sexual harassment, stalking, and sexual assault during military service compared to their non-LGBT peers (Schuyler et al., 2020). As the U.S. military works to better understand the well-being of LGBT service members, insight into how these victimization experiences affect individuals’ health and readiness to serve can inform military leaders and health care providers and help direct tailored intervention strategies. The current study explored the health and service-related impact of sexual harassment, stalking, and sexual assault during military service in a first-of-its-kind sample of active-duty LGBT U.S. service members.
Health Impact of Sexual and Stalking Victimization During Military Service
Scant research has examined links between sexual and stalking victimization during military service and health outcomes among samples of active-duty U.S. service members. Extant literature has suggested an increased likelihood of posttraumatic stress disorder (PTSD), depression, anxiety, substance use, somatic symptoms, and health care utilization among those who experienced sexual victimization during service, including sexual harassment and sexual assault (Magley et al., 1999; Maguen et al., 2012; Morgan et al., 2017; Parnell et al., 2018; Seelig et al., 2017; Shipherd et al., 2009; Wolfe et al., 1998). Among female U.S. service members, sexual victimization has been found to more strongly predict PTSD than combat exposure or other types of traumatic violence (Kearns et al., 2016; Parnell et al., 2018; Wolfe et al., 1998). Among male military personnel (mixed service members and veterans), sexual victimization during military service has been found to predict PTSD when adjusting for combat exposure, though more research is needed on the relationship between sexual victimization in the military and PTSD among male military personnel (Wolfe-Clark et al., 2017).
It is unclear whether findings with heterosexual and cisgender U.S. service members might reflect the experiences of LGBT U.S. service members, given that sexual identity (e.g., lesbian, gay, or bisexual identity) and gender identity (i.e., internal sense of gender; e.g., man, woman, transgender man, genderqueer) have seldom been assessed or reported in active-duty samples. Research among LGBT veterans, however, has explored health outcomes associated with experiences of sexual victimization during military service (Beckman et al., 2018; Lucas et al., 2018). Several studies have noted an increased risk of PTSD, depression, substance use, psychiatric disorders (e.g., bipolar disorder, personality disorder), and suicidality among LGBT veterans who experienced sexual victimization (Beckman et al., 2018; Lehavot & Simpson, 2014; Lindsay et al., 2016; Sexton et al., 2018). In some cases, relationships between sexual victimization and mental health outcomes were more pronounced in LGBT veterans than non-LGBT veterans (Lehavot & Simpson, 2014; Sexton et al., 2018). However, without additional information on symptom onset or time since victimization, it is difficult to discern whether health problems associated with these victimization experiences among LGBT veterans also occurred during active military service.
In addition, LGBT U.S. service members may face unique challenges associated with disclosing sexual and stalking victimization experiences and seeking care for victimization-related health issues. Among civilian women, those identifying as bisexual have reported a greater impact of social reactions to disclosure of sexual trauma and greater difficulties with recovery compared to heterosexual women (Sigurvinsdottir & Ullman, 2015). In the wake of DADT, environments of sexual stigma and resistance to institutional change may persist; this could not only enhance the negative impact of sexual and stalking victimization during military service for LGBT U.S. service members, but it may also discourage reporting and help seeking among those who do not wish to disclose their sexual orientation or gender identity (Johnson et al., 2015). Concurrently, military health care providers may be unprepared to care for LGBT U.S. service members or be reluctant to discuss nonheterosexual sexual behaviors (Johnson et al., 2015; Rerucha et al., 2018), further hindering provision of care for those who have experienced or are experiencing sexual harassment, stalking or sexual assault during their military service. Thus, LGBT U.S. service members may face unique barriers that prevent linkages to care and perpetuate the negative impacts of sexual and stalking victimization during military service.
Minority Stress: The Perpetrator Hypothesis
Research on LGBT populations’ behavioral health experiences has largely relied on a minority stress perspective. This perspective was initially applied to sexual minorities (i.e., LGB; Meyer, 2003), and later to gender minorities (i.e., transgender; Hendricks & Testa, 2012). It suggests that the high burden of negative behavioral health experiences observed among LGBT people is likely the result of structural stigma and targeted violence based on individuals’ minority identities. More specifically, the minority stress model posits that due to structural stigma LGBT people encounter in their social environment (e.g., discriminatory policies or institutional practices, overt violence), they often internalize negative expectations about themselves (e.g., they will encounter rejection or violence due to their LGBT identity). The chronic burden of these negative expectations tax coping and resilience and are thought to result in many negative behavioral health experiences, including depression and anxiety. Several existing studies have added empirical support for this mechanistic pathway from discrimination to negative behavioral health outcomes (Chodzen et al., 2019; Klemmer et al., 2018; Lehavot & Simpson, 2014).
Recently, a sexual and gender minority stress perspective was adapted for use in the military environment through a “perpetrator hypothesis” (Castro & Goldbach, 2018), with implications to guide future intervention and program planning efforts. The foundation of the perpetrator hypothesis is that the structural environment of the military may create a context of elevated discrimination and violence perpetration against LGBT U.S. service members due to, for example, organizational policies and programs that overtly discriminate based on LGBT identity (e.g., DADT and the military’s current policy barring transgender individuals who wish to serve in alignment with their gender identity). Targeted discrimination and sexual victimization are hypothesized to have a direct negative impact on the functional domains of LGBT U.S. service members (e.g., health, help seeking, and military performance and readiness).
Presently, due to a lack of research on LGBT active-duty U.S. service members’ experiences of sexual harassment, stalking, and sexual assault, the impact of these victimization experiences on functional domains remains unknown (Burrelli, 2010). The present study addressed this gap in understanding by examining the relationship between sexual harassment, stalking, and sexual assault during military service (i.e., sexual and stalking victimization in the military) and health (i.e., anxiety, depression, PTSD, and somatic symptomatology), help-seeking behavior (i.e., last-year physician and last-year mental health provider visits), and lost duty days due to illness among LGBT active-duty U.S. military service members.
Methods
Procedures
Recruitment.
Our study utilized a respondent-driven sampling (RDS) methodology. Complete details of our RDS recruitment and enrollment procedures are reported elsewhere (Schuyler et al., 2020). We began with recruitment of seeds using referrals from an expert advisory panel. Primary investigators and study staff members also provided seed recruits through their military network contacts. When seed recruitment through panel and study staff referrals slowed, we expanded recruitment by promoting the study through popular military-related social media and on college campuses. Each strategy was accompanied by a unique referral code. Referral codes were used to monitor and track referral effectiveness and to differentiate study respondents who were enrolled through advertisements on social media and on college campuses, and those who were referred by a study participant. Survey respondents received unique referral codes following study participation to refer others to the study. Survey participants received a $25 electronic gift card (if they completed the online survey off duty) and a $10 gift card for each referral who completed the survey. Checks for fraud were conducted both throughout study recruitment and during analysis. Examples of fraudulent survey responses included those with IP addresses outside the United States or in areas where no military base existed; duplicate IP addresses with similar response patterns; two or more incorrect attention-control measures; and short survey duration. Four-hundred forty-eight persons accessed our survey website, of which 73.2% were eligible. Of those, 327 (99.6%) agreed to participate in the survey. Fraud checks eliminated 79 (24.0%) surveys. The resulting sample consisted of 248 LGBT active-duty U.S. service members. The institutional review boards at the University of Southern California and the University of California Los Angeles approved all study data collection methods and procedures.
Measures
Sociodemographic information.
Sexual identity was measured using one item, “What is your sexual identity?” Response options were heterosexual or straight, gay or lesbian, bisexual, and sexual orientation not listed here. For analysis, this item was coded as binary for LGBT, with heterosexual or straight set as the reference group. Sex assigned at birth was reported via one item asking, “What sex were you assigned at birth (i.e., what sex is on your birth certificate)?” Gender identity was assessed by an item with six response options: male, female, transgender male or trans man, transgender female or trans woman, genderqueer or gender nonconforming, and gender identity not listed. For analysis, gender identity was coded as a binary variable for transgender, with the reference group set as cisgender (assigned sex matched their gender identity). Respondents also reported their age and racial and ethnic identity. Racial and ethnic identity response options were Black or African American, Latino or Hispanic, White or Caucasian, Native American or Alaska Native, Asian or Pacific Islander, multiracial, and other. Due to small sample sizes of all categories but White, Latino or Hispanic, and Black or African American, all other racial and ethnic categories were condensed into one category of other racial and ethnic identities for analysis (reference group: White).
Characteristics of military service.
Respondents reported the number of years they had been serving in the military and in which service branch. Military service branch was reported using the following options: U.S. Air Force, U.S. Army, U.S. Marine Corps, and U.S. Navy. For analysis, we chose the U.S. Air Force as the reference group, given its reported rates of sexual assault are consistently lower relative to other branches (Department of Defense, 2017, 2019). We used one item assessing current pay grade (responses from E-1 to O-6) to determine officer rank in a binary variable for analysis (reference: enlisted).
Health experiences.
Respondents answered two binary response items assessing if they had seen a medical provider and a mental health provider in the past year (i.e., “During the past year, did you receive any care for a physical injury or ailment?” and “During the past year did you receive counseling/mental health services?”). Survey takers responded to a measure assessing somatic symptom severity using the 15-item patient health questionnaire (PHQ-15; Cronbach’s alpha = .80; Kroenke et al., 2002). Respondents were first presented with the following question, “During the past 7 days, how much have you been bothered by any of the following problems?” and then a list of 15 somatic symptoms including, for example, stomach pain, back pain, and headaches. Response options for each were “Not bothered at all (0),” “Bothered a little (1)” or “Bothered a lot (2).” Responses to each somatic symptom were summed to obtain the scale score (range: 0–30), with symptom severity being minimal (0–4), low (5–9), medium (10–15), or high (15–30). For regression analysis, the somatic symptom severity scale was binary coded for high somatic symptom severity following validated scoring procedures (reference group: <15). Respondents answered questions on their anxiety severity using the 7-item Generalized Anxiety Disorder scale (GAD-7; Cronbach’s alpha = .99; Spitzer et al., 2006). Respondents were first presented with the question, “Over the last two weeks, how often have you been presented with the following problems?” Then, using a Likert scale with response options “Not at all (0),” “Several days (1),” “More than half the days (2),” and “Nearly every day (3),” survey takers responded to seven problems (e.g., “feeling nervous anxious or on edge,” “trouble relaxing,” and “becoming easily annoyed or irritable”). Responses were summed (range: 0–21) and represented minimal (0–4), mild (5–9), moderate (10–14), or high (15–21) anxiety. For regression analysis, the scale score was made binary for elevated anxiety (reference group: <5). Cutoff scores were based on DSM-5 criteria.
Respondents also answered questions on their PTSD symptom severity using the PTSD Checklist for DSM-5 (PCL-5; Cronbach’s alpha = .92; Blevins et al., 2015). Respondents were asked to rate how bothered they had been by each of 20 items in the past month due to a very stressful experience (e.g., Item example: “Repeated, disturbing and unwanted memories of the stressful experience”). Responses were recoded using a 5-point Likert scale with response options, “Not at all (0),” “A little bit (1),” “Moderately (2),” “Quite a bit (3),” and “Extremely (4).” Items were summed to provide a total severity score (range = 0–80). Following current psychometric work and DSM-5 criteria, the scale was made binary for regression analysis, with scores of 33 or greater representing clinically signifying PTSD symptom severity (reference group: <33). Additionally, respondents answered questions regarding their suicidality risk using the Suicide Behaviors Questionnaire–Revised (SBQ-R; Cronbach’s alpha = .65; Osman et al., 2001). Respondents answered four distinct questions about their suicidal thoughts and behaviors (e.g., “Have you ever thought about or attempted to kill yourself?”). Responses for each question were scored using a numerical scale that varied by question and ranged from 1 to 4, 1 to 5, 1 to 3, and 0 to 6. Responses to each item were summed for the total scale score (range: 3–18). Following existing psychometric work, scores of 7 or greater represented high suicidality risk (reference < 7).
Lost duty days.
Respondents reported the number of days that they missed duty via an item asking, “In the last year, approximately how many days did you not report for duty because you were sick, injured, or on limited duty?” For analysis, this item was dichotomized using a cutoff of 5 or more lost duty days based on the mean number of lost duty days (reference group: <5 lost duty days).
Sexual harassment, stalking, and sexual assault during military service.
Details on all sexual and stalking victimization items, and descriptive data for each item in the current sample, have been reported previously (Schuyler et al., 2020). A composite variable was created representing the number of types of sexual and stalking victimization experienced by study respondents (ranging from 1 to 3 types). First, sexual harassment was assessed with six questions adapted from the definition used by the Military Equal Opportunity Program (Department of Defense, 2015) and a RAND Corporation military workplace survey (Morral et al., 2015). Questions referenced repeated experiences encountered during military service, both on- or off-duty and on- or off-base, such as “someone repeatedly telling jokes of a sexual nature” or “someone repeatedly making sexual comments, gestures, or body movements.” All six items were combined into one variable and dichotomized for analysis, wherein a single endorsement of any item constituted an experience of sexual harassment (reference group: no reported sexual harassment). Stalking was assessed with four questions based on a review of the Uniform Code of Military Justice (UCMJ, 2006), and Love is Respect, a dating violence intervention (Love is Respect, n.d.). Questions referenced experiences encountered during military service, both on- or off-duty and on- or off-base (e.g., “Someone showing up at your home or workplace unannounced or uninvited;” “Someone sending you unwanted messages, emails, or phone calls”). Participants were asked how often each type of stalking occurred (never, 1 time, 2–4 times, or 5+ times). Two or more experiences of any stalking experience constituted stalking. Sexual assault was assessed with six questions adapted from a U.S. Department of Justice special report on rape and sexual assault victimization among college women and the UCMJ (Sinozich & Langton, 2014; UCMJ, 2006). The questions asked about nonconsensual or unwanted sexual contact experienced during military service involving a military member or civilian, someone known to the participant, or a stranger (questions are detailed in total in a companionate analysis; author reference). Examples of dichotomous items included are “oral sex,” “anal intercourse,” and “vaginal intercourse” (with respective definitions provided for sexual behaviors; e.g., “Forced touching of a sexual nature [i.e., forced kissing, touching of private parts, groping, fondling]”). Finally, sexual harassment, stalking, and sexual assault responses were coded into binary items wherein the presence of any victimization was coded as 1 (reference group: no victimization). The three items were then summed into a single count variable demonstrating the total number of types of sexual and stalking victimization experienced during military service reported by study respondents, ranging from no types to all three types.
Analysis
Descriptive statistics were obtained for sociodemographic and military service characteristics; reports of sexual harassment, stalking, and sexual assault during military service; and health- and service-related outcome variables. Logistic regression was used to examine bivariate associations between sexual and stalking victimization experiences during military service, health, and lost service days. Variables found to be significant at the bivariate level were then tested together in adjusted multivariable mixed-effects logistic regression models to include terms adjusting for the potential effect of clustering in RDS recruitment chains. Variables that became nonsignificant were removed to obtain the most parsimonious final models. All analyses were conducted using Stata version 14 (StataCorp, 2015). Additionally, we assessed the amount and interference of missing data. Because relatively little data were missing overall, we used listwise deletion to include all available data across analyses. We also assessed for multicollinearity; diagnostic tests revealed acceptable tolerance values for all independent variables (mean VIF = 1.21).
Results
Participants
Table 1 summarizes the demographic characteristics of the sample. Two-fifths of the active-duty LGBT sample were officers (n = 100, 40.3%), and the service branch with the most enrolled respondents was the U.S. Army (n = 105, 42.3%). More than half of the sample was White (n = 164, 66.2%), with Latinos (n = 33, 13.3%) and Blacks or African Americans (n = 20, 8.1%) as the second and third largest groups, respectively. Transgender U.S. service members (n = 58, 23.4%) were well represented in the sample, and many participants were assigned male sex at birth (n = 142, 57.3%).
Sociodemographic Characteristics Among Active Duty LGBT Military Service Members Recruited Using Respondent-driven Sampling (n = 248).
Table 2 summarizes participants’ responses to sexual harassment, stalking, and sexual assault during military service; health and service utilization questionnaires; and lost duty days. More than one-third of the sample reported having experienced stalking (n = 84; 36.7%), and more than three quarters reported having been sexually harassed (n = 186; 81.6%). Approximately one quarter reported having been sexually assaulted (n = 64; 25.8%). The mean number of types of sexual and stalking victimization experienced by the sample overall was 1.4 (SD = 0.9). More than half of respondents saw a medical provider in the last year (n = 138; 56.1%), whereas slightly fewer saw a mental health clinician in the last year (n = 103; 41.9%). One quarter (n = 61) of the sample had somatic symptom severity that was either medium or high, whereas 33.5% (n = 76) of the sample experienced more than minimal anxiety. The majority of the sample reported low PTSD symptomatology (84.8%, n = 207), and about one third had high levels of suicidality (33.2%, n = 82). The mean number of lost service days due to illness, injury, or limited-duty status was 3.8 days (SD = 11.4).
Sexual Harassment, Stalking, and Sexual Assault During Military Service, Health, Service Utilization, and Lost Service Days Among LGBT Service Members Recruited Using Respondent-driven Sampling.
Regression Analyses
Bivariate analyses revealed significant correlates of health, service utilization, and lost duty days. Sexual and stalking victimization experiences were positively associated with seeing a mental health clinician in the last year, high somatic symptom severity, high anxiety, PTSD symptomatology, and suicidality in our sample. There was no relationship between sexual and stalking victimization experiences in the military and having seen a physician in the last year, nor with lost duty days due to illness. Table 3 summarizes all findings in adjusted mixed-effects logistic regression models. In our final model for having visited a mental health clinician in the last year (χ2[3] = 20.16, p < .001), we found that each additional sexual and stalking victimization experience reported was correlated with a 1.9 increase in odds of having seen a mental health clinician (p = .017). In the final model for somatic symptom severity (χ2[1] = 6.67, p = .009), there was a 2.7 increase in odds of symptomatology for each additional report of sexual or stalking victimization (p = .010). As for elevated anxiety, our final model (χ2[1] = 11.32, p < .001) showed that each additional report of sexual or stalking victimization was related to a 76% increase in odds of having elevated anxiety (OR = 1.7, p = .001). In the final model for PTSD symptomatology (χ2[2] = 16.81, p < .001), there was two times greater odds of having elevated PTSD symptoms for each additional type of sexual or stalking victimization encountered during military service (OR = 2.0, p = .001). Finally, the adjusted model for suicidality (χ2[2] = 23.43, p < .001) showed that participants had a 1.8 increase in odds of exhibiting suicidality for each additional type of sexual or stalking victimization encountered during their military service (OR = 1.8, p < .001).
Adjusted regression results among LGBT service members recruited using respondent-driven sampling.†
Note. Models included only variables significant at the bivariate level. Sexual orientation, racial and ethnic identity, officer rank, and branch were significant at the bivariate level and included in adjusted models but fell from significance and were removed from each final model.
(–) denotes predictor that fell from significance in multivariate models.
*p ≤ .05. **p ≤ .01. ***p ≤ .001.
AFAB = Assigned Female Sex at Birth, Reference = Assigned Male at Birth.
†Sample recruited using respondent-driven sampling. Multivariate models included a term to adjust for cluster effects of RDS recruitment chains.
Discussion
We sought to understand the relationship between sexual and stalking victimization experienced during military service and help seeking, health, and lost service days in a sample of active-duty LGBT military U.S. service members. The findings of the present study show that sexual and stalking victimization was associated with a greater likelihood of seeing a mental health clinician in the last year and increased odds of reporting clinical levels of negative mental health sequelae (i.e., anxiety, PTSD symptomatology, and suicidality) and a high degree of somatic symptoms. We did not find a relationship in our data between sexual and stalking victimization experiences and having seen a physical health clinician. Further, sexual and stalking victimization was not related to having missed any service days due to reported illness.
Our findings largely align with civilian literature highlighting the unique nature of sexual victimization and related negative health outcomes among LGBT populations. For one, experiences of trauma or discrimination related to sexual or gender identity may accumulate over time, potentially contributing to an enhanced risk of psychological health issues and a lower likelihood of seeking care (Ard & Makadon, 2011; Mustanski et al., 2016). Among LGBT U.S. service members, cumulative experiences of sexuality or gender identity-related discrimination, stigma, harassment, and violence may hinder health, work, and social functioning (Murdoch et al., 2007), and this study provides insight into the impact of sexual victimization experiences in the military context. The findings of the present study highlight sexual and stalking victimization among LGBT U.S. service members as an important concern with implications on health and for military readiness, which warrants further research.
Additional correlates of mental health outcomes in the current sample are important to note. Controlling for experiences of sexual and stalking victimization, transgender identity remained a significant predictor of suicidality, and female sex assigned at birth significantly predicted PTSD symptomatology. Although limited by the small number of transgender participants, our findings align with previous work suggesting an elevated risk of suicidality among transgender veterans, compared to the larger veteran population (Blosnich et al., 2013). One study identified factors related to suicidality among transgender veterans, which included stigmatization during military service (Lehavot & Simpson, 2014). Additional research is needed to further examine disparities in suicidality among transgender U.S. service members and identify strategies for mitigating suicide risk, including those focused on military and unit climate factors.
Research examining differences in PTSD symptoms among active-duty men and women has produced mixed results (Hourani et al., 2014; Jacobson et al., 2015; Maguen et al., 2012), similar to findings among veteran samples (e.g., Crum-Cianflone & Jacobson, 2014; Kang et al., 2005). Among active-duty samples, men and women report PTSD related to military trauma at similar rates (Jacobson et al., 2015; Maguen et al., 2012). However, there may be variations in the degree of PTSD symptoms related to specific types of trauma; in particular, female U.S. service members have reported higher PTSD symptoms related to combat-related trauma than male U.S. service members (Hourani et al., 2014). Our findings extend this literature by identifying disparities in PTSD symptoms among LGBT U.S. service members; controlling for sexual and stalking victimization during service suggests that other military traumas (e.g., combat exposure) may play an important role in the development of PTSD for LGBT U.S. service members assigned female sex at birth. Additional research among LGBT U.S. service members is warranted.
Sexual and Stalking Victimization and LGBT U.S. Service Members' Health
In addressing the mental and physical health of LGBT U.S. service members, it is important to assess for sexual and stalking victimization and quickly intervene to prevent the progression of symptoms (e.g., anxiety) into increasingly adverse and severe outcomes (e.g., PTSD, suicidality). The identification of sexual and stalking victimization and the treatment of associated negative health sequelae can also prevent potential future revictimization. Our findings underscore a need for competent military medical and mental health providers who can assess and discuss minority sexual orientation, gender identity, and sexual and stalking victimization with U.S. service members. Training to improve the ability of military health care providers to appropriately serve LGBT U.S. service members is indicated. Military mental and physical health providers should be prepared to treat LGBT U.S. service members by providing care that is empathic, is nonbiased, and deliberately affirms their minority identities and status as competent and important members of the military (Johnson et al., 2015). Recent research has shown that many military health care providers may be unprepared to care for LGBT U.S. service members in this way, given reluctance to ask about sexual behaviors in this group (Johnson et al., 2015; Rerucha et al., 2018). Similar to efforts being implemented in the veteran health care system (e.g., Kauth & Shipherd, 2016), programs to train military health providers on the competent and appropriate care of LGBT U.S. service members should be pursued (Johnson et al., 2015; Shrader et al., 2017). As our data showed, experiences of sexual and stalking victimization were related to increased odds of having seen a mental health provider. Military health care providers should therefore be well prepared to address sexual and stalking victimization with LGBT U.S. service members. This, in part, includes having a nuanced understanding of the needs of this minority group.
Structural Implications
Examination of health disparities with military and veteran communities has been especially limited among LGBT populations and therefore, research with this group has been recommended as an essential next step to improve military health systems overall (Kondo et al., 2017). Our study begins to answer this call for knowledge development among minority groups in military service units. The results of a recent companion analysis (Schuyler et al., 2020) and the present study suggest that sexual and stalking victimization is elevated among LGBT U.S. service members and negatively affects their well-being and service readiness and by extension, that of the U.S. military overall. For example, sexual and stalking victimization has been shown to contribute to lower work satisfaction, military commitment, and work productivity among U.S. service members (Magley et al., 1999) and higher rates of demotion and greater attrition from the military relative to those who have not experienced sexual and stalking victimization (Rosellini et al., 2017). Therefore, military leaders and decision-makers must intervene to improve the structural environment and climate of the military to mitigate environmental factors that contribute to sexual and stalking victimization risk, so as to promote the well-being of both LGBT U.S. service members and the U.S. military as a whole.
Studies have reviewed likely etiological factors of sexual and stalking victimization and its associated health sequalae (Campbell et al., 2009; Castro & Goldbach, 2018), and these studies provide clear paths toward the creation of an affirming military climate that is likely to generate reduced victimization and improved readiness. These studies lay out future research directions toward evidence-based policy interventions to improve military climate for all U.S. service members, including LGBT individuals. Campbell et al. (2009) highlighted the pervasiveness of self-blame and shame in the etiology and perpetuation of elevated sexual and stalking victimization. They argued that messages promoting self-blame and concealment of sexual and stalking victimization are pervasive across all levels of the ecological model in and outside of the military context (e.g., shaming messages encountered by individuals in their social networks, in their interactions with legal and medical systems, and from the overall culture and climate of the military). To address the negative impact of victimization experiences on U.S. service members’ well-being and military readiness, shaming messages at each level should be replaced with positive messages that convey the harm caused by and intolerance of sexual and stalking perpetration toward all U.S. service members. Creating environments that mitigate risk of sexual and stalking victimization by reducing organizational sexual stigma and sexual prejudice and that provide care to address potential health and career-related effects of victimization are essential (Burks, 2011; Johnson et al., 2015).
Future Research Directions
The role of military leaders, supportive peers, and organizational responses and support (i.e., peer and organizational factors), as well as LGBT U.S. service members’ individual characteristics (i.e., concealment of minority identity, internalized homophobia, and fear of rejection), have been theorized as likely important moderators of the negative impact of sexual and stalking victimization on the functional domains of LGBT U.S. service members (Castro & Goldbach, 2018; Daniel et al., 2019; Martin et al., 2000). Future research should aim to understand how individual characteristics and peer and organizational factors affect the relationship between sexual and stalking victimization experiences and U.S. service members’ functional domains. Such research can highlight factors that are amenable targets for interventions to reduce sexual and stalking victimization and its impact on LGBT U.S. service members’ well-being.
Furthermore, research should seek to better understand contextual factors (e.g., number and characteristics of perpetrators, location of sexual and stalking victimization experiences) that may contribute to the severity and nature of victimization-associated health outcomes (Bennett et al., 2019), especially among LGBT U.S. service members. For example, some literature has suggested that perceived social support and reactions to sexual and stalking victimization disclosure may be more negative among LGBT U.S. service members than among their cisgender heterosexual peers who have experienced such victimization (Sigurvinsdottir & Ullman, 2015). Additionally, the nature of leadership and unit climate and environment regarding sexual violence, including responses to reporting of sexual and stalking victimization, may affect service member health and well-being.
Limitations
The present study has limitations. The impact of sexual and stalking victimization on functional domains may change over time after the first victimization experience. The present study was cross-sectional, and the temporality of relationships cannot be inferred. Longitudinal studies that attempt to understand time until onset and severity of symptoms following sexual and stalking victimization are needed to establish causality between victimization and behavioral health outcomes. Furthermore, given the relatively small sample of individuals reporting each form of sexual and stalking victimization, we could not examine the health- and service-related impact of each individual form of sexual and stalking victimization on our health outcomes of due to limited statistical power. It will be necessary to enroll larger samples of LGBT U.S. service members in future research to more fully explore sexual and stalking victimization in this minority group. A continuous relationship might exist between variables (e.g., higher rates of victimization are associated with worse health outcomes), but again, due to limited statistical power, we could not analyze the relationship of each type of sexual victimization with each outcome.
Additionally, our study has a potential self-report bias, given the sensitive nature of questions regarding sexuality and victimization. Given potential repercussions of reporting sexual and stalking victimization during military service, members may have underreported their experiences of victimization. Limitations in our measures are also notable. Items on sexual and stalking victimization, although informed by military language and policy, have not been utilized as frequently as other measures, such as the military sexual trauma screener used in the Veterans Health Administration (Mengeling et al., 2019). At once, sexual assault researchers have recommended the use of multiple behaviorally specific items to assess unique victimization experiences (e.g., Ullman & Brecklin, 2002) and thus, our measures may have captured a wider breadth of victimization experiences than a brief clinical screening measure. Furthermore, our measure assessing for stalking victimization may be overly inclusive, given that due to time constraints, we could not assess characteristics of stalking experiences (e.g., identity of perpetrator, degree of persistence or threat), which would have allowed us to differentiate more severe stalking experiences. We also recognize that our suicidality measure may be limited, as reflected by a low Cronbach’s alpha statistic. This may be related to the low prevalence of suicidality in the sample, the small number of items in the scale, or the aggregation of suicidal ideation and attempt in a single measure. Future work should investigate more sophisticated suicidality measures that examine ideation and behavior as separate constructs.
In addition, the present study did not include a comparison group of non-LGBT service members. Therefore, the prevalence and magnitude of the effects of sexual and stalking victimization on the health of LGBT U.S. service members in comparison to the general population of U.S service members cannot be established. Research that enrolls sufficient samples of LGBT and non-LGBT U.S. service members to examine their differential health experiences and the relationship of structural factors (e.g., organizational culture and affirming or discriminatory LGBT policies) on health are needed.
There were also some limitations related to our data-handling procedures. Despite fraud-prevention measures, survey responses might have included invalid entries (e.g., some individuals may not have been active-duty U.S. service members). Additionally, though RDS has been strongly indicated for use with LGBT populations that are hard to reach for research, it may not provide a sample as representative as one using probability-based methods. Finally, the majority of our sample was serving in the U.S. Army and U.S. Air Force, and although this is relatively characteristic of the military’s distribution of U.S. service members across branches (Department of Defense, 2017), findings may not generalize to U.S. military service members as a whole.
Conclusion
A high level of military performance and readiness relies on U.S. service members with strong physical and mental health and positive evaluations of the military workplace and their role in the military. Sexual and stalking victimization during military service is a serious barrier to military readiness that has negative impacts on the health and satisfaction of U.S. service members. Reports of sexual and stalking victimization are prevalent among LGBT U.S. service members, and this must be addressed by military leaders and those charged with the health care of these minority U.S. service members. Addressing sexual and stalking victimization will require interventions that reduce individuals’ shame regarding these victimization experiences and promote and affirm LGBT U.S. service members as essential at every level of the military. This is underscored given recent policy directives that prohibit transgender U.S. service members (Gaouette, 2019), signaling to LGBT U.S. service members that they may not be worthy of U.S. military service. It should be noted that given this policy context, it remains extremely difficult to conduct research with LGBT U.S. military service members, even though they are allowed to serve openly. This is a major issue prohibiting health and well-being in this group of U.S. service members.
Footnotes
Authors’ Note
Cary Leonard Klemmer is also affiliated with DePaul Family and Community Services, DePaul University, Chicago, IL, USA.
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 work was supported by the Office of the Assistant Secretary of Defense for Health Affairs through the Psychological Health/Traumatic Brain Injury Research Program (W81XWH-15-1-0699, W81XWH-15-1-0700, and W81XWH-15-1-0701). The U.S. Army Medical Research Acquisition Activity (Fort Detrick, MD) was the awarding and administering acquisition office. Opinions, interpretations, conclusions, and recommendations are those of the author and are not necessarily endorsed by the U.S. Department of Defense.
